--- /dev/null
+mtest
+mtest[23]
+testdb
+mdb_stat
+*.[ao]
+*.so
+*[~#]
+*.bak
+*.orig
+*.rej
+core
+core.*
+valgrind.*
--- /dev/null
+Copyright 2011 Howard Chu, Symas Corp.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted only as authorized by the OpenLDAP
+Public License.
+
+A copy of this license is available in the file LICENSE in the
+top-level directory of the distribution or, alternatively, at
+<http://www.OpenLDAP.org/license.html>.
+
+OpenLDAP is a registered trademark of the OpenLDAP Foundation.
+
+Individual files and/or contributed packages may be copyright by
+other parties and/or subject to additional restrictions.
+
+This work also contains materials derived from public sources.
+
+Additional information about OpenLDAP can be obtained at
+<http://www.openldap.org/>.
--- /dev/null
+# Doxyfile 1.7.1
+
+# This file describes the settings to be used by the documentation system
+# doxygen (www.doxygen.org) for a project
+#
+# All text after a hash (#) is considered a comment and will be ignored
+# The format is:
+# TAG = value [value, ...]
+# For lists items can also be appended using:
+# TAG += value [value, ...]
+# Values that contain spaces should be placed between quotes (" ")
+
+#---------------------------------------------------------------------------
+# Project related configuration options
+#---------------------------------------------------------------------------
+
+# This tag specifies the encoding used for all characters in the config file
+# that follow. The default is UTF-8 which is also the encoding used for all
+# text before the first occurrence of this tag. Doxygen uses libiconv (or the
+# iconv built into libc) for the transcoding. See
+# http://www.gnu.org/software/libiconv for the list of possible encodings.
+
+DOXYFILE_ENCODING = UTF-8
+
+# The PROJECT_NAME tag is a single word (or a sequence of words surrounded
+# by quotes) that should identify the project.
+
+PROJECT_NAME = MDB
+
+# The PROJECT_NUMBER tag can be used to enter a project or revision number.
+# This could be handy for archiving the generated documentation or
+# if some version control system is used.
+
+PROJECT_NUMBER =
+
+# The OUTPUT_DIRECTORY tag is used to specify the (relative or absolute)
+# base path where the generated documentation will be put.
+# If a relative path is entered, it will be relative to the location
+# where doxygen was started. If left blank the current directory will be used.
+
+OUTPUT_DIRECTORY =
+
+# If the CREATE_SUBDIRS tag is set to YES, then doxygen will create
+# 4096 sub-directories (in 2 levels) under the output directory of each output
+# format and will distribute the generated files over these directories.
+# Enabling this option can be useful when feeding doxygen a huge amount of
+# source files, where putting all generated files in the same directory would
+# otherwise cause performance problems for the file system.
+
+CREATE_SUBDIRS = NO
+
+# The OUTPUT_LANGUAGE tag is used to specify the language in which all
+# documentation generated by doxygen is written. Doxygen will use this
+# information to generate all constant output in the proper language.
+# The default language is English, other supported languages are:
+# Afrikaans, Arabic, Brazilian, Catalan, Chinese, Chinese-Traditional,
+# Croatian, Czech, Danish, Dutch, Esperanto, Farsi, Finnish, French, German,
+# Greek, Hungarian, Italian, Japanese, Japanese-en (Japanese with English
+# messages), Korean, Korean-en, Lithuanian, Norwegian, Macedonian, Persian,
+# Polish, Portuguese, Romanian, Russian, Serbian, Serbian-Cyrilic, Slovak,
+# Slovene, Spanish, Swedish, Ukrainian, and Vietnamese.
+
+OUTPUT_LANGUAGE = English
+
+# If the BRIEF_MEMBER_DESC tag is set to YES (the default) Doxygen will
+# include brief member descriptions after the members that are listed in
+# the file and class documentation (similar to JavaDoc).
+# Set to NO to disable this.
+
+BRIEF_MEMBER_DESC = YES
+
+# If the REPEAT_BRIEF tag is set to YES (the default) Doxygen will prepend
+# the brief description of a member or function before the detailed description.
+# Note: if both HIDE_UNDOC_MEMBERS and BRIEF_MEMBER_DESC are set to NO, the
+# brief descriptions will be completely suppressed.
+
+REPEAT_BRIEF = YES
+
+# This tag implements a quasi-intelligent brief description abbreviator
+# that is used to form the text in various listings. Each string
+# in this list, if found as the leading text of the brief description, will be
+# stripped from the text and the result after processing the whole list, is
+# used as the annotated text. Otherwise, the brief description is used as-is.
+# If left blank, the following values are used ("$name" is automatically
+# replaced with the name of the entity): "The $name class" "The $name widget"
+# "The $name file" "is" "provides" "specifies" "contains"
+# "represents" "a" "an" "the"
+
+ABBREVIATE_BRIEF =
+
+# If the ALWAYS_DETAILED_SEC and REPEAT_BRIEF tags are both set to YES then
+# Doxygen will generate a detailed section even if there is only a brief
+# description.
+
+ALWAYS_DETAILED_SEC = NO
+
+# If the INLINE_INHERITED_MEMB tag is set to YES, doxygen will show all
+# inherited members of a class in the documentation of that class as if those
+# members were ordinary class members. Constructors, destructors and assignment
+# operators of the base classes will not be shown.
+
+INLINE_INHERITED_MEMB = NO
+
+# If the FULL_PATH_NAMES tag is set to YES then Doxygen will prepend the full
+# path before files name in the file list and in the header files. If set
+# to NO the shortest path that makes the file name unique will be used.
+
+FULL_PATH_NAMES = YES
+
+# If the FULL_PATH_NAMES tag is set to YES then the STRIP_FROM_PATH tag
+# can be used to strip a user-defined part of the path. Stripping is
+# only done if one of the specified strings matches the left-hand part of
+# the path. The tag can be used to show relative paths in the file list.
+# If left blank the directory from which doxygen is run is used as the
+# path to strip.
+
+STRIP_FROM_PATH =
+
+# The STRIP_FROM_INC_PATH tag can be used to strip a user-defined part of
+# the path mentioned in the documentation of a class, which tells
+# the reader which header file to include in order to use a class.
+# If left blank only the name of the header file containing the class
+# definition is used. Otherwise one should specify the include paths that
+# are normally passed to the compiler using the -I flag.
+
+STRIP_FROM_INC_PATH =
+
+# If the SHORT_NAMES tag is set to YES, doxygen will generate much shorter
+# (but less readable) file names. This can be useful is your file systems
+# doesn't support long names like on DOS, Mac, or CD-ROM.
+
+SHORT_NAMES = NO
+
+# If the JAVADOC_AUTOBRIEF tag is set to YES then Doxygen
+# will interpret the first line (until the first dot) of a JavaDoc-style
+# comment as the brief description. If set to NO, the JavaDoc
+# comments will behave just like regular Qt-style comments
+# (thus requiring an explicit @brief command for a brief description.)
+
+JAVADOC_AUTOBRIEF = NO
+
+# If the QT_AUTOBRIEF tag is set to YES then Doxygen will
+# interpret the first line (until the first dot) of a Qt-style
+# comment as the brief description. If set to NO, the comments
+# will behave just like regular Qt-style comments (thus requiring
+# an explicit \brief command for a brief description.)
+
+QT_AUTOBRIEF = NO
+
+# The MULTILINE_CPP_IS_BRIEF tag can be set to YES to make Doxygen
+# treat a multi-line C++ special comment block (i.e. a block of //! or ///
+# comments) as a brief description. This used to be the default behaviour.
+# The new default is to treat a multi-line C++ comment block as a detailed
+# description. Set this tag to YES if you prefer the old behaviour instead.
+
+MULTILINE_CPP_IS_BRIEF = NO
+
+# If the INHERIT_DOCS tag is set to YES (the default) then an undocumented
+# member inherits the documentation from any documented member that it
+# re-implements.
+
+INHERIT_DOCS = YES
+
+# If the SEPARATE_MEMBER_PAGES tag is set to YES, then doxygen will produce
+# a new page for each member. If set to NO, the documentation of a member will
+# be part of the file/class/namespace that contains it.
+
+SEPARATE_MEMBER_PAGES = NO
+
+# The TAB_SIZE tag can be used to set the number of spaces in a tab.
+# Doxygen uses this value to replace tabs by spaces in code fragments.
+
+TAB_SIZE = 4
+
+# This tag can be used to specify a number of aliases that acts
+# as commands in the documentation. An alias has the form "name=value".
+# For example adding "sideeffect=\par Side Effects:\n" will allow you to
+# put the command \sideeffect (or @sideeffect) in the documentation, which
+# will result in a user-defined paragraph with heading "Side Effects:".
+# You can put \n's in the value part of an alias to insert newlines.
+
+ALIASES =
+
+# Set the OPTIMIZE_OUTPUT_FOR_C tag to YES if your project consists of C
+# sources only. Doxygen will then generate output that is more tailored for C.
+# For instance, some of the names that are used will be different. The list
+# of all members will be omitted, etc.
+
+OPTIMIZE_OUTPUT_FOR_C = YES
+
+# Set the OPTIMIZE_OUTPUT_JAVA tag to YES if your project consists of Java
+# sources only. Doxygen will then generate output that is more tailored for
+# Java. For instance, namespaces will be presented as packages, qualified
+# scopes will look different, etc.
+
+OPTIMIZE_OUTPUT_JAVA = NO
+
+# Set the OPTIMIZE_FOR_FORTRAN tag to YES if your project consists of Fortran
+# sources only. Doxygen will then generate output that is more tailored for
+# Fortran.
+
+OPTIMIZE_FOR_FORTRAN = NO
+
+# Set the OPTIMIZE_OUTPUT_VHDL tag to YES if your project consists of VHDL
+# sources. Doxygen will then generate output that is tailored for
+# VHDL.
+
+OPTIMIZE_OUTPUT_VHDL = NO
+
+# Doxygen selects the parser to use depending on the extension of the files it
+# parses. With this tag you can assign which parser to use for a given extension.
+# Doxygen has a built-in mapping, but you can override or extend it using this
+# tag. The format is ext=language, where ext is a file extension, and language
+# is one of the parsers supported by doxygen: IDL, Java, Javascript, CSharp, C,
+# C++, D, PHP, Objective-C, Python, Fortran, VHDL, C, C++. For instance to make
+# doxygen treat .inc files as Fortran files (default is PHP), and .f files as C
+# (default is Fortran), use: inc=Fortran f=C. Note that for custom extensions
+# you also need to set FILE_PATTERNS otherwise the files are not read by doxygen.
+
+EXTENSION_MAPPING =
+
+# If you use STL classes (i.e. std::string, std::vector, etc.) but do not want
+# to include (a tag file for) the STL sources as input, then you should
+# set this tag to YES in order to let doxygen match functions declarations and
+# definitions whose arguments contain STL classes (e.g. func(std::string); v.s.
+# func(std::string) {}). This also make the inheritance and collaboration
+# diagrams that involve STL classes more complete and accurate.
+
+BUILTIN_STL_SUPPORT = NO
+
+# If you use Microsoft's C++/CLI language, you should set this option to YES to
+# enable parsing support.
+
+CPP_CLI_SUPPORT = NO
+
+# Set the SIP_SUPPORT tag to YES if your project consists of sip sources only.
+# Doxygen will parse them like normal C++ but will assume all classes use public
+# instead of private inheritance when no explicit protection keyword is present.
+
+SIP_SUPPORT = NO
+
+# For Microsoft's IDL there are propget and propput attributes to indicate getter
+# and setter methods for a property. Setting this option to YES (the default)
+# will make doxygen to replace the get and set methods by a property in the
+# documentation. This will only work if the methods are indeed getting or
+# setting a simple type. If this is not the case, or you want to show the
+# methods anyway, you should set this option to NO.
+
+IDL_PROPERTY_SUPPORT = YES
+
+# If member grouping is used in the documentation and the DISTRIBUTE_GROUP_DOC
+# tag is set to YES, then doxygen will reuse the documentation of the first
+# member in the group (if any) for the other members of the group. By default
+# all members of a group must be documented explicitly.
+
+DISTRIBUTE_GROUP_DOC = NO
+
+# Set the SUBGROUPING tag to YES (the default) to allow class member groups of
+# the same type (for instance a group of public functions) to be put as a
+# subgroup of that type (e.g. under the Public Functions section). Set it to
+# NO to prevent subgrouping. Alternatively, this can be done per class using
+# the \nosubgrouping command.
+
+SUBGROUPING = YES
+
+# When TYPEDEF_HIDES_STRUCT is enabled, a typedef of a struct, union, or enum
+# is documented as struct, union, or enum with the name of the typedef. So
+# typedef struct TypeS {} TypeT, will appear in the documentation as a struct
+# with name TypeT. When disabled the typedef will appear as a member of a file,
+# namespace, or class. And the struct will be named TypeS. This can typically
+# be useful for C code in case the coding convention dictates that all compound
+# types are typedef'ed and only the typedef is referenced, never the tag name.
+
+TYPEDEF_HIDES_STRUCT = NO
+
+# The SYMBOL_CACHE_SIZE determines the size of the internal cache use to
+# determine which symbols to keep in memory and which to flush to disk.
+# When the cache is full, less often used symbols will be written to disk.
+# For small to medium size projects (<1000 input files) the default value is
+# probably good enough. For larger projects a too small cache size can cause
+# doxygen to be busy swapping symbols to and from disk most of the time
+# causing a significant performance penality.
+# If the system has enough physical memory increasing the cache will improve the
+# performance by keeping more symbols in memory. Note that the value works on
+# a logarithmic scale so increasing the size by one will rougly double the
+# memory usage. The cache size is given by this formula:
+# 2^(16+SYMBOL_CACHE_SIZE). The valid range is 0..9, the default is 0,
+# corresponding to a cache size of 2^16 = 65536 symbols
+
+SYMBOL_CACHE_SIZE = 0
+
+#---------------------------------------------------------------------------
+# Build related configuration options
+#---------------------------------------------------------------------------
+
+# If the EXTRACT_ALL tag is set to YES doxygen will assume all entities in
+# documentation are documented, even if no documentation was available.
+# Private class members and static file members will be hidden unless
+# the EXTRACT_PRIVATE and EXTRACT_STATIC tags are set to YES
+
+EXTRACT_ALL = NO
+
+# If the EXTRACT_PRIVATE tag is set to YES all private members of a class
+# will be included in the documentation.
+
+EXTRACT_PRIVATE = NO
+
+# If the EXTRACT_STATIC tag is set to YES all static members of a file
+# will be included in the documentation.
+
+EXTRACT_STATIC = NO
+
+# If the EXTRACT_LOCAL_CLASSES tag is set to YES classes (and structs)
+# defined locally in source files will be included in the documentation.
+# If set to NO only classes defined in header files are included.
+
+EXTRACT_LOCAL_CLASSES = NO
+
+# This flag is only useful for Objective-C code. When set to YES local
+# methods, which are defined in the implementation section but not in
+# the interface are included in the documentation.
+# If set to NO (the default) only methods in the interface are included.
+
+EXTRACT_LOCAL_METHODS = NO
+
+# If this flag is set to YES, the members of anonymous namespaces will be
+# extracted and appear in the documentation as a namespace called
+# 'anonymous_namespace{file}', where file will be replaced with the base
+# name of the file that contains the anonymous namespace. By default
+# anonymous namespace are hidden.
+
+EXTRACT_ANON_NSPACES = NO
+
+# If the HIDE_UNDOC_MEMBERS tag is set to YES, Doxygen will hide all
+# undocumented members of documented classes, files or namespaces.
+# If set to NO (the default) these members will be included in the
+# various overviews, but no documentation section is generated.
+# This option has no effect if EXTRACT_ALL is enabled.
+
+HIDE_UNDOC_MEMBERS = NO
+
+# If the HIDE_UNDOC_CLASSES tag is set to YES, Doxygen will hide all
+# undocumented classes that are normally visible in the class hierarchy.
+# If set to NO (the default) these classes will be included in the various
+# overviews. This option has no effect if EXTRACT_ALL is enabled.
+
+HIDE_UNDOC_CLASSES = NO
+
+# If the HIDE_FRIEND_COMPOUNDS tag is set to YES, Doxygen will hide all
+# friend (class|struct|union) declarations.
+# If set to NO (the default) these declarations will be included in the
+# documentation.
+
+HIDE_FRIEND_COMPOUNDS = NO
+
+# If the HIDE_IN_BODY_DOCS tag is set to YES, Doxygen will hide any
+# documentation blocks found inside the body of a function.
+# If set to NO (the default) these blocks will be appended to the
+# function's detailed documentation block.
+
+HIDE_IN_BODY_DOCS = NO
+
+# The INTERNAL_DOCS tag determines if documentation
+# that is typed after a \internal command is included. If the tag is set
+# to NO (the default) then the documentation will be excluded.
+# Set it to YES to include the internal documentation.
+
+INTERNAL_DOCS = NO
+
+# If the CASE_SENSE_NAMES tag is set to NO then Doxygen will only generate
+# file names in lower-case letters. If set to YES upper-case letters are also
+# allowed. This is useful if you have classes or files whose names only differ
+# in case and if your file system supports case sensitive file names. Windows
+# and Mac users are advised to set this option to NO.
+
+CASE_SENSE_NAMES = YES
+
+# If the HIDE_SCOPE_NAMES tag is set to NO (the default) then Doxygen
+# will show members with their full class and namespace scopes in the
+# documentation. If set to YES the scope will be hidden.
+
+HIDE_SCOPE_NAMES = NO
+
+# If the SHOW_INCLUDE_FILES tag is set to YES (the default) then Doxygen
+# will put a list of the files that are included by a file in the documentation
+# of that file.
+
+SHOW_INCLUDE_FILES = YES
+
+# If the FORCE_LOCAL_INCLUDES tag is set to YES then Doxygen
+# will list include files with double quotes in the documentation
+# rather than with sharp brackets.
+
+FORCE_LOCAL_INCLUDES = NO
+
+# If the INLINE_INFO tag is set to YES (the default) then a tag [inline]
+# is inserted in the documentation for inline members.
+
+INLINE_INFO = YES
+
+# If the SORT_MEMBER_DOCS tag is set to YES (the default) then doxygen
+# will sort the (detailed) documentation of file and class members
+# alphabetically by member name. If set to NO the members will appear in
+# declaration order.
+
+SORT_MEMBER_DOCS = YES
+
+# If the SORT_BRIEF_DOCS tag is set to YES then doxygen will sort the
+# brief documentation of file, namespace and class members alphabetically
+# by member name. If set to NO (the default) the members will appear in
+# declaration order.
+
+SORT_BRIEF_DOCS = NO
+
+# If the SORT_MEMBERS_CTORS_1ST tag is set to YES then doxygen
+# will sort the (brief and detailed) documentation of class members so that
+# constructors and destructors are listed first. If set to NO (the default)
+# the constructors will appear in the respective orders defined by
+# SORT_MEMBER_DOCS and SORT_BRIEF_DOCS.
+# This tag will be ignored for brief docs if SORT_BRIEF_DOCS is set to NO
+# and ignored for detailed docs if SORT_MEMBER_DOCS is set to NO.
+
+SORT_MEMBERS_CTORS_1ST = NO
+
+# If the SORT_GROUP_NAMES tag is set to YES then doxygen will sort the
+# hierarchy of group names into alphabetical order. If set to NO (the default)
+# the group names will appear in their defined order.
+
+SORT_GROUP_NAMES = NO
+
+# If the SORT_BY_SCOPE_NAME tag is set to YES, the class list will be
+# sorted by fully-qualified names, including namespaces. If set to
+# NO (the default), the class list will be sorted only by class name,
+# not including the namespace part.
+# Note: This option is not very useful if HIDE_SCOPE_NAMES is set to YES.
+# Note: This option applies only to the class list, not to the
+# alphabetical list.
+
+SORT_BY_SCOPE_NAME = NO
+
+# The GENERATE_TODOLIST tag can be used to enable (YES) or
+# disable (NO) the todo list. This list is created by putting \todo
+# commands in the documentation.
+
+GENERATE_TODOLIST = YES
+
+# The GENERATE_TESTLIST tag can be used to enable (YES) or
+# disable (NO) the test list. This list is created by putting \test
+# commands in the documentation.
+
+GENERATE_TESTLIST = YES
+
+# The GENERATE_BUGLIST tag can be used to enable (YES) or
+# disable (NO) the bug list. This list is created by putting \bug
+# commands in the documentation.
+
+GENERATE_BUGLIST = YES
+
+# The GENERATE_DEPRECATEDLIST tag can be used to enable (YES) or
+# disable (NO) the deprecated list. This list is created by putting
+# \deprecated commands in the documentation.
+
+GENERATE_DEPRECATEDLIST= YES
+
+# The ENABLED_SECTIONS tag can be used to enable conditional
+# documentation sections, marked by \if sectionname ... \endif.
+
+ENABLED_SECTIONS =
+
+# The MAX_INITIALIZER_LINES tag determines the maximum number of lines
+# the initial value of a variable or define consists of for it to appear in
+# the documentation. If the initializer consists of more lines than specified
+# here it will be hidden. Use a value of 0 to hide initializers completely.
+# The appearance of the initializer of individual variables and defines in the
+# documentation can be controlled using \showinitializer or \hideinitializer
+# command in the documentation regardless of this setting.
+
+MAX_INITIALIZER_LINES = 30
+
+# Set the SHOW_USED_FILES tag to NO to disable the list of files generated
+# at the bottom of the documentation of classes and structs. If set to YES the
+# list will mention the files that were used to generate the documentation.
+
+SHOW_USED_FILES = YES
+
+# If the sources in your project are distributed over multiple directories
+# then setting the SHOW_DIRECTORIES tag to YES will show the directory hierarchy
+# in the documentation. The default is NO.
+
+SHOW_DIRECTORIES = NO
+
+# Set the SHOW_FILES tag to NO to disable the generation of the Files page.
+# This will remove the Files entry from the Quick Index and from the
+# Folder Tree View (if specified). The default is YES.
+
+SHOW_FILES = YES
+
+# Set the SHOW_NAMESPACES tag to NO to disable the generation of the
+# Namespaces page.
+# This will remove the Namespaces entry from the Quick Index
+# and from the Folder Tree View (if specified). The default is YES.
+
+SHOW_NAMESPACES = YES
+
+# The FILE_VERSION_FILTER tag can be used to specify a program or script that
+# doxygen should invoke to get the current version for each file (typically from
+# the version control system). Doxygen will invoke the program by executing (via
+# popen()) the command <command> <input-file>, where <command> is the value of
+# the FILE_VERSION_FILTER tag, and <input-file> is the name of an input file
+# provided by doxygen. Whatever the program writes to standard output
+# is used as the file version. See the manual for examples.
+
+FILE_VERSION_FILTER =
+
+# The LAYOUT_FILE tag can be used to specify a layout file which will be parsed
+# by doxygen. The layout file controls the global structure of the generated
+# output files in an output format independent way. The create the layout file
+# that represents doxygen's defaults, run doxygen with the -l option.
+# You can optionally specify a file name after the option, if omitted
+# DoxygenLayout.xml will be used as the name of the layout file.
+
+LAYOUT_FILE =
+
+#---------------------------------------------------------------------------
+# configuration options related to warning and progress messages
+#---------------------------------------------------------------------------
+
+# The QUIET tag can be used to turn on/off the messages that are generated
+# by doxygen. Possible values are YES and NO. If left blank NO is used.
+
+QUIET = NO
+
+# The WARNINGS tag can be used to turn on/off the warning messages that are
+# generated by doxygen. Possible values are YES and NO. If left blank
+# NO is used.
+
+WARNINGS = YES
+
+# If WARN_IF_UNDOCUMENTED is set to YES, then doxygen will generate warnings
+# for undocumented members. If EXTRACT_ALL is set to YES then this flag will
+# automatically be disabled.
+
+WARN_IF_UNDOCUMENTED = YES
+
+# If WARN_IF_DOC_ERROR is set to YES, doxygen will generate warnings for
+# potential errors in the documentation, such as not documenting some
+# parameters in a documented function, or documenting parameters that
+# don't exist or using markup commands wrongly.
+
+WARN_IF_DOC_ERROR = YES
+
+# This WARN_NO_PARAMDOC option can be abled to get warnings for
+# functions that are documented, but have no documentation for their parameters
+# or return value. If set to NO (the default) doxygen will only warn about
+# wrong or incomplete parameter documentation, but not about the absence of
+# documentation.
+
+WARN_NO_PARAMDOC = NO
+
+# The WARN_FORMAT tag determines the format of the warning messages that
+# doxygen can produce. The string should contain the $file, $line, and $text
+# tags, which will be replaced by the file and line number from which the
+# warning originated and the warning text. Optionally the format may contain
+# $version, which will be replaced by the version of the file (if it could
+# be obtained via FILE_VERSION_FILTER)
+
+WARN_FORMAT = "$file:$line: $text"
+
+# The WARN_LOGFILE tag can be used to specify a file to which warning
+# and error messages should be written. If left blank the output is written
+# to stderr.
+
+WARN_LOGFILE =
+
+#---------------------------------------------------------------------------
+# configuration options related to the input files
+#---------------------------------------------------------------------------
+
+# The INPUT tag can be used to specify the files and/or directories that contain
+# documented source files. You may enter file names like "myfile.cpp" or
+# directories like "/usr/src/myproject". Separate the files or directories
+# with spaces.
+
+INPUT = mdb.h
+
+# This tag can be used to specify the character encoding of the source files
+# that doxygen parses. Internally doxygen uses the UTF-8 encoding, which is
+# also the default input encoding. Doxygen uses libiconv (or the iconv built
+# into libc) for the transcoding. See http://www.gnu.org/software/libiconv for
+# the list of possible encodings.
+
+INPUT_ENCODING = UTF-8
+
+# If the value of the INPUT tag contains directories, you can use the
+# FILE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp
+# and *.h) to filter out the source-files in the directories. If left
+# blank the following patterns are tested:
+# *.c *.cc *.cxx *.cpp *.c++ *.java *.ii *.ixx *.ipp *.i++ *.inl *.h *.hh *.hxx
+# *.hpp *.h++ *.idl *.odl *.cs *.php *.php3 *.inc *.m *.mm *.py *.f90
+
+FILE_PATTERNS =
+
+# The RECURSIVE tag can be used to turn specify whether or not subdirectories
+# should be searched for input files as well. Possible values are YES and NO.
+# If left blank NO is used.
+
+RECURSIVE = NO
+
+# The EXCLUDE tag can be used to specify files and/or directories that should
+# excluded from the INPUT source files. This way you can easily exclude a
+# subdirectory from a directory tree whose root is specified with the INPUT tag.
+
+EXCLUDE =
+
+# The EXCLUDE_SYMLINKS tag can be used select whether or not files or
+# directories that are symbolic links (a Unix filesystem feature) are excluded
+# from the input.
+
+EXCLUDE_SYMLINKS = NO
+
+# If the value of the INPUT tag contains directories, you can use the
+# EXCLUDE_PATTERNS tag to specify one or more wildcard patterns to exclude
+# certain files from those directories. Note that the wildcards are matched
+# against the file with absolute path, so to exclude all test directories
+# for example use the pattern */test/*
+
+EXCLUDE_PATTERNS =
+
+# The EXCLUDE_SYMBOLS tag can be used to specify one or more symbol names
+# (namespaces, classes, functions, etc.) that should be excluded from the
+# output. The symbol name can be a fully qualified name, a word, or if the
+# wildcard * is used, a substring. Examples: ANamespace, AClass,
+# AClass::ANamespace, ANamespace::*Test
+
+EXCLUDE_SYMBOLS =
+
+# The EXAMPLE_PATH tag can be used to specify one or more files or
+# directories that contain example code fragments that are included (see
+# the \include command).
+
+EXAMPLE_PATH =
+
+# If the value of the EXAMPLE_PATH tag contains directories, you can use the
+# EXAMPLE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp
+# and *.h) to filter out the source-files in the directories. If left
+# blank all files are included.
+
+EXAMPLE_PATTERNS =
+
+# If the EXAMPLE_RECURSIVE tag is set to YES then subdirectories will be
+# searched for input files to be used with the \include or \dontinclude
+# commands irrespective of the value of the RECURSIVE tag.
+# Possible values are YES and NO. If left blank NO is used.
+
+EXAMPLE_RECURSIVE = NO
+
+# The IMAGE_PATH tag can be used to specify one or more files or
+# directories that contain image that are included in the documentation (see
+# the \image command).
+
+IMAGE_PATH =
+
+# The INPUT_FILTER tag can be used to specify a program that doxygen should
+# invoke to filter for each input file. Doxygen will invoke the filter program
+# by executing (via popen()) the command <filter> <input-file>, where <filter>
+# is the value of the INPUT_FILTER tag, and <input-file> is the name of an
+# input file. Doxygen will then use the output that the filter program writes
+# to standard output.
+# If FILTER_PATTERNS is specified, this tag will be
+# ignored.
+
+INPUT_FILTER =
+
+# The FILTER_PATTERNS tag can be used to specify filters on a per file pattern
+# basis.
+# Doxygen will compare the file name with each pattern and apply the
+# filter if there is a match.
+# The filters are a list of the form:
+# pattern=filter (like *.cpp=my_cpp_filter). See INPUT_FILTER for further
+# info on how filters are used. If FILTER_PATTERNS is empty, INPUT_FILTER
+# is applied to all files.
+
+FILTER_PATTERNS =
+
+# If the FILTER_SOURCE_FILES tag is set to YES, the input filter (if set using
+# INPUT_FILTER) will be used to filter the input files when producing source
+# files to browse (i.e. when SOURCE_BROWSER is set to YES).
+
+FILTER_SOURCE_FILES = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to source browsing
+#---------------------------------------------------------------------------
+
+# If the SOURCE_BROWSER tag is set to YES then a list of source files will
+# be generated. Documented entities will be cross-referenced with these sources.
+# Note: To get rid of all source code in the generated output, make sure also
+# VERBATIM_HEADERS is set to NO.
+
+SOURCE_BROWSER = NO
+
+# Setting the INLINE_SOURCES tag to YES will include the body
+# of functions and classes directly in the documentation.
+
+INLINE_SOURCES = NO
+
+# Setting the STRIP_CODE_COMMENTS tag to YES (the default) will instruct
+# doxygen to hide any special comment blocks from generated source code
+# fragments. Normal C and C++ comments will always remain visible.
+
+STRIP_CODE_COMMENTS = YES
+
+# If the REFERENCED_BY_RELATION tag is set to YES
+# then for each documented function all documented
+# functions referencing it will be listed.
+
+REFERENCED_BY_RELATION = NO
+
+# If the REFERENCES_RELATION tag is set to YES
+# then for each documented function all documented entities
+# called/used by that function will be listed.
+
+REFERENCES_RELATION = NO
+
+# If the REFERENCES_LINK_SOURCE tag is set to YES (the default)
+# and SOURCE_BROWSER tag is set to YES, then the hyperlinks from
+# functions in REFERENCES_RELATION and REFERENCED_BY_RELATION lists will
+# link to the source code.
+# Otherwise they will link to the documentation.
+
+REFERENCES_LINK_SOURCE = YES
+
+# If the USE_HTAGS tag is set to YES then the references to source code
+# will point to the HTML generated by the htags(1) tool instead of doxygen
+# built-in source browser. The htags tool is part of GNU's global source
+# tagging system (see http://www.gnu.org/software/global/global.html). You
+# will need version 4.8.6 or higher.
+
+USE_HTAGS = NO
+
+# If the VERBATIM_HEADERS tag is set to YES (the default) then Doxygen
+# will generate a verbatim copy of the header file for each class for
+# which an include is specified. Set to NO to disable this.
+
+VERBATIM_HEADERS = YES
+
+#---------------------------------------------------------------------------
+# configuration options related to the alphabetical class index
+#---------------------------------------------------------------------------
+
+# If the ALPHABETICAL_INDEX tag is set to YES, an alphabetical index
+# of all compounds will be generated. Enable this if the project
+# contains a lot of classes, structs, unions or interfaces.
+
+ALPHABETICAL_INDEX = YES
+
+# If the alphabetical index is enabled (see ALPHABETICAL_INDEX) then
+# the COLS_IN_ALPHA_INDEX tag can be used to specify the number of columns
+# in which this list will be split (can be a number in the range [1..20])
+
+COLS_IN_ALPHA_INDEX = 5
+
+# In case all classes in a project start with a common prefix, all
+# classes will be put under the same header in the alphabetical index.
+# The IGNORE_PREFIX tag can be used to specify one or more prefixes that
+# should be ignored while generating the index headers.
+
+IGNORE_PREFIX =
+
+#---------------------------------------------------------------------------
+# configuration options related to the HTML output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_HTML tag is set to YES (the default) Doxygen will
+# generate HTML output.
+
+GENERATE_HTML = YES
+
+# The HTML_OUTPUT tag is used to specify where the HTML docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `html' will be used as the default path.
+
+HTML_OUTPUT = html
+
+# The HTML_FILE_EXTENSION tag can be used to specify the file extension for
+# each generated HTML page (for example: .htm,.php,.asp). If it is left blank
+# doxygen will generate files with .html extension.
+
+HTML_FILE_EXTENSION = .html
+
+# The HTML_HEADER tag can be used to specify a personal HTML header for
+# each generated HTML page. If it is left blank doxygen will generate a
+# standard header.
+
+HTML_HEADER =
+
+# The HTML_FOOTER tag can be used to specify a personal HTML footer for
+# each generated HTML page. If it is left blank doxygen will generate a
+# standard footer.
+
+HTML_FOOTER =
+
+# The HTML_STYLESHEET tag can be used to specify a user-defined cascading
+# style sheet that is used by each HTML page. It can be used to
+# fine-tune the look of the HTML output. If the tag is left blank doxygen
+# will generate a default style sheet. Note that doxygen will try to copy
+# the style sheet file to the HTML output directory, so don't put your own
+# stylesheet in the HTML output directory as well, or it will be erased!
+
+HTML_STYLESHEET =
+
+# The HTML_COLORSTYLE_HUE tag controls the color of the HTML output.
+# Doxygen will adjust the colors in the stylesheet and background images
+# according to this color. Hue is specified as an angle on a colorwheel,
+# see http://en.wikipedia.org/wiki/Hue for more information.
+# For instance the value 0 represents red, 60 is yellow, 120 is green,
+# 180 is cyan, 240 is blue, 300 purple, and 360 is red again.
+# The allowed range is 0 to 359.
+
+HTML_COLORSTYLE_HUE = 220
+
+# The HTML_COLORSTYLE_SAT tag controls the purity (or saturation) of
+# the colors in the HTML output. For a value of 0 the output will use
+# grayscales only. A value of 255 will produce the most vivid colors.
+
+HTML_COLORSTYLE_SAT = 100
+
+# The HTML_COLORSTYLE_GAMMA tag controls the gamma correction applied to
+# the luminance component of the colors in the HTML output. Values below
+# 100 gradually make the output lighter, whereas values above 100 make
+# the output darker. The value divided by 100 is the actual gamma applied,
+# so 80 represents a gamma of 0.8, The value 220 represents a gamma of 2.2,
+# and 100 does not change the gamma.
+
+HTML_COLORSTYLE_GAMMA = 80
+
+# If the HTML_TIMESTAMP tag is set to YES then the footer of each generated HTML
+# page will contain the date and time when the page was generated. Setting
+# this to NO can help when comparing the output of multiple runs.
+
+HTML_TIMESTAMP = YES
+
+# If the HTML_ALIGN_MEMBERS tag is set to YES, the members of classes,
+# files or namespaces will be aligned in HTML using tables. If set to
+# NO a bullet list will be used.
+
+HTML_ALIGN_MEMBERS = YES
+
+# If the HTML_DYNAMIC_SECTIONS tag is set to YES then the generated HTML
+# documentation will contain sections that can be hidden and shown after the
+# page has loaded. For this to work a browser that supports
+# JavaScript and DHTML is required (for instance Mozilla 1.0+, Firefox
+# Netscape 6.0+, Internet explorer 5.0+, Konqueror, or Safari).
+
+HTML_DYNAMIC_SECTIONS = NO
+
+# If the GENERATE_DOCSET tag is set to YES, additional index files
+# will be generated that can be used as input for Apple's Xcode 3
+# integrated development environment, introduced with OSX 10.5 (Leopard).
+# To create a documentation set, doxygen will generate a Makefile in the
+# HTML output directory. Running make will produce the docset in that
+# directory and running "make install" will install the docset in
+# ~/Library/Developer/Shared/Documentation/DocSets so that Xcode will find
+# it at startup.
+# See http://developer.apple.com/tools/creatingdocsetswithdoxygen.html
+# for more information.
+
+GENERATE_DOCSET = NO
+
+# When GENERATE_DOCSET tag is set to YES, this tag determines the name of the
+# feed. A documentation feed provides an umbrella under which multiple
+# documentation sets from a single provider (such as a company or product suite)
+# can be grouped.
+
+DOCSET_FEEDNAME = "Doxygen generated docs"
+
+# When GENERATE_DOCSET tag is set to YES, this tag specifies a string that
+# should uniquely identify the documentation set bundle. This should be a
+# reverse domain-name style string, e.g. com.mycompany.MyDocSet. Doxygen
+# will append .docset to the name.
+
+DOCSET_BUNDLE_ID = org.doxygen.Project
+
+# When GENERATE_PUBLISHER_ID tag specifies a string that should uniquely identify
+# the documentation publisher. This should be a reverse domain-name style
+# string, e.g. com.mycompany.MyDocSet.documentation.
+
+DOCSET_PUBLISHER_ID = org.doxygen.Publisher
+
+# The GENERATE_PUBLISHER_NAME tag identifies the documentation publisher.
+
+DOCSET_PUBLISHER_NAME = Publisher
+
+# If the GENERATE_HTMLHELP tag is set to YES, additional index files
+# will be generated that can be used as input for tools like the
+# Microsoft HTML help workshop to generate a compiled HTML help file (.chm)
+# of the generated HTML documentation.
+
+GENERATE_HTMLHELP = NO
+
+# If the GENERATE_HTMLHELP tag is set to YES, the CHM_FILE tag can
+# be used to specify the file name of the resulting .chm file. You
+# can add a path in front of the file if the result should not be
+# written to the html output directory.
+
+CHM_FILE =
+
+# If the GENERATE_HTMLHELP tag is set to YES, the HHC_LOCATION tag can
+# be used to specify the location (absolute path including file name) of
+# the HTML help compiler (hhc.exe). If non-empty doxygen will try to run
+# the HTML help compiler on the generated index.hhp.
+
+HHC_LOCATION =
+
+# If the GENERATE_HTMLHELP tag is set to YES, the GENERATE_CHI flag
+# controls if a separate .chi index file is generated (YES) or that
+# it should be included in the master .chm file (NO).
+
+GENERATE_CHI = NO
+
+# If the GENERATE_HTMLHELP tag is set to YES, the CHM_INDEX_ENCODING
+# is used to encode HtmlHelp index (hhk), content (hhc) and project file
+# content.
+
+CHM_INDEX_ENCODING =
+
+# If the GENERATE_HTMLHELP tag is set to YES, the BINARY_TOC flag
+# controls whether a binary table of contents is generated (YES) or a
+# normal table of contents (NO) in the .chm file.
+
+BINARY_TOC = NO
+
+# The TOC_EXPAND flag can be set to YES to add extra items for group members
+# to the contents of the HTML help documentation and to the tree view.
+
+TOC_EXPAND = NO
+
+# If the GENERATE_QHP tag is set to YES and both QHP_NAMESPACE and
+# QHP_VIRTUAL_FOLDER are set, an additional index file will be generated
+# that can be used as input for Qt's qhelpgenerator to generate a
+# Qt Compressed Help (.qch) of the generated HTML documentation.
+
+GENERATE_QHP = NO
+
+# If the QHG_LOCATION tag is specified, the QCH_FILE tag can
+# be used to specify the file name of the resulting .qch file.
+# The path specified is relative to the HTML output folder.
+
+QCH_FILE =
+
+# The QHP_NAMESPACE tag specifies the namespace to use when generating
+# Qt Help Project output. For more information please see
+# http://doc.trolltech.com/qthelpproject.html#namespace
+
+QHP_NAMESPACE = org.doxygen.Project
+
+# The QHP_VIRTUAL_FOLDER tag specifies the namespace to use when generating
+# Qt Help Project output. For more information please see
+# http://doc.trolltech.com/qthelpproject.html#virtual-folders
+
+QHP_VIRTUAL_FOLDER = doc
+
+# If QHP_CUST_FILTER_NAME is set, it specifies the name of a custom filter to
+# add. For more information please see
+# http://doc.trolltech.com/qthelpproject.html#custom-filters
+
+QHP_CUST_FILTER_NAME =
+
+# The QHP_CUST_FILT_ATTRS tag specifies the list of the attributes of the
+# custom filter to add. For more information please see
+# <a href="http://doc.trolltech.com/qthelpproject.html#custom-filters">
+# Qt Help Project / Custom Filters</a>.
+
+QHP_CUST_FILTER_ATTRS =
+
+# The QHP_SECT_FILTER_ATTRS tag specifies the list of the attributes this
+# project's
+# filter section matches.
+# <a href="http://doc.trolltech.com/qthelpproject.html#filter-attributes">
+# Qt Help Project / Filter Attributes</a>.
+
+QHP_SECT_FILTER_ATTRS =
+
+# If the GENERATE_QHP tag is set to YES, the QHG_LOCATION tag can
+# be used to specify the location of Qt's qhelpgenerator.
+# If non-empty doxygen will try to run qhelpgenerator on the generated
+# .qhp file.
+
+QHG_LOCATION =
+
+# If the GENERATE_ECLIPSEHELP tag is set to YES, additional index files
+# will be generated, which together with the HTML files, form an Eclipse help
+# plugin. To install this plugin and make it available under the help contents
+# menu in Eclipse, the contents of the directory containing the HTML and XML
+# files needs to be copied into the plugins directory of eclipse. The name of
+# the directory within the plugins directory should be the same as
+# the ECLIPSE_DOC_ID value. After copying Eclipse needs to be restarted before
+# the help appears.
+
+GENERATE_ECLIPSEHELP = NO
+
+# A unique identifier for the eclipse help plugin. When installing the plugin
+# the directory name containing the HTML and XML files should also have
+# this name.
+
+ECLIPSE_DOC_ID = org.doxygen.Project
+
+# The DISABLE_INDEX tag can be used to turn on/off the condensed index at
+# top of each HTML page. The value NO (the default) enables the index and
+# the value YES disables it.
+
+DISABLE_INDEX = NO
+
+# This tag can be used to set the number of enum values (range [1..20])
+# that doxygen will group on one line in the generated HTML documentation.
+
+ENUM_VALUES_PER_LINE = 4
+
+# The GENERATE_TREEVIEW tag is used to specify whether a tree-like index
+# structure should be generated to display hierarchical information.
+# If the tag value is set to YES, a side panel will be generated
+# containing a tree-like index structure (just like the one that
+# is generated for HTML Help). For this to work a browser that supports
+# JavaScript, DHTML, CSS and frames is required (i.e. any modern browser).
+# Windows users are probably better off using the HTML help feature.
+
+GENERATE_TREEVIEW = NO
+
+# By enabling USE_INLINE_TREES, doxygen will generate the Groups, Directories,
+# and Class Hierarchy pages using a tree view instead of an ordered list.
+
+USE_INLINE_TREES = NO
+
+# If the treeview is enabled (see GENERATE_TREEVIEW) then this tag can be
+# used to set the initial width (in pixels) of the frame in which the tree
+# is shown.
+
+TREEVIEW_WIDTH = 250
+
+# When the EXT_LINKS_IN_WINDOW option is set to YES doxygen will open
+# links to external symbols imported via tag files in a separate window.
+
+EXT_LINKS_IN_WINDOW = NO
+
+# Use this tag to change the font size of Latex formulas included
+# as images in the HTML documentation. The default is 10. Note that
+# when you change the font size after a successful doxygen run you need
+# to manually remove any form_*.png images from the HTML output directory
+# to force them to be regenerated.
+
+FORMULA_FONTSIZE = 10
+
+# Use the FORMULA_TRANPARENT tag to determine whether or not the images
+# generated for formulas are transparent PNGs. Transparent PNGs are
+# not supported properly for IE 6.0, but are supported on all modern browsers.
+# Note that when changing this option you need to delete any form_*.png files
+# in the HTML output before the changes have effect.
+
+FORMULA_TRANSPARENT = YES
+
+# When the SEARCHENGINE tag is enabled doxygen will generate a search box
+# for the HTML output. The underlying search engine uses javascript
+# and DHTML and should work on any modern browser. Note that when using
+# HTML help (GENERATE_HTMLHELP), Qt help (GENERATE_QHP), or docsets
+# (GENERATE_DOCSET) there is already a search function so this one should
+# typically be disabled. For large projects the javascript based search engine
+# can be slow, then enabling SERVER_BASED_SEARCH may provide a better solution.
+
+SEARCHENGINE = YES
+
+# When the SERVER_BASED_SEARCH tag is enabled the search engine will be
+# implemented using a PHP enabled web server instead of at the web client
+# using Javascript. Doxygen will generate the search PHP script and index
+# file to put on the web server. The advantage of the server
+# based approach is that it scales better to large projects and allows
+# full text search. The disadvances is that it is more difficult to setup
+# and does not have live searching capabilities.
+
+SERVER_BASED_SEARCH = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the LaTeX output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_LATEX tag is set to YES (the default) Doxygen will
+# generate Latex output.
+
+GENERATE_LATEX = NO
+
+# The LATEX_OUTPUT tag is used to specify where the LaTeX docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `latex' will be used as the default path.
+
+LATEX_OUTPUT = latex
+
+# The LATEX_CMD_NAME tag can be used to specify the LaTeX command name to be
+# invoked. If left blank `latex' will be used as the default command name.
+# Note that when enabling USE_PDFLATEX this option is only used for
+# generating bitmaps for formulas in the HTML output, but not in the
+# Makefile that is written to the output directory.
+
+LATEX_CMD_NAME = latex
+
+# The MAKEINDEX_CMD_NAME tag can be used to specify the command name to
+# generate index for LaTeX. If left blank `makeindex' will be used as the
+# default command name.
+
+MAKEINDEX_CMD_NAME = makeindex
+
+# If the COMPACT_LATEX tag is set to YES Doxygen generates more compact
+# LaTeX documents. This may be useful for small projects and may help to
+# save some trees in general.
+
+COMPACT_LATEX = NO
+
+# The PAPER_TYPE tag can be used to set the paper type that is used
+# by the printer. Possible values are: a4, a4wide, letter, legal and
+# executive. If left blank a4wide will be used.
+
+PAPER_TYPE = a4wide
+
+# The EXTRA_PACKAGES tag can be to specify one or more names of LaTeX
+# packages that should be included in the LaTeX output.
+
+EXTRA_PACKAGES =
+
+# The LATEX_HEADER tag can be used to specify a personal LaTeX header for
+# the generated latex document. The header should contain everything until
+# the first chapter. If it is left blank doxygen will generate a
+# standard header. Notice: only use this tag if you know what you are doing!
+
+LATEX_HEADER =
+
+# If the PDF_HYPERLINKS tag is set to YES, the LaTeX that is generated
+# is prepared for conversion to pdf (using ps2pdf). The pdf file will
+# contain links (just like the HTML output) instead of page references
+# This makes the output suitable for online browsing using a pdf viewer.
+
+PDF_HYPERLINKS = YES
+
+# If the USE_PDFLATEX tag is set to YES, pdflatex will be used instead of
+# plain latex in the generated Makefile. Set this option to YES to get a
+# higher quality PDF documentation.
+
+USE_PDFLATEX = YES
+
+# If the LATEX_BATCHMODE tag is set to YES, doxygen will add the \\batchmode.
+# command to the generated LaTeX files. This will instruct LaTeX to keep
+# running if errors occur, instead of asking the user for help.
+# This option is also used when generating formulas in HTML.
+
+LATEX_BATCHMODE = NO
+
+# If LATEX_HIDE_INDICES is set to YES then doxygen will not
+# include the index chapters (such as File Index, Compound Index, etc.)
+# in the output.
+
+LATEX_HIDE_INDICES = NO
+
+# If LATEX_SOURCE_CODE is set to YES then doxygen will include
+# source code with syntax highlighting in the LaTeX output.
+# Note that which sources are shown also depends on other settings
+# such as SOURCE_BROWSER.
+
+LATEX_SOURCE_CODE = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the RTF output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_RTF tag is set to YES Doxygen will generate RTF output
+# The RTF output is optimized for Word 97 and may not look very pretty with
+# other RTF readers or editors.
+
+GENERATE_RTF = NO
+
+# The RTF_OUTPUT tag is used to specify where the RTF docs will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `rtf' will be used as the default path.
+
+RTF_OUTPUT = rtf
+
+# If the COMPACT_RTF tag is set to YES Doxygen generates more compact
+# RTF documents. This may be useful for small projects and may help to
+# save some trees in general.
+
+COMPACT_RTF = NO
+
+# If the RTF_HYPERLINKS tag is set to YES, the RTF that is generated
+# will contain hyperlink fields. The RTF file will
+# contain links (just like the HTML output) instead of page references.
+# This makes the output suitable for online browsing using WORD or other
+# programs which support those fields.
+# Note: wordpad (write) and others do not support links.
+
+RTF_HYPERLINKS = NO
+
+# Load stylesheet definitions from file. Syntax is similar to doxygen's
+# config file, i.e. a series of assignments. You only have to provide
+# replacements, missing definitions are set to their default value.
+
+RTF_STYLESHEET_FILE =
+
+# Set optional variables used in the generation of an rtf document.
+# Syntax is similar to doxygen's config file.
+
+RTF_EXTENSIONS_FILE =
+
+#---------------------------------------------------------------------------
+# configuration options related to the man page output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_MAN tag is set to YES (the default) Doxygen will
+# generate man pages
+
+GENERATE_MAN = YES
+
+# The MAN_OUTPUT tag is used to specify where the man pages will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `man' will be used as the default path.
+
+MAN_OUTPUT = man
+
+# The MAN_EXTENSION tag determines the extension that is added to
+# the generated man pages (default is the subroutine's section .3)
+
+MAN_EXTENSION = .3
+
+# If the MAN_LINKS tag is set to YES and Doxygen generates man output,
+# then it will generate one additional man file for each entity
+# documented in the real man page(s). These additional files
+# only source the real man page, but without them the man command
+# would be unable to find the correct page. The default is NO.
+
+MAN_LINKS = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the XML output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_XML tag is set to YES Doxygen will
+# generate an XML file that captures the structure of
+# the code including all documentation.
+
+GENERATE_XML = NO
+
+# The XML_OUTPUT tag is used to specify where the XML pages will be put.
+# If a relative path is entered the value of OUTPUT_DIRECTORY will be
+# put in front of it. If left blank `xml' will be used as the default path.
+
+XML_OUTPUT = xml
+
+# The XML_SCHEMA tag can be used to specify an XML schema,
+# which can be used by a validating XML parser to check the
+# syntax of the XML files.
+
+XML_SCHEMA =
+
+# The XML_DTD tag can be used to specify an XML DTD,
+# which can be used by a validating XML parser to check the
+# syntax of the XML files.
+
+XML_DTD =
+
+# If the XML_PROGRAMLISTING tag is set to YES Doxygen will
+# dump the program listings (including syntax highlighting
+# and cross-referencing information) to the XML output. Note that
+# enabling this will significantly increase the size of the XML output.
+
+XML_PROGRAMLISTING = YES
+
+#---------------------------------------------------------------------------
+# configuration options for the AutoGen Definitions output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_AUTOGEN_DEF tag is set to YES Doxygen will
+# generate an AutoGen Definitions (see autogen.sf.net) file
+# that captures the structure of the code including all
+# documentation. Note that this feature is still experimental
+# and incomplete at the moment.
+
+GENERATE_AUTOGEN_DEF = NO
+
+#---------------------------------------------------------------------------
+# configuration options related to the Perl module output
+#---------------------------------------------------------------------------
+
+# If the GENERATE_PERLMOD tag is set to YES Doxygen will
+# generate a Perl module file that captures the structure of
+# the code including all documentation. Note that this
+# feature is still experimental and incomplete at the
+# moment.
+
+GENERATE_PERLMOD = NO
+
+# If the PERLMOD_LATEX tag is set to YES Doxygen will generate
+# the necessary Makefile rules, Perl scripts and LaTeX code to be able
+# to generate PDF and DVI output from the Perl module output.
+
+PERLMOD_LATEX = NO
+
+# If the PERLMOD_PRETTY tag is set to YES the Perl module output will be
+# nicely formatted so it can be parsed by a human reader.
+# This is useful
+# if you want to understand what is going on.
+# On the other hand, if this
+# tag is set to NO the size of the Perl module output will be much smaller
+# and Perl will parse it just the same.
+
+PERLMOD_PRETTY = YES
+
+# The names of the make variables in the generated doxyrules.make file
+# are prefixed with the string contained in PERLMOD_MAKEVAR_PREFIX.
+# This is useful so different doxyrules.make files included by the same
+# Makefile don't overwrite each other's variables.
+
+PERLMOD_MAKEVAR_PREFIX =
+
+#---------------------------------------------------------------------------
+# Configuration options related to the preprocessor
+#---------------------------------------------------------------------------
+
+# If the ENABLE_PREPROCESSING tag is set to YES (the default) Doxygen will
+# evaluate all C-preprocessor directives found in the sources and include
+# files.
+
+ENABLE_PREPROCESSING = YES
+
+# If the MACRO_EXPANSION tag is set to YES Doxygen will expand all macro
+# names in the source code. If set to NO (the default) only conditional
+# compilation will be performed. Macro expansion can be done in a controlled
+# way by setting EXPAND_ONLY_PREDEF to YES.
+
+MACRO_EXPANSION = NO
+
+# If the EXPAND_ONLY_PREDEF and MACRO_EXPANSION tags are both set to YES
+# then the macro expansion is limited to the macros specified with the
+# PREDEFINED and EXPAND_AS_DEFINED tags.
+
+EXPAND_ONLY_PREDEF = NO
+
+# If the SEARCH_INCLUDES tag is set to YES (the default) the includes files
+# in the INCLUDE_PATH (see below) will be search if a #include is found.
+
+SEARCH_INCLUDES = YES
+
+# The INCLUDE_PATH tag can be used to specify one or more directories that
+# contain include files that are not input files but should be processed by
+# the preprocessor.
+
+INCLUDE_PATH =
+
+# You can use the INCLUDE_FILE_PATTERNS tag to specify one or more wildcard
+# patterns (like *.h and *.hpp) to filter out the header-files in the
+# directories. If left blank, the patterns specified with FILE_PATTERNS will
+# be used.
+
+INCLUDE_FILE_PATTERNS =
+
+# The PREDEFINED tag can be used to specify one or more macro names that
+# are defined before the preprocessor is started (similar to the -D option of
+# gcc). The argument of the tag is a list of macros of the form: name
+# or name=definition (no spaces). If the definition and the = are
+# omitted =1 is assumed. To prevent a macro definition from being
+# undefined via #undef or recursively expanded use the := operator
+# instead of the = operator.
+
+PREDEFINED =
+
+# If the MACRO_EXPANSION and EXPAND_ONLY_PREDEF tags are set to YES then
+# this tag can be used to specify a list of macro names that should be expanded.
+# The macro definition that is found in the sources will be used.
+# Use the PREDEFINED tag if you want to use a different macro definition.
+
+EXPAND_AS_DEFINED =
+
+# If the SKIP_FUNCTION_MACROS tag is set to YES (the default) then
+# doxygen's preprocessor will remove all function-like macros that are alone
+# on a line, have an all uppercase name, and do not end with a semicolon. Such
+# function macros are typically used for boiler-plate code, and will confuse
+# the parser if not removed.
+
+SKIP_FUNCTION_MACROS = YES
+
+#---------------------------------------------------------------------------
+# Configuration::additions related to external references
+#---------------------------------------------------------------------------
+
+# The TAGFILES option can be used to specify one or more tagfiles.
+# Optionally an initial location of the external documentation
+# can be added for each tagfile. The format of a tag file without
+# this location is as follows:
+#
+# TAGFILES = file1 file2 ...
+# Adding location for the tag files is done as follows:
+#
+# TAGFILES = file1=loc1 "file2 = loc2" ...
+# where "loc1" and "loc2" can be relative or absolute paths or
+# URLs. If a location is present for each tag, the installdox tool
+# does not have to be run to correct the links.
+# Note that each tag file must have a unique name
+# (where the name does NOT include the path)
+# If a tag file is not located in the directory in which doxygen
+# is run, you must also specify the path to the tagfile here.
+
+TAGFILES =
+
+# When a file name is specified after GENERATE_TAGFILE, doxygen will create
+# a tag file that is based on the input files it reads.
+
+GENERATE_TAGFILE =
+
+# If the ALLEXTERNALS tag is set to YES all external classes will be listed
+# in the class index. If set to NO only the inherited external classes
+# will be listed.
+
+ALLEXTERNALS = NO
+
+# If the EXTERNAL_GROUPS tag is set to YES all external groups will be listed
+# in the modules index. If set to NO, only the current project's groups will
+# be listed.
+
+EXTERNAL_GROUPS = YES
+
+# The PERL_PATH should be the absolute path and name of the perl script
+# interpreter (i.e. the result of `which perl').
+
+PERL_PATH = /usr/bin/perl
+
+#---------------------------------------------------------------------------
+# Configuration options related to the dot tool
+#---------------------------------------------------------------------------
+
+# If the CLASS_DIAGRAMS tag is set to YES (the default) Doxygen will
+# generate a inheritance diagram (in HTML, RTF and LaTeX) for classes with base
+# or super classes. Setting the tag to NO turns the diagrams off. Note that
+# this option is superseded by the HAVE_DOT option below. This is only a
+# fallback. It is recommended to install and use dot, since it yields more
+# powerful graphs.
+
+CLASS_DIAGRAMS = YES
+
+# You can define message sequence charts within doxygen comments using the \msc
+# command. Doxygen will then run the mscgen tool (see
+# http://www.mcternan.me.uk/mscgen/) to produce the chart and insert it in the
+# documentation. The MSCGEN_PATH tag allows you to specify the directory where
+# the mscgen tool resides. If left empty the tool is assumed to be found in the
+# default search path.
+
+MSCGEN_PATH =
+
+# If set to YES, the inheritance and collaboration graphs will hide
+# inheritance and usage relations if the target is undocumented
+# or is not a class.
+
+HIDE_UNDOC_RELATIONS = YES
+
+# If you set the HAVE_DOT tag to YES then doxygen will assume the dot tool is
+# available from the path. This tool is part of Graphviz, a graph visualization
+# toolkit from AT&T and Lucent Bell Labs. The other options in this section
+# have no effect if this option is set to NO (the default)
+
+HAVE_DOT = NO
+
+# The DOT_NUM_THREADS specifies the number of dot invocations doxygen is
+# allowed to run in parallel. When set to 0 (the default) doxygen will
+# base this on the number of processors available in the system. You can set it
+# explicitly to a value larger than 0 to get control over the balance
+# between CPU load and processing speed.
+
+DOT_NUM_THREADS = 0
+
+# By default doxygen will write a font called FreeSans.ttf to the output
+# directory and reference it in all dot files that doxygen generates. This
+# font does not include all possible unicode characters however, so when you need
+# these (or just want a differently looking font) you can specify the font name
+# using DOT_FONTNAME. You need need to make sure dot is able to find the font,
+# which can be done by putting it in a standard location or by setting the
+# DOTFONTPATH environment variable or by setting DOT_FONTPATH to the directory
+# containing the font.
+
+DOT_FONTNAME = FreeSans.ttf
+
+# The DOT_FONTSIZE tag can be used to set the size of the font of dot graphs.
+# The default size is 10pt.
+
+DOT_FONTSIZE = 10
+
+# By default doxygen will tell dot to use the output directory to look for the
+# FreeSans.ttf font (which doxygen will put there itself). If you specify a
+# different font using DOT_FONTNAME you can set the path where dot
+# can find it using this tag.
+
+DOT_FONTPATH =
+
+# If the CLASS_GRAPH and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for each documented class showing the direct and
+# indirect inheritance relations. Setting this tag to YES will force the
+# the CLASS_DIAGRAMS tag to NO.
+
+CLASS_GRAPH = YES
+
+# If the COLLABORATION_GRAPH and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for each documented class showing the direct and
+# indirect implementation dependencies (inheritance, containment, and
+# class references variables) of the class with other documented classes.
+
+COLLABORATION_GRAPH = YES
+
+# If the GROUP_GRAPHS and HAVE_DOT tags are set to YES then doxygen
+# will generate a graph for groups, showing the direct groups dependencies
+
+GROUP_GRAPHS = YES
+
+# If the UML_LOOK tag is set to YES doxygen will generate inheritance and
+# collaboration diagrams in a style similar to the OMG's Unified Modeling
+# Language.
+
+UML_LOOK = NO
+
+# If set to YES, the inheritance and collaboration graphs will show the
+# relations between templates and their instances.
+
+TEMPLATE_RELATIONS = NO
+
+# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDE_GRAPH, and HAVE_DOT
+# tags are set to YES then doxygen will generate a graph for each documented
+# file showing the direct and indirect include dependencies of the file with
+# other documented files.
+
+INCLUDE_GRAPH = YES
+
+# If the ENABLE_PREPROCESSING, SEARCH_INCLUDES, INCLUDED_BY_GRAPH, and
+# HAVE_DOT tags are set to YES then doxygen will generate a graph for each
+# documented header file showing the documented files that directly or
+# indirectly include this file.
+
+INCLUDED_BY_GRAPH = YES
+
+# If the CALL_GRAPH and HAVE_DOT options are set to YES then
+# doxygen will generate a call dependency graph for every global function
+# or class method. Note that enabling this option will significantly increase
+# the time of a run. So in most cases it will be better to enable call graphs
+# for selected functions only using the \callgraph command.
+
+CALL_GRAPH = NO
+
+# If the CALLER_GRAPH and HAVE_DOT tags are set to YES then
+# doxygen will generate a caller dependency graph for every global function
+# or class method. Note that enabling this option will significantly increase
+# the time of a run. So in most cases it will be better to enable caller
+# graphs for selected functions only using the \callergraph command.
+
+CALLER_GRAPH = NO
+
+# If the GRAPHICAL_HIERARCHY and HAVE_DOT tags are set to YES then doxygen
+# will graphical hierarchy of all classes instead of a textual one.
+
+GRAPHICAL_HIERARCHY = YES
+
+# If the DIRECTORY_GRAPH, SHOW_DIRECTORIES and HAVE_DOT tags are set to YES
+# then doxygen will show the dependencies a directory has on other directories
+# in a graphical way. The dependency relations are determined by the #include
+# relations between the files in the directories.
+
+DIRECTORY_GRAPH = YES
+
+# The DOT_IMAGE_FORMAT tag can be used to set the image format of the images
+# generated by dot. Possible values are png, jpg, or gif
+# If left blank png will be used.
+
+DOT_IMAGE_FORMAT = png
+
+# The tag DOT_PATH can be used to specify the path where the dot tool can be
+# found. If left blank, it is assumed the dot tool can be found in the path.
+
+DOT_PATH =
+
+# The DOTFILE_DIRS tag can be used to specify one or more directories that
+# contain dot files that are included in the documentation (see the
+# \dotfile command).
+
+DOTFILE_DIRS =
+
+# The DOT_GRAPH_MAX_NODES tag can be used to set the maximum number of
+# nodes that will be shown in the graph. If the number of nodes in a graph
+# becomes larger than this value, doxygen will truncate the graph, which is
+# visualized by representing a node as a red box. Note that doxygen if the
+# number of direct children of the root node in a graph is already larger than
+# DOT_GRAPH_MAX_NODES then the graph will not be shown at all. Also note
+# that the size of a graph can be further restricted by MAX_DOT_GRAPH_DEPTH.
+
+DOT_GRAPH_MAX_NODES = 50
+
+# The MAX_DOT_GRAPH_DEPTH tag can be used to set the maximum depth of the
+# graphs generated by dot. A depth value of 3 means that only nodes reachable
+# from the root by following a path via at most 3 edges will be shown. Nodes
+# that lay further from the root node will be omitted. Note that setting this
+# option to 1 or 2 may greatly reduce the computation time needed for large
+# code bases. Also note that the size of a graph can be further restricted by
+# DOT_GRAPH_MAX_NODES. Using a depth of 0 means no depth restriction.
+
+MAX_DOT_GRAPH_DEPTH = 0
+
+# Set the DOT_TRANSPARENT tag to YES to generate images with a transparent
+# background. This is disabled by default, because dot on Windows does not
+# seem to support this out of the box. Warning: Depending on the platform used,
+# enabling this option may lead to badly anti-aliased labels on the edges of
+# a graph (i.e. they become hard to read).
+
+DOT_TRANSPARENT = NO
+
+# Set the DOT_MULTI_TARGETS tag to YES allow dot to generate multiple output
+# files in one run (i.e. multiple -o and -T options on the command line). This
+# makes dot run faster, but since only newer versions of dot (>1.8.10)
+# support this, this feature is disabled by default.
+
+DOT_MULTI_TARGETS = YES
+
+# If the GENERATE_LEGEND tag is set to YES (the default) Doxygen will
+# generate a legend page explaining the meaning of the various boxes and
+# arrows in the dot generated graphs.
+
+GENERATE_LEGEND = YES
+
+# If the DOT_CLEANUP tag is set to YES (the default) Doxygen will
+# remove the intermediate dot files that are used to generate
+# the various graphs.
+
+DOT_CLEANUP = YES
--- /dev/null
+The OpenLDAP Public License
+ Version 2.8, 17 August 2003
+
+Redistribution and use of this software and associated documentation
+("Software"), with or without modification, are permitted provided
+that the following conditions are met:
+
+1. Redistributions in source form must retain copyright statements
+ and notices,
+
+2. Redistributions in binary form must reproduce applicable copyright
+ statements and notices, this list of conditions, and the following
+ disclaimer in the documentation and/or other materials provided
+ with the distribution, and
+
+3. Redistributions must contain a verbatim copy of this document.
+
+The OpenLDAP Foundation may revise this license from time to time.
+Each revision is distinguished by a version number. You may use
+this Software under terms of this license revision or under the
+terms of any subsequent revision of the license.
+
+THIS SOFTWARE IS PROVIDED BY THE OPENLDAP FOUNDATION AND ITS
+CONTRIBUTORS ``AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES,
+INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
+AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
+SHALL THE OPENLDAP FOUNDATION, ITS CONTRIBUTORS, OR THE AUTHOR(S)
+OR OWNER(S) OF THE SOFTWARE BE LIABLE FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.
+
+The names of the authors and copyright holders must not be used in
+advertising or otherwise to promote the sale, use or other dealing
+in this Software without specific, written prior permission. Title
+to copyright in this Software shall at all times remain with copyright
+holders.
+
+OpenLDAP is a registered trademark of the OpenLDAP Foundation.
+
+Copyright 1999-2003 The OpenLDAP Foundation, Redwood City,
+California, USA. All Rights Reserved. Permission to copy and
+distribute verbatim copies of this document is granted.
--- /dev/null
+CC = gcc
+W = -W -Wall -Wno-unused-parameter -Wbad-function-cast
+OPT = -O2 -g
+CFLAGS = -pthread $(OPT) $(W) $(XCFLAGS)
+LDLIBS =
+
+PROGS = mdb_stat mtest mtest2 mtest3
+all: libmdb.a libmdb.so $(PROGS)
+
+clean:
+ rm -rf $(PROGS) *.[ao] *.so *~ testdb
+
+test: all
+ mkdir testdb
+ ./mtest && ./mdb_stat testdb
+
+libmdb.a: mdb.o midl.o
+ ar rs $@ mdb.o midl.o
+
+libmdb.so: mdb.o midl.o
+ gcc -shared -o $@ mdb.o midl.o
+
+mdb_stat: mdb_stat.o libmdb.a
+mtest: mtest.o libmdb.a
+mtest2: mtest2.o libmdb.a
+mtest3: mtest3.o libmdb.a
+mtest4: mtest4.o libmdb.a
+
+mdb.o: mdb.c mdb.h midl.h
+ $(CC) $(CFLAGS) -fPIC $(CPPFLAGS) -c mdb.c
+
+midl.o: midl.c midl.h
+ $(CC) $(CFLAGS) -fPIC $(CPPFLAGS) -c midl.c
+
+%: %.o
+ $(CC) $(CFLAGS) $(LDFLAGS) $^ $(LDLIBS) -o $@
+
+%.o: %.c mdb.h
+ $(CC) $(CFLAGS) $(CPPFLAGS) -c $<
--- /dev/null
+#!/bin/sh
+warning_filter() {
+ egrep -v ': In function .*:$|: warning: (ISO C99 requires rest arguments to be used|format .*%p.* has type .*struct MDB_.*\*)'
+}
+
+exit `{
+ { XCFLAGS="-std=c99 -pedantic" make "$@" 2>&1; echo $? >&3; } |
+ warning_filter >&2
+} 3>&1`
--- /dev/null
+/** @file mdb.c
+ * @brief memory-mapped database library
+ *
+ * A Btree-based database management library modeled loosely on the
+ * BerkeleyDB API, but much simplified.
+ */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ *
+ * This code is derived from btree.c written by Martin Hedenfalk.
+ *
+ * Copyright (c) 2009, 2010 Martin Hedenfalk <martin@bzero.se>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/param.h>
+#ifdef _WIN32
+#include <windows.h>
+#else
+#include <sys/uio.h>
+#include <sys/mman.h>
+#ifdef HAVE_SYS_FILE_H
+#include <sys/file.h>
+#endif
+#include <fcntl.h>
+#endif
+
+#include <assert.h>
+#include <errno.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <time.h>
+#include <unistd.h>
+#ifdef _WIN32
+#define pthread_t DWORD
+#define pthread_mutex_t HANDLE
+#define pthread_key_t DWORD
+#define pthread_self() GetCurrentThreadId()
+#define pthread_key_create(x,y) *(x) = TlsAlloc()
+#define pthread_key_delete(x) TlsFree(x)
+#define pthread_getspecific(x) TlsGetValue(x)
+#define pthread_setspecific(x,y) TlsSetValue(x,y)
+#define pthread_mutex_unlock(x) ReleaseMutex(x)
+#define pthread_mutex_lock(x) WaitForSingleObject(x, INFINITE)
+#define LOCK_MUTEX_R(env) pthread_mutex_lock(env->me_rmutex)
+#define UNLOCK_MUTEX_R(env) pthread_mutex_unlock(env->me_rmutex)
+#define LOCK_MUTEX_W(env) pthread_mutex_lock(env->me_wmutex)
+#define UNLOCK_MUTEX_W(env) pthread_mutex_unlock(env->me_wmutex)
+#define getpid() GetCurrentProcessId()
+#else
+#include <pthread.h>
+#define LOCK_MUTEX_R(env) pthread_mutex_lock(&env->me_txns->mti_mutex)
+#define UNLOCK_MUTEX_R(env) pthread_mutex_unlock(&env->me_txns->mti_mutex)
+#define LOCK_MUTEX_W(env) pthread_mutex_lock(&env->me_txns->mti_wmutex)
+#define UNLOCK_MUTEX_W(env) pthread_mutex_unlock(&env->me_txns->mti_wmutex)
+#endif
+
+#ifdef _WIN32
+#define fdatasync(fd) !FlushFileBuffers(fd)
+#define ErrCode() GetLastError()
+#define GetPageSize(x) {SYSTEM_INFO si; GetSystemInfo(&si); (x) = si.dwPageSize;}
+#define close(fd) CloseHandle(fd)
+#define munmap(ptr,len) UnmapViewOfFile(ptr)
+#else
+#define ErrCode() errno
+#define HANDLE int
+#define INVALID_HANDLE_VALUE -1
+#define GetPageSize(x) (x) = sysconf(_SC_PAGE_SIZE)
+#endif
+#include "mdb.h"
+
+#define ULONG unsigned long
+typedef ULONG pgno_t;
+
+#include "midl.h"
+
+/* Note: If O_DSYNC is undefined but exists in /usr/include,
+ * preferably set some compiler flag to get the definition.
+ * Otherwise compile with the less efficient -DMDB_DSYNC=O_SYNC.
+ */
+#ifndef MDB_DSYNC
+# define MDB_DSYNC O_DSYNC
+#endif
+
+#ifndef DEBUG
+#define DEBUG 0
+#endif
+
+#if !(__STDC_VERSION__ >= 199901L || defined(__GNUC__))
+# define DPRINTF (void) /* Vararg macros may be unsupported */
+#elif DEBUG
+# define DPRINTF(fmt, ...) /* Requires 2 or more args */ \
+ fprintf(stderr, "%s:%d:(%p) " fmt "\n", __func__, __LINE__, pthread_self(), __VA_ARGS__)
+#else
+# define DPRINTF(fmt, ...) ((void) 0)
+#endif
+#define DPUTS(arg) DPRINTF("%s", arg)
+
+#define PAGESIZE 4096
+#define MDB_MINKEYS 2
+#define MDB_MAGIC 0xBEEFC0DE
+#define MDB_VERSION 1
+#define MAXKEYSIZE 511
+#if DEBUG
+#define KBUF (MAXKEYSIZE*2+1)
+#define DKBUF char kbuf[KBUF]
+#define DKEY(x) mdb_dkey(x, kbuf)
+#else
+#define DKBUF
+#define DKEY(x)
+#endif
+
+/* The DB view is always consistent because all writes are wrapped in
+ * the wmutex. Finer-grained locks aren't necessary.
+ */
+#ifndef LAZY_LOCKS
+#define LAZY_LOCKS 1
+#endif
+#if LAZY_LOCKS
+#define LAZY_MUTEX_LOCK(x)
+#define LAZY_MUTEX_UNLOCK(x)
+#define LAZY_RWLOCK_UNLOCK(x)
+#define LAZY_RWLOCK_WRLOCK(x)
+#define LAZY_RWLOCK_RDLOCK(x)
+#define LAZY_RWLOCK_DEF(x)
+#define LAZY_RWLOCK_INIT(x,y)
+#define LAZY_RWLOCK_DESTROY(x)
+#else
+#define LAZY_MUTEX_LOCK(x) pthread_mutex_lock(x)
+#define LAZY_MUTEX_UNLOCK(x) pthread_mutex_unlock(x)
+#define LAZY_RWLOCK_UNLOCK(x) pthread_rwlock_unlock(x)
+#define LAZY_RWLOCK_WRLOCK(x) pthread_rwlock_wrlock(x)
+#define LAZY_RWLOCK_RDLOCK(x) pthread_rwlock_rdlock(x)
+#define LAZY_RWLOCK_DEF(x) pthread_rwlock_t x
+#define LAZY_RWLOCK_INIT(x,y) pthread_rwlock_init(x,y)
+#define LAZY_RWLOCK_DESTROY(x) pthread_rwlock_destroy(x)
+#endif
+
+#define P_INVALID (~0UL)
+
+#define F_ISSET(w, f) (((w) & (f)) == (f))
+
+typedef uint16_t indx_t;
+
+#define DEFAULT_READERS 126
+#define DEFAULT_MAPSIZE 1048576
+
+/* Lock descriptor stuff */
+#ifndef CACHELINE
+#define CACHELINE 64 /* most CPUs. Itanium uses 128 */
+#endif
+
+typedef struct MDB_rxbody {
+ ULONG mrb_txnid;
+ pid_t mrb_pid;
+ pthread_t mrb_tid;
+} MDB_rxbody;
+
+typedef struct MDB_reader {
+ union {
+ MDB_rxbody mrx;
+#define mr_txnid mru.mrx.mrb_txnid
+#define mr_pid mru.mrx.mrb_pid
+#define mr_tid mru.mrx.mrb_tid
+ /* cache line alignment */
+ char pad[(sizeof(MDB_rxbody)+CACHELINE-1) & ~(CACHELINE-1)];
+ } mru;
+} MDB_reader;
+
+typedef struct MDB_txbody {
+ uint32_t mtb_magic;
+ uint32_t mtb_version;
+/* For POSIX the actual mutexes reside in shared memory.
+ * On Windows, mutexes are allocated by the kernel; we store
+ * the name in shared memory so that other processes can
+ * grab them.
+ */
+#ifdef _WIN32
+ char mtb_rmname[32];
+#else
+ pthread_mutex_t mtb_mutex;
+#endif
+ ULONG mtb_txnid;
+ uint32_t mtb_numreaders;
+ uint32_t mtb_me_toggle;
+} MDB_txbody;
+
+typedef struct MDB_txninfo {
+ union {
+ MDB_txbody mtb;
+#define mti_magic mt1.mtb.mtb_magic
+#define mti_version mt1.mtb.mtb_version
+#define mti_mutex mt1.mtb.mtb_mutex
+#define mti_rmname mt1.mtb.mtb_rmname
+#define mti_txnid mt1.mtb.mtb_txnid
+#define mti_numreaders mt1.mtb.mtb_numreaders
+#define mti_me_toggle mt1.mtb.mtb_me_toggle
+ char pad[(sizeof(MDB_txbody)+CACHELINE-1) & ~(CACHELINE-1)];
+ } mt1;
+ union {
+#ifdef _WIN32
+ char mt2_wmname[32];
+#define mti_wmname mt2.mt2_wmname
+#else
+ pthread_mutex_t mt2_wmutex;
+#define mti_wmutex mt2.mt2_wmutex
+#endif
+ char pad[(sizeof(pthread_mutex_t)+CACHELINE-1) & ~(CACHELINE-1)];
+ } mt2;
+ MDB_reader mti_readers[1];
+} MDB_txninfo;
+
+/* Common header for all page types. Overflow pages
+ * occupy a number of contiguous pages with no
+ * headers on any page after the first.
+ */
+typedef struct MDB_page { /* represents a page of storage */
+#define mp_pgno mp_p.p_pgno
+ union padded {
+ pgno_t p_pgno; /* page number */
+ void * p_align; /* for IL32P64 */
+ } mp_p;
+#define P_BRANCH 0x01 /* branch page */
+#define P_LEAF 0x02 /* leaf page */
+#define P_OVERFLOW 0x04 /* overflow page */
+#define P_META 0x08 /* meta page */
+#define P_DIRTY 0x10 /* dirty page */
+#define P_LEAF2 0x20 /* DB with small, fixed size keys and no data */
+ uint32_t mp_flags;
+#define mp_lower mp_pb.pb.pb_lower
+#define mp_upper mp_pb.pb.pb_upper
+#define mp_pages mp_pb.pb_pages
+ union page_bounds {
+ struct {
+ indx_t pb_lower; /* lower bound of free space */
+ indx_t pb_upper; /* upper bound of free space */
+ } pb;
+ uint32_t pb_pages; /* number of overflow pages */
+ } mp_pb;
+ indx_t mp_ptrs[1]; /* dynamic size */
+} MDB_page;
+
+#define PAGEHDRSZ ((unsigned) offsetof(MDB_page, mp_ptrs))
+
+#define NUMKEYS(p) (((p)->mp_lower - PAGEHDRSZ) >> 1)
+#define SIZELEFT(p) (indx_t)((p)->mp_upper - (p)->mp_lower)
+#define PAGEFILL(env, p) (1000L * ((env)->me_psize - PAGEHDRSZ - SIZELEFT(p)) / \
+ ((env)->me_psize - PAGEHDRSZ))
+#define IS_LEAF(p) F_ISSET((p)->mp_flags, P_LEAF)
+#define IS_LEAF2(p) F_ISSET((p)->mp_flags, P_LEAF2)
+#define IS_BRANCH(p) F_ISSET((p)->mp_flags, P_BRANCH)
+#define IS_OVERFLOW(p) F_ISSET((p)->mp_flags, P_OVERFLOW)
+
+#define OVPAGES(size, psize) ((PAGEHDRSZ-1 + (size)) / (psize) + 1)
+
+typedef struct MDB_db {
+ uint32_t md_pad; /* also ksize for LEAF2 pages */
+ uint16_t md_flags;
+ uint16_t md_depth;
+ ULONG md_branch_pages;
+ ULONG md_leaf_pages;
+ ULONG md_overflow_pages;
+ ULONG md_entries;
+ pgno_t md_root;
+} MDB_db;
+
+#define FREE_DBI 0
+#define MAIN_DBI 1
+
+typedef struct MDB_meta { /* meta (footer) page content */
+ uint32_t mm_magic;
+ uint32_t mm_version;
+ void *mm_address; /* address for fixed mapping */
+ size_t mm_mapsize; /* size of mmap region */
+ MDB_db mm_dbs[2]; /* first is free space, 2nd is main db */
+#define mm_psize mm_dbs[0].md_pad
+#define mm_flags mm_dbs[0].md_flags
+ pgno_t mm_last_pg; /* last used page in file */
+ ULONG mm_txnid; /* txnid that committed this page */
+} MDB_meta;
+
+typedef struct MDB_dhead { /* a dirty page */
+ MDB_page *md_parent;
+ unsigned md_pi; /* parent index */
+ int md_num;
+} MDB_dhead;
+
+typedef struct MDB_dpage {
+ MDB_dhead h;
+ MDB_page p;
+} MDB_dpage;
+
+typedef struct MDB_oldpages {
+ struct MDB_oldpages *mo_next;
+ ULONG mo_txnid;
+ pgno_t mo_pages[1]; /* dynamic */
+} MDB_oldpages;
+
+typedef struct MDB_pageparent {
+ MDB_page *mp_page;
+ MDB_page *mp_parent;
+ unsigned mp_pi;
+} MDB_pageparent;
+
+static MDB_dpage *mdb_alloc_page(MDB_txn *txn, MDB_dbi dbi, MDB_page *parent, unsigned int parent_idx, int num);
+static int mdb_touch(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *mp);
+
+typedef struct MDB_ppage { /* ordered list of pages */
+ MDB_page *mp_page;
+ unsigned int mp_ki; /* cursor index on page */
+} MDB_ppage;
+
+#define CURSOR_TOP(c) (&(c)->mc_stack[(c)->mc_snum-1])
+#define CURSOR_PARENT(c) (&(c)->mc_stack[(c)->mc_snum-2])
+#define CURSOR_STACK 32
+
+struct MDB_xcursor;
+
+struct MDB_cursor {
+ MDB_txn *mc_txn;
+ MDB_ppage mc_stack[CURSOR_STACK]; /* stack of parent pages */
+ unsigned int mc_snum; /* number of pushed pages */
+ MDB_dbi mc_dbi;
+ short mc_initialized; /* 1 if initialized */
+ short mc_eof; /* 1 if end is reached */
+ struct MDB_xcursor *mc_xcursor;
+};
+
+#define METADATA(p) ((void *)((char *)(p) + PAGEHDRSZ))
+
+typedef struct MDB_node {
+#define mn_pgno mn_p.np_pgno
+#define mn_dsize mn_p.np_dsize
+ union {
+ pgno_t np_pgno; /* child page number */
+ uint32_t np_dsize; /* leaf data size */
+ } mn_p;
+ unsigned int mn_flags:4;
+ unsigned int mn_ksize:12; /* key size */
+#define F_BIGDATA 0x01 /* data put on overflow page */
+#define F_SUBDATA 0x02 /* data is a sub-database */
+#define F_DUPDATA 0x04 /* data has duplicates */
+ char mn_data[1];
+} MDB_node;
+
+typedef struct MDB_dbx {
+ MDB_val md_name;
+ MDB_cmp_func *md_cmp; /* user compare function */
+ MDB_cmp_func *md_dcmp; /* user dupsort function */
+ MDB_rel_func *md_rel; /* user relocate function */
+ MDB_dbi md_parent;
+ unsigned int md_dirty;
+} MDB_dbx;
+
+struct MDB_txn {
+ pgno_t mt_next_pgno; /* next unallocated page */
+ ULONG mt_txnid;
+ MDB_env *mt_env;
+ pgno_t *mt_free_pgs; /* this is an IDL */
+ union {
+ MIDL2 *dirty_list; /* modified pages */
+ MDB_reader *reader;
+ } mt_u;
+ MDB_dbx *mt_dbxs; /* array */
+ MDB_db *mt_dbs;
+ unsigned int mt_numdbs;
+
+#define MDB_TXN_RDONLY 0x01 /* read-only transaction */
+#define MDB_TXN_ERROR 0x02 /* an error has occurred */
+ unsigned int mt_flags;
+ unsigned int mt_toggle;
+};
+
+/* Context for sorted-dup records */
+typedef struct MDB_xcursor {
+ MDB_cursor mx_cursor;
+ MDB_txn mx_txn;
+ MDB_dbx mx_dbxs[4];
+ MDB_db mx_dbs[4];
+} MDB_xcursor;
+
+struct MDB_env {
+ HANDLE me_fd;
+ HANDLE me_lfd;
+ HANDLE me_mfd; /* just for writing the meta pages */
+#define MDB_FATAL_ERROR 0x80000000U
+ uint32_t me_flags;
+ uint32_t me_extrapad; /* unused for now */
+ unsigned int me_maxreaders;
+ unsigned int me_numdbs;
+ unsigned int me_maxdbs;
+ char *me_path;
+ char *me_map;
+ MDB_txninfo *me_txns;
+ MDB_meta *me_metas[2];
+ MDB_txn *me_txn; /* current write transaction */
+ size_t me_mapsize;
+ off_t me_size; /* current file size */
+ pgno_t me_maxpg; /* me_mapsize / me_psize */
+ unsigned int me_psize;
+ unsigned int me_db_toggle;
+ MDB_dbx *me_dbxs; /* array */
+ MDB_db *me_dbs[2];
+ MDB_oldpages *me_pghead;
+ pthread_key_t me_txkey; /* thread-key for readers */
+ MDB_dpage *me_dpages;
+ pgno_t me_free_pgs[MDB_IDL_UM_SIZE];
+ MIDL2 me_dirty_list[MDB_IDL_DB_SIZE];
+ LAZY_RWLOCK_DEF(me_dblock);
+#ifdef _WIN32
+ HANDLE me_rmutex; /* Windows mutexes don't reside in shared mem */
+ HANDLE me_wmutex;
+#endif
+};
+
+#define NODESIZE offsetof(MDB_node, mn_data)
+
+#define INDXSIZE(k) (NODESIZE + ((k) == NULL ? 0 : (k)->mv_size))
+#define LEAFSIZE(k, d) (NODESIZE + (k)->mv_size + (d)->mv_size)
+#define NODEPTR(p, i) ((MDB_node *)((char *)(p) + (p)->mp_ptrs[i]))
+#define NODEKEY(node) (void *)((node)->mn_data)
+#define NODEDATA(node) (void *)((char *)(node)->mn_data + (node)->mn_ksize)
+#define NODEPGNO(node) ((node)->mn_pgno)
+#define NODEDSZ(node) ((node)->mn_dsize)
+#define NODEKSZ(node) ((node)->mn_ksize)
+#define LEAF2KEY(p, i, ks) ((char *)(p) + PAGEHDRSZ + ((i)*(ks)))
+
+#define MDB_SET_KEY(node, key) if (key!=NULL) {(key)->mv_size = NODEKSZ(node); (key)->mv_data = NODEKEY(node);}
+
+#define MDB_COMMIT_PAGES 64 /* max number of pages to write in one commit */
+
+static int mdb_search_page_root(MDB_txn *txn,
+ MDB_dbi dbi, MDB_val *key,
+ MDB_cursor *cursor, int modify,
+ MDB_pageparent *mpp);
+static int mdb_search_page(MDB_txn *txn,
+ MDB_dbi dbi, MDB_val *key,
+ MDB_cursor *cursor, int modify,
+ MDB_pageparent *mpp);
+
+static int mdb_env_read_header(MDB_env *env, MDB_meta *meta);
+static int mdb_env_read_meta(MDB_env *env, int *which);
+static int mdb_env_write_meta(MDB_txn *txn);
+static int mdb_get_page(MDB_txn *txn, pgno_t pgno, MDB_page **mp);
+
+static MDB_node *mdb_search_node(MDB_txn *txn, MDB_dbi dbi, MDB_page *mp,
+ MDB_val *key, int *exactp, unsigned int *kip);
+static int mdb_add_node(MDB_txn *txn, MDB_dbi dbi, MDB_page *mp,
+ indx_t indx, MDB_val *key, MDB_val *data,
+ pgno_t pgno, uint8_t flags);
+static void mdb_del_node(MDB_page *mp, indx_t indx, int ksize);
+static int mdb_del0(MDB_txn *txn, MDB_dbi dbi, unsigned int ki,
+ MDB_pageparent *mpp, MDB_node *leaf);
+static int mdb_put0(MDB_txn *txn, MDB_dbi dbi,
+ MDB_val *key, MDB_val *data, unsigned int flags);
+static int mdb_read_data(MDB_txn *txn, MDB_node *leaf, MDB_val *data);
+
+static int mdb_rebalance(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *mp);
+static int mdb_update_key(MDB_page *mp, indx_t indx, MDB_val *key);
+static int mdb_move_node(MDB_txn *txn, MDB_dbi dbi,
+ MDB_pageparent *src, indx_t srcindx,
+ MDB_pageparent *dst, indx_t dstindx);
+static int mdb_merge(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *src,
+ MDB_pageparent *dst);
+static int mdb_split(MDB_txn *txn, MDB_dbi dbi, MDB_page **mpp,
+ unsigned int *newindxp, MDB_val *newkey,
+ MDB_val *newdata, pgno_t newpgno);
+static MDB_dpage *mdb_new_page(MDB_txn *txn, MDB_dbi dbi, uint32_t flags, int num);
+
+static void cursor_pop_page(MDB_cursor *cursor);
+static MDB_ppage *cursor_push_page(MDB_cursor *cursor,
+ MDB_page *mp);
+
+static int mdb_sibling(MDB_cursor *cursor, int move_right);
+static int mdb_cursor_next(MDB_cursor *cursor,
+ MDB_val *key, MDB_val *data, MDB_cursor_op op);
+static int mdb_cursor_prev(MDB_cursor *cursor,
+ MDB_val *key, MDB_val *data, MDB_cursor_op op);
+static int mdb_cursor_set(MDB_cursor *cursor,
+ MDB_val *key, MDB_val *data, MDB_cursor_op op, int *exactp);
+static int mdb_cursor_first(MDB_cursor *cursor,
+ MDB_val *key, MDB_val *data);
+static int mdb_cursor_last(MDB_cursor *cursor,
+ MDB_val *key, MDB_val *data);
+
+static void mdb_xcursor_init0(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx);
+static void mdb_xcursor_init1(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx,
+ MDB_page *mp, MDB_node *node);
+static void mdb_xcursor_fini(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx);
+
+static size_t mdb_leaf_size(MDB_env *env, MDB_val *key,
+ MDB_val *data);
+static size_t mdb_branch_size(MDB_env *env, MDB_val *key);
+
+static int memncmp(const void *s1, size_t n1,
+ const void *s2, size_t n2);
+static int memnrcmp(const void *s1, size_t n1,
+ const void *s2, size_t n2);
+
+#ifdef _WIN32
+static SECURITY_DESCRIPTOR mdb_null_sd;
+static SECURITY_ATTRIBUTES mdb_all_sa;
+static int mdb_sec_inited;
+#endif
+
+static int
+memncmp(const void *s1, size_t n1, const void *s2, size_t n2)
+{
+ int diff, len_diff = -1;
+
+ if (n1 >= n2) {
+ len_diff = (n1 > n2);
+ n1 = n2;
+ }
+ diff = memcmp(s1, s2, n1);
+ return diff ? diff : len_diff;
+}
+
+static int
+memnrcmp(const void *s1, size_t n1, const void *s2, size_t n2)
+{
+ const unsigned char *p1, *p2, *p1_lim;
+
+ if (n2 == 0)
+ return n1 != 0;
+ if (n1 == 0)
+ return -1;
+
+ p1 = (const unsigned char *)s1 + n1 - 1;
+ p2 = (const unsigned char *)s2 + n2 - 1;
+
+ for (p1_lim = (n1 <= n2 ? s1 : s2); *p1 == *p2; p1--, p2--) {
+ if (p1 == p1_lim)
+ return (p1 != s1) ? (p1 != p2) : (p2 != s2) ? -1 : 0;
+ }
+ return *p1 - *p2;
+}
+
+char *
+mdb_version(int *maj, int *min, int *pat)
+{
+ if (maj) *maj = MDB_VERSION_MAJOR;
+ if (min) *min = MDB_VERSION_MINOR;
+ if (pat) *pat = MDB_VERSION_PATCH;
+ return MDB_VERSION_STRING;
+}
+
+static char *const mdb_errstr[] = {
+ "MDB_KEYEXIST: Key/data pair already exists",
+ "MDB_NOTFOUND: No matching key/data pair found",
+ "MDB_PAGE_NOTFOUND: Requested page not found",
+ "MDB_CORRUPTED: Located page was wrong type",
+ "MDB_PANIC: Update of meta page failed",
+ "MDB_VERSION_MISMATCH: Database environment version mismatch"
+};
+
+char *
+mdb_strerror(int err)
+{
+ if (!err)
+ return ("Successful return: 0");
+
+ if (err >= MDB_KEYEXIST && err <= MDB_VERSION_MISMATCH)
+ return mdb_errstr[err - MDB_KEYEXIST];
+
+ return strerror(err);
+}
+
+#if DEBUG
+static char *
+mdb_dkey(MDB_val *key, char *buf)
+{
+ char *ptr = buf;
+ unsigned char *c = key->mv_data;
+ unsigned int i;
+ if (key->mv_size > MAXKEYSIZE)
+ return "MAXKEYSIZE";
+ for (i=0; i<key->mv_size; i++)
+ ptr += sprintf(ptr, "%02x", *c++);
+ return buf;
+}
+#endif
+
+int
+mdb_cmp(MDB_txn *txn, MDB_dbi dbi, const MDB_val *a, const MDB_val *b)
+{
+ if (txn->mt_dbxs[dbi].md_cmp)
+ return txn->mt_dbxs[dbi].md_cmp(a, b);
+
+ if (txn->mt_dbs[dbi].md_flags & (MDB_REVERSEKEY
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+ |MDB_INTEGERKEY
+#endif
+ ))
+ return memnrcmp(a->mv_data, a->mv_size, b->mv_data, b->mv_size);
+ else
+ return memncmp((char *)a->mv_data, a->mv_size, b->mv_data, b->mv_size);
+}
+
+int
+mdb_dcmp(MDB_txn *txn, MDB_dbi dbi, const MDB_val *a, const MDB_val *b)
+{
+ if (txn->mt_dbxs[dbi].md_dcmp)
+ return txn->mt_dbxs[dbi].md_dcmp(a, b);
+
+ if (txn->mt_dbs[dbi].md_flags & (0
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+ |MDB_INTEGERDUP
+#endif
+ ))
+ return memnrcmp(a->mv_data, a->mv_size, b->mv_data, b->mv_size);
+ else
+ return memncmp((char *)a->mv_data, a->mv_size, b->mv_data, b->mv_size);
+}
+
+/* Allocate new page(s) for writing */
+static MDB_dpage *
+mdb_alloc_page(MDB_txn *txn, MDB_dbi dbi, MDB_page *parent, unsigned int parent_idx, int num)
+{
+ MDB_dpage *dp;
+ pgno_t pgno = P_INVALID;
+ MIDL2 mid;
+
+ if (txn->mt_txnid > 2) {
+
+ if (!txn->mt_env->me_pghead && dbi != FREE_DBI &&
+ txn->mt_dbs[FREE_DBI].md_root != P_INVALID) {
+ /* See if there's anything in the free DB */
+ MDB_pageparent mpp;
+ MDB_node *leaf;
+ ULONG *kptr, oldest;
+
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ mdb_search_page(txn, FREE_DBI, NULL, NULL, 0, &mpp);
+ leaf = NODEPTR(mpp.mp_page, 0);
+ kptr = (ULONG *)NODEKEY(leaf);
+
+ {
+ unsigned int i;
+ oldest = txn->mt_txnid - 1;
+ for (i=0; i<txn->mt_env->me_txns->mti_numreaders; i++) {
+ ULONG mr = txn->mt_env->me_txns->mti_readers[i].mr_txnid;
+ if (mr && mr < oldest)
+ oldest = mr;
+ }
+ }
+
+ if (oldest > *kptr) {
+ /* It's usable, grab it.
+ */
+ MDB_oldpages *mop;
+ MDB_val data;
+ pgno_t *idl;
+
+ mdb_read_data(txn, leaf, &data);
+ idl = (ULONG *)data.mv_data;
+ mop = malloc(sizeof(MDB_oldpages) + MDB_IDL_SIZEOF(idl) - sizeof(pgno_t));
+ mop->mo_next = txn->mt_env->me_pghead;
+ mop->mo_txnid = *kptr;
+ txn->mt_env->me_pghead = mop;
+ memcpy(mop->mo_pages, idl, MDB_IDL_SIZEOF(idl));
+
+#if DEBUG > 1
+ {
+ unsigned int i;
+ DPRINTF("IDL read txn %lu root %lu num %lu",
+ mop->mo_txnid, txn->mt_dbs[FREE_DBI].md_root, idl[0]);
+ for (i=0; i<idl[0]; i++) {
+ DPRINTF("IDL %lu", idl[i+1]);
+ }
+ }
+#endif
+ /* drop this IDL from the DB */
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ mdb_search_page(txn, FREE_DBI, NULL, NULL, 1, &mpp);
+ leaf = NODEPTR(mpp.mp_page, 0);
+ mdb_del0(txn, FREE_DBI, 0, &mpp, leaf);
+ }
+ }
+ if (txn->mt_env->me_pghead) {
+ MDB_oldpages *mop = txn->mt_env->me_pghead;
+ if (num > 1) {
+ /* FIXME: For now, always use fresh pages. We
+ * really ought to search the free list for a
+ * contiguous range.
+ */
+ ;
+ } else {
+ /* peel pages off tail, so we only have to truncate the list */
+ pgno = MDB_IDL_LAST(mop->mo_pages);
+ if (MDB_IDL_IS_RANGE(mop->mo_pages)) {
+ mop->mo_pages[2]++;
+ if (mop->mo_pages[2] > mop->mo_pages[1])
+ mop->mo_pages[0] = 0;
+ } else {
+ mop->mo_pages[0]--;
+ }
+ if (MDB_IDL_IS_ZERO(mop->mo_pages)) {
+ txn->mt_env->me_pghead = mop->mo_next;
+ free(mop);
+ }
+ }
+ }
+ }
+
+ if (pgno == P_INVALID) {
+ /* DB size is maxed out */
+ if (txn->mt_next_pgno + num >= txn->mt_env->me_maxpg)
+ return NULL;
+ }
+ if (txn->mt_env->me_dpages && num == 1) {
+ dp = txn->mt_env->me_dpages;
+ txn->mt_env->me_dpages = (MDB_dpage *)dp->h.md_parent;
+ } else {
+ if ((dp = malloc(txn->mt_env->me_psize * num + sizeof(MDB_dhead))) == NULL)
+ return NULL;
+ }
+ dp->h.md_num = num;
+ dp->h.md_parent = parent;
+ dp->h.md_pi = parent_idx;
+ if (pgno == P_INVALID) {
+ dp->p.mp_pgno = txn->mt_next_pgno;
+ txn->mt_next_pgno += num;
+ } else {
+ dp->p.mp_pgno = pgno;
+ }
+ mid.mid = dp->p.mp_pgno;
+ mid.mptr = dp;
+ mdb_midl2_insert(txn->mt_u.dirty_list, &mid);
+
+ return dp;
+}
+
+/* Touch a page: make it dirty and re-insert into tree with updated pgno.
+ */
+static int
+mdb_touch(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *pp)
+{
+ MDB_page *mp = pp->mp_page;
+ pgno_t pgno;
+ assert(txn != NULL);
+ assert(pp != NULL);
+
+ if (!F_ISSET(mp->mp_flags, P_DIRTY)) {
+ MDB_dpage *dp;
+ if ((dp = mdb_alloc_page(txn, dbi, pp->mp_parent, pp->mp_pi, 1)) == NULL)
+ return ENOMEM;
+ DPRINTF("touched db %u page %lu -> %lu", dbi, mp->mp_pgno, dp->p.mp_pgno);
+ assert(mp->mp_pgno != dp->p.mp_pgno);
+ mdb_midl_insert(txn->mt_free_pgs, mp->mp_pgno);
+ pgno = dp->p.mp_pgno;
+ memcpy(&dp->p, mp, txn->mt_env->me_psize);
+ mp = &dp->p;
+ mp->mp_pgno = pgno;
+ mp->mp_flags |= P_DIRTY;
+
+ /* Update the page number to new touched page. */
+ if (pp->mp_parent != NULL)
+ NODEPGNO(NODEPTR(pp->mp_parent, pp->mp_pi)) = mp->mp_pgno;
+ pp->mp_page = mp;
+ }
+ return 0;
+}
+
+int
+mdb_env_sync(MDB_env *env, int force)
+{
+ int rc = 0;
+ if (force || !F_ISSET(env->me_flags, MDB_NOSYNC)) {
+ if (fdatasync(env->me_fd))
+ rc = ErrCode();
+ }
+ return rc;
+}
+
+static inline void
+mdb_txn_reset0(MDB_txn *txn);
+
+static inline int
+mdb_txn_renew0(MDB_txn *txn)
+{
+ MDB_env *env = txn->mt_env;
+
+ if (txn->mt_flags & MDB_TXN_RDONLY) {
+ MDB_reader *r = pthread_getspecific(env->me_txkey);
+ if (!r) {
+ unsigned int i;
+ pid_t pid = getpid();
+ pthread_t tid = pthread_self();
+
+ LOCK_MUTEX_R(env);
+ for (i=0; i<env->me_txns->mti_numreaders; i++)
+ if (env->me_txns->mti_readers[i].mr_pid == 0)
+ break;
+ if (i == env->me_maxreaders) {
+ UNLOCK_MUTEX_R(env);
+ return ENOMEM;
+ }
+ env->me_txns->mti_readers[i].mr_pid = pid;
+ env->me_txns->mti_readers[i].mr_tid = tid;
+ if (i >= env->me_txns->mti_numreaders)
+ env->me_txns->mti_numreaders = i+1;
+ UNLOCK_MUTEX_R(env);
+ r = &env->me_txns->mti_readers[i];
+ pthread_setspecific(env->me_txkey, r);
+ }
+ txn->mt_txnid = env->me_txns->mti_txnid;
+ txn->mt_toggle = env->me_txns->mti_me_toggle;
+ r->mr_txnid = txn->mt_txnid;
+ txn->mt_u.reader = r;
+ } else {
+ LOCK_MUTEX_W(env);
+
+ txn->mt_txnid = env->me_txns->mti_txnid+1;
+ txn->mt_toggle = env->me_txns->mti_me_toggle;
+ txn->mt_u.dirty_list = env->me_dirty_list;
+ txn->mt_u.dirty_list[0].mid = 0;
+ txn->mt_free_pgs = env->me_free_pgs;
+ txn->mt_free_pgs[0] = 0;
+ txn->mt_next_pgno = env->me_metas[txn->mt_toggle]->mm_last_pg+1;
+ env->me_txn = txn;
+ }
+
+ /* Copy the DB arrays */
+ LAZY_RWLOCK_RDLOCK(&env->me_dblock);
+ txn->mt_numdbs = env->me_numdbs;
+ txn->mt_dbxs = env->me_dbxs; /* mostly static anyway */
+ memcpy(txn->mt_dbs, env->me_metas[txn->mt_toggle]->mm_dbs, 2 * sizeof(MDB_db));
+ if (txn->mt_numdbs > 2)
+ memcpy(txn->mt_dbs+2, env->me_dbs[env->me_db_toggle]+2,
+ (txn->mt_numdbs - 2) * sizeof(MDB_db));
+ LAZY_RWLOCK_UNLOCK(&env->me_dblock);
+
+ return MDB_SUCCESS;
+}
+
+int
+mdb_txn_renew(MDB_txn *txn)
+{
+ int rc;
+
+ if (!txn)
+ return EINVAL;
+
+ if (txn->mt_env->me_flags & MDB_FATAL_ERROR) {
+ DPUTS("environment had fatal error, must shutdown!");
+ return MDB_PANIC;
+ }
+
+ rc = mdb_txn_renew0(txn);
+ if (rc == MDB_SUCCESS) {
+ DPRINTF("renew txn %lu%c %p on mdbenv %p, root page %lu",
+ txn->mt_txnid, (txn->mt_flags & MDB_TXN_RDONLY) ? 'r' : 'w', txn,
+ (void *)txn->mt_env, txn->mt_dbs[MAIN_DBI].md_root);
+ }
+ return rc;
+}
+
+int
+mdb_txn_begin(MDB_env *env, unsigned int flags, MDB_txn **ret)
+{
+ MDB_txn *txn;
+ int rc;
+
+ if (env->me_flags & MDB_FATAL_ERROR) {
+ DPUTS("environment had fatal error, must shutdown!");
+ return MDB_PANIC;
+ }
+ if ((txn = calloc(1, sizeof(MDB_txn) + env->me_maxdbs * sizeof(MDB_db))) == NULL) {
+ DPRINTF("calloc: %s", strerror(ErrCode()));
+ return ENOMEM;
+ }
+ txn->mt_dbs = (MDB_db *)(txn+1);
+ if (flags & MDB_RDONLY) {
+ txn->mt_flags |= MDB_TXN_RDONLY;
+ }
+ txn->mt_env = env;
+
+ rc = mdb_txn_renew0(txn);
+ if (rc)
+ free(txn);
+ else {
+ *ret = txn;
+ DPRINTF("begin txn %lu%c %p on mdbenv %p, root page %lu",
+ txn->mt_txnid, (txn->mt_flags & MDB_TXN_RDONLY) ? 'r' : 'w', txn,
+ (void *) env, txn->mt_dbs[MAIN_DBI].md_root);
+ }
+
+ return rc;
+}
+
+static inline void
+mdb_txn_reset0(MDB_txn *txn)
+{
+ MDB_env *env = txn->mt_env;
+
+ if (F_ISSET(txn->mt_flags, MDB_TXN_RDONLY)) {
+ txn->mt_u.reader->mr_txnid = 0;
+ } else {
+ MDB_oldpages *mop;
+ MDB_dpage *dp;
+ unsigned int i;
+
+ /* return all dirty pages to dpage list */
+ for (i=1; i<=txn->mt_u.dirty_list[0].mid; i++) {
+ dp = txn->mt_u.dirty_list[i].mptr;
+ if (dp->h.md_num == 1) {
+ dp->h.md_parent = (MDB_page *)txn->mt_env->me_dpages;
+ txn->mt_env->me_dpages = dp;
+ } else {
+ /* large pages just get freed directly */
+ free(dp);
+ }
+ }
+
+ while ((mop = txn->mt_env->me_pghead)) {
+ txn->mt_env->me_pghead = mop->mo_next;
+ free(mop);
+ }
+
+ env->me_txn = NULL;
+ for (i=2; i<env->me_numdbs; i++)
+ env->me_dbxs[i].md_dirty = 0;
+ UNLOCK_MUTEX_W(env);
+ }
+}
+
+void
+mdb_txn_reset(MDB_txn *txn)
+{
+ if (txn == NULL)
+ return;
+
+ DPRINTF("reset txn %lu%c %p on mdbenv %p, root page %lu",
+ txn->mt_txnid, (txn->mt_flags & MDB_TXN_RDONLY) ? 'r' : 'w', txn,
+ (void *)txn->mt_env, txn->mt_dbs[MAIN_DBI].md_root);
+
+ mdb_txn_reset0(txn);
+}
+
+void
+mdb_txn_abort(MDB_txn *txn)
+{
+ if (txn == NULL)
+ return;
+
+ DPRINTF("abort txn %lu%c %p on mdbenv %p, root page %lu",
+ txn->mt_txnid, (txn->mt_flags & MDB_TXN_RDONLY) ? 'r' : 'w', txn,
+ (void *)txn->mt_env, txn->mt_dbs[MAIN_DBI].md_root);
+
+ mdb_txn_reset0(txn);
+ free(txn);
+}
+
+int
+mdb_txn_commit(MDB_txn *txn)
+{
+ int n, done;
+ unsigned int i;
+ ssize_t rc;
+ off_t size;
+ MDB_dpage *dp;
+ MDB_env *env;
+ pgno_t next;
+
+ assert(txn != NULL);
+ assert(txn->mt_env != NULL);
+
+ env = txn->mt_env;
+
+ if (F_ISSET(txn->mt_flags, MDB_TXN_RDONLY)) {
+ mdb_txn_abort(txn);
+ return MDB_SUCCESS;
+ }
+
+ if (txn != env->me_txn) {
+ DPUTS("attempt to commit unknown transaction");
+ mdb_txn_abort(txn);
+ return EINVAL;
+ }
+
+ if (F_ISSET(txn->mt_flags, MDB_TXN_ERROR)) {
+ DPUTS("error flag is set, can't commit");
+ mdb_txn_abort(txn);
+ return EINVAL;
+ }
+
+ if (!txn->mt_u.dirty_list[0].mid)
+ goto done;
+
+ DPRINTF("committing txn %lu %p on mdbenv %p, root page %lu",
+ txn->mt_txnid, txn, (void *)env, txn->mt_dbs[MAIN_DBI].md_root);
+
+ /* should only be one record now */
+ if (env->me_pghead) {
+ MDB_pageparent mpp;
+
+ /* make sure first page of freeDB is touched and on freelist */
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ mdb_search_page(txn, FREE_DBI, NULL, NULL, 1, &mpp);
+ }
+ /* save to free list */
+ if (!MDB_IDL_IS_ZERO(txn->mt_free_pgs)) {
+ MDB_val key, data;
+ MDB_pageparent mpp;
+ ULONG i;
+
+ /* make sure last page of freeDB is touched and on freelist */
+ key.mv_size = MAXKEYSIZE+1;
+ key.mv_data = NULL;
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ mdb_search_page(txn, FREE_DBI, &key, NULL, 1, &mpp);
+
+#if DEBUG > 1
+ {
+ unsigned int i;
+ ULONG *idl = txn->mt_free_pgs;
+ DPRINTF("IDL write txn %lu root %lu num %lu",
+ txn->mt_txnid, txn->mt_dbs[FREE_DBI].md_root, idl[0]);
+ for (i=0; i<idl[0]; i++) {
+ DPRINTF("IDL %lu", idl[i+1]);
+ }
+ }
+#endif
+ /* write to last page of freeDB */
+ key.mv_size = sizeof(pgno_t);
+ key.mv_data = (char *)&txn->mt_txnid;
+ data.mv_data = txn->mt_free_pgs;
+ /* The free list can still grow during this call,
+ * despite the pre-emptive touches above. So check
+ * and make sure the entire thing got written.
+ */
+ do {
+ i = txn->mt_free_pgs[0];
+ data.mv_size = MDB_IDL_SIZEOF(txn->mt_free_pgs);
+ rc = mdb_put0(txn, FREE_DBI, &key, &data, 0);
+ if (rc) {
+ mdb_txn_abort(txn);
+ return rc;
+ }
+ } while (i != txn->mt_free_pgs[0]);
+ }
+ /* should only be one record now */
+ if (env->me_pghead) {
+ MDB_val key, data;
+ MDB_oldpages *mop;
+
+ mop = env->me_pghead;
+ key.mv_size = sizeof(pgno_t);
+ key.mv_data = (char *)&mop->mo_txnid;
+ data.mv_size = MDB_IDL_SIZEOF(mop->mo_pages);
+ data.mv_data = mop->mo_pages;
+ mdb_put0(txn, FREE_DBI, &key, &data, 0);
+ free(env->me_pghead);
+ env->me_pghead = NULL;
+ }
+
+ /* Update DB root pointers. Their pages have already been
+ * touched so this is all in-place and cannot fail.
+ */
+ {
+ MDB_val data;
+ data.mv_size = sizeof(MDB_db);
+
+ for (i = 2; i < txn->mt_numdbs; i++) {
+ if (txn->mt_dbxs[i].md_dirty) {
+ data.mv_data = &txn->mt_dbs[i];
+ mdb_put0(txn, MAIN_DBI, &txn->mt_dbxs[i].md_name, &data, 0);
+ }
+ }
+ }
+
+ /* Commit up to MDB_COMMIT_PAGES dirty pages to disk until done.
+ */
+ next = 0;
+ i = 1;
+ do {
+#ifdef _WIN32
+ /* Windows actually supports scatter/gather I/O, but only on
+ * unbuffered file handles. Since we're relying on the OS page
+ * cache for all our data, that's self-defeating. So we just
+ * write pages one at a time. We use the ov structure to set
+ * the write offset, to at least save the overhead of a Seek
+ * system call.
+ */
+ OVERLAPPED ov;
+ memset(&ov, 0, sizeof(ov));
+ for (; i<=txn->mt_u.dirty_list[0].mid; i++) {
+ dp = txn->mt_u.dirty_list[i].mptr;
+ DPRINTF("committing page %lu", dp->p.mp_pgno);
+ size = dp->p.mp_pgno * env->me_psize;
+ ov.Offset = size & 0xffffffff;
+ ov.OffsetHigh = size >> 16;
+ ov.OffsetHigh >>= 16;
+ /* clear dirty flag */
+ dp->p.mp_flags &= ~P_DIRTY;
+ rc = WriteFile(env->me_fd, &dp->p, env->me_psize * dp->h.md_num, NULL, &ov);
+ if (!rc) {
+ n = ErrCode();
+ DPRINTF("WriteFile: %d", n);
+ mdb_txn_abort(txn);
+ return n;
+ }
+ }
+ done = 1;;
+#else
+ struct iovec iov[MDB_COMMIT_PAGES];
+ n = 0;
+ done = 1;
+ size = 0;
+ for (; i<=txn->mt_u.dirty_list[0].mid; i++) {
+ dp = txn->mt_u.dirty_list[i].mptr;
+ if (dp->p.mp_pgno != next) {
+ if (n) {
+ DPRINTF("committing %u dirty pages", n);
+ rc = writev(env->me_fd, iov, n);
+ if (rc != size) {
+ n = ErrCode();
+ if (rc > 0)
+ DPUTS("short write, filesystem full?");
+ else
+ DPRINTF("writev: %s", strerror(n));
+ mdb_txn_abort(txn);
+ return n;
+ }
+ n = 0;
+ size = 0;
+ }
+ lseek(env->me_fd, dp->p.mp_pgno * env->me_psize, SEEK_SET);
+ next = dp->p.mp_pgno;
+ }
+ DPRINTF("committing page %lu", dp->p.mp_pgno);
+ iov[n].iov_len = env->me_psize * dp->h.md_num;
+ iov[n].iov_base = &dp->p;
+ size += iov[n].iov_len;
+ next = dp->p.mp_pgno + dp->h.md_num;
+ /* clear dirty flag */
+ dp->p.mp_flags &= ~P_DIRTY;
+ if (++n >= MDB_COMMIT_PAGES) {
+ done = 0;
+ i++;
+ break;
+ }
+ }
+
+ if (n == 0)
+ break;
+
+ DPRINTF("committing %u dirty pages", n);
+ rc = writev(env->me_fd, iov, n);
+ if (rc != size) {
+ n = ErrCode();
+ if (rc > 0)
+ DPUTS("short write, filesystem full?");
+ else
+ DPRINTF("writev: %s", strerror(n));
+ mdb_txn_abort(txn);
+ return n;
+ }
+#endif
+ } while (!done);
+
+ /* Drop the dirty pages.
+ */
+ for (i=1; i<=txn->mt_u.dirty_list[0].mid; i++) {
+ dp = txn->mt_u.dirty_list[i].mptr;
+ if (dp->h.md_num == 1) {
+ dp->h.md_parent = (MDB_page *)txn->mt_env->me_dpages;
+ txn->mt_env->me_dpages = dp;
+ } else {
+ free(dp);
+ }
+ txn->mt_u.dirty_list[i].mid = 0;
+ }
+ txn->mt_u.dirty_list[0].mid = 0;
+
+ if ((n = mdb_env_sync(env, 0)) != 0 ||
+ (n = mdb_env_write_meta(txn)) != MDB_SUCCESS) {
+ mdb_txn_abort(txn);
+ return n;
+ }
+
+done:
+ env->me_txn = NULL;
+ /* update the DB tables */
+ {
+ int toggle = !env->me_db_toggle;
+ MDB_db *ip, *jp;
+
+ ip = &env->me_dbs[toggle][2];
+ jp = &txn->mt_dbs[2];
+ LAZY_RWLOCK_WRLOCK(&env->me_dblock);
+ for (i = 2; i < txn->mt_numdbs; i++) {
+ if (ip->md_root != jp->md_root)
+ *ip = *jp;
+ ip++; jp++;
+ }
+
+ for (i = 2; i < txn->mt_numdbs; i++) {
+ if (txn->mt_dbxs[i].md_dirty)
+ txn->mt_dbxs[i].md_dirty = 0;
+ }
+ env->me_db_toggle = toggle;
+ env->me_numdbs = txn->mt_numdbs;
+ LAZY_RWLOCK_UNLOCK(&env->me_dblock);
+ }
+
+ UNLOCK_MUTEX_W(env);
+ free(txn);
+
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_env_read_header(MDB_env *env, MDB_meta *meta)
+{
+ char page[PAGESIZE];
+ MDB_page *p;
+ MDB_meta *m;
+ int rc, err;
+
+ /* We don't know the page size yet, so use a minimum value.
+ */
+
+#ifdef _WIN32
+ if (!ReadFile(env->me_fd, page, PAGESIZE, (DWORD *)&rc, NULL) || rc == 0)
+#else
+ if ((rc = read(env->me_fd, page, PAGESIZE)) == 0)
+#endif
+ {
+ return ENOENT;
+ }
+ else if (rc != PAGESIZE) {
+ err = ErrCode();
+ if (rc > 0)
+ err = EINVAL;
+ DPRINTF("read: %s", strerror(err));
+ return err;
+ }
+
+ p = (MDB_page *)page;
+
+ if (!F_ISSET(p->mp_flags, P_META)) {
+ DPRINTF("page %lu not a meta page", p->mp_pgno);
+ return EINVAL;
+ }
+
+ m = METADATA(p);
+ if (m->mm_magic != MDB_MAGIC) {
+ DPUTS("meta has invalid magic");
+ return EINVAL;
+ }
+
+ if (m->mm_version != MDB_VERSION) {
+ DPRINTF("database is version %u, expected version %u",
+ m->mm_version, MDB_VERSION);
+ return MDB_VERSION_MISMATCH;
+ }
+
+ memcpy(meta, m, sizeof(*m));
+ return 0;
+}
+
+static int
+mdb_env_init_meta(MDB_env *env, MDB_meta *meta)
+{
+ MDB_page *p, *q;
+ MDB_meta *m;
+ int rc;
+ unsigned int psize;
+
+ DPUTS("writing new meta page");
+
+ GetPageSize(psize);
+
+ meta->mm_magic = MDB_MAGIC;
+ meta->mm_version = MDB_VERSION;
+ meta->mm_psize = psize;
+ meta->mm_last_pg = 1;
+ meta->mm_flags = env->me_flags & 0xffff;
+ meta->mm_flags |= MDB_INTEGERKEY;
+ meta->mm_dbs[0].md_root = P_INVALID;
+ meta->mm_dbs[1].md_root = P_INVALID;
+
+ p = calloc(2, psize);
+ p->mp_pgno = 0;
+ p->mp_flags = P_META;
+
+ m = METADATA(p);
+ memcpy(m, meta, sizeof(*meta));
+
+ q = (MDB_page *)((char *)p + psize);
+
+ q->mp_pgno = 1;
+ q->mp_flags = P_META;
+
+ m = METADATA(q);
+ memcpy(m, meta, sizeof(*meta));
+
+#ifdef _WIN32
+ {
+ DWORD len;
+ rc = WriteFile(env->me_fd, p, psize * 2, &len, NULL);
+ rc = (len == psize * 2) ? MDB_SUCCESS : ErrCode();
+ }
+#else
+ rc = write(env->me_fd, p, psize * 2);
+ rc = (rc == (int)psize * 2) ? MDB_SUCCESS : ErrCode();
+#endif
+ free(p);
+ return rc;
+}
+
+static int
+mdb_env_write_meta(MDB_txn *txn)
+{
+ MDB_env *env;
+ MDB_meta meta, metab;
+ off_t off;
+ int rc, len, toggle;
+ char *ptr;
+#ifdef _WIN32
+ OVERLAPPED ov;
+#endif
+
+ assert(txn != NULL);
+ assert(txn->mt_env != NULL);
+
+ toggle = !txn->mt_toggle;
+ DPRINTF("writing meta page %d for root page %lu",
+ toggle, txn->mt_dbs[MAIN_DBI].md_root);
+
+ env = txn->mt_env;
+
+ metab.mm_txnid = env->me_metas[toggle]->mm_txnid;
+ metab.mm_last_pg = env->me_metas[toggle]->mm_last_pg;
+
+ ptr = (char *)&meta;
+ off = offsetof(MDB_meta, mm_dbs[0].md_depth);
+ len = sizeof(MDB_meta) - off;
+
+ ptr += off;
+ meta.mm_dbs[0] = txn->mt_dbs[0];
+ meta.mm_dbs[1] = txn->mt_dbs[1];
+ meta.mm_last_pg = txn->mt_next_pgno - 1;
+ meta.mm_txnid = txn->mt_txnid;
+
+ if (toggle)
+ off += env->me_psize;
+ off += PAGEHDRSZ;
+
+ /* Write to the SYNC fd */
+#ifdef _WIN32
+ {
+ memset(&ov, 0, sizeof(ov));
+ ov.Offset = off;
+ WriteFile(env->me_mfd, ptr, len, (DWORD *)&rc, &ov);
+ }
+#else
+ rc = pwrite(env->me_mfd, ptr, len, off);
+#endif
+ if (rc != len) {
+ int r2;
+ rc = ErrCode();
+ DPUTS("write failed, disk error?");
+ /* On a failure, the pagecache still contains the new data.
+ * Write some old data back, to prevent it from being used.
+ * Use the non-SYNC fd; we know it will fail anyway.
+ */
+ meta.mm_last_pg = metab.mm_last_pg;
+ meta.mm_txnid = metab.mm_txnid;
+#ifdef _WIN32
+ WriteFile(env->me_fd, ptr, len, NULL, &ov);
+#else
+ r2 = pwrite(env->me_fd, ptr, len, off);
+#endif
+ env->me_flags |= MDB_FATAL_ERROR;
+ return rc;
+ }
+ /* Memory ordering issues are irrelevant; since the entire writer
+ * is wrapped by wmutex, all of these changes will become visible
+ * after the wmutex is unlocked. Since the DB is multi-version,
+ * readers will get consistent data regardless of how fresh or
+ * how stale their view of these values is.
+ */
+ LAZY_MUTEX_LOCK(&env->me_txns->mti_mutex);
+ txn->mt_env->me_txns->mti_me_toggle = toggle;
+ txn->mt_env->me_txns->mti_txnid = txn->mt_txnid;
+ LAZY_MUTEX_UNLOCK(&env->me_txns->mti_mutex);
+
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_env_read_meta(MDB_env *env, int *which)
+{
+ int toggle = 0;
+
+ assert(env != NULL);
+
+ if (env->me_metas[0]->mm_txnid < env->me_metas[1]->mm_txnid)
+ toggle = 1;
+
+ DPRINTF("Using meta page %d", toggle);
+ *which = toggle;
+
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_create(MDB_env **env)
+{
+ MDB_env *e;
+
+ e = calloc(1, sizeof(MDB_env));
+ if (!e) return ENOMEM;
+
+ e->me_maxreaders = DEFAULT_READERS;
+ e->me_maxdbs = 2;
+ e->me_fd = INVALID_HANDLE_VALUE;
+ e->me_lfd = INVALID_HANDLE_VALUE;
+ e->me_mfd = INVALID_HANDLE_VALUE;
+ *env = e;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_set_mapsize(MDB_env *env, size_t size)
+{
+ if (env->me_map)
+ return EINVAL;
+ env->me_mapsize = size;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_set_maxdbs(MDB_env *env, int dbs)
+{
+ if (env->me_map)
+ return EINVAL;
+ env->me_maxdbs = dbs;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_set_maxreaders(MDB_env *env, int readers)
+{
+ if (env->me_map)
+ return EINVAL;
+ env->me_maxreaders = readers;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_get_maxreaders(MDB_env *env, int *readers)
+{
+ if (!env || !readers)
+ return EINVAL;
+ *readers = env->me_maxreaders;
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_env_open2(MDB_env *env, unsigned int flags)
+{
+ int i, newenv = 0, toggle;
+ MDB_meta meta;
+ MDB_page *p;
+
+ env->me_flags = flags;
+
+ memset(&meta, 0, sizeof(meta));
+
+ if ((i = mdb_env_read_header(env, &meta)) != 0) {
+ if (i != ENOENT)
+ return i;
+ DPUTS("new mdbenv");
+ newenv = 1;
+ }
+
+ if (!env->me_mapsize) {
+ env->me_mapsize = newenv ? DEFAULT_MAPSIZE : meta.mm_mapsize;
+ }
+
+#ifdef _WIN32
+ {
+ HANDLE mh;
+ LONG sizelo, sizehi;
+ sizelo = env->me_mapsize & 0xffffffff;
+ sizehi = env->me_mapsize >> 16; /* pointless on WIN32, only needed on W64 */
+ sizehi >>= 16;
+ /* Windows won't create mappings for zero length files.
+ * Just allocate the maxsize right now.
+ */
+ if (newenv) {
+ SetFilePointer(env->me_fd, sizelo, sizehi ? &sizehi : NULL, 0);
+ if (!SetEndOfFile(env->me_fd))
+ return ErrCode();
+ SetFilePointer(env->me_fd, 0, NULL, 0);
+ }
+ mh = CreateFileMapping(env->me_fd, NULL, PAGE_READONLY,
+ sizehi, sizelo, NULL);
+ if (!mh)
+ return ErrCode();
+ env->me_map = MapViewOfFileEx(mh, FILE_MAP_READ, 0, 0, env->me_mapsize,
+ meta.mm_address);
+ CloseHandle(mh);
+ if (!env->me_map)
+ return ErrCode();
+ }
+#else
+ i = MAP_SHARED;
+ if (meta.mm_address && (flags & MDB_FIXEDMAP))
+ i |= MAP_FIXED;
+ env->me_map = mmap(meta.mm_address, env->me_mapsize, PROT_READ, i,
+ env->me_fd, 0);
+ if (env->me_map == MAP_FAILED)
+ return ErrCode();
+#endif
+
+ if (newenv) {
+ meta.mm_mapsize = env->me_mapsize;
+ if (flags & MDB_FIXEDMAP)
+ meta.mm_address = env->me_map;
+ i = mdb_env_init_meta(env, &meta);
+ if (i != MDB_SUCCESS) {
+ munmap(env->me_map, env->me_mapsize);
+ return i;
+ }
+ }
+ env->me_psize = meta.mm_psize;
+
+ env->me_maxpg = env->me_mapsize / env->me_psize;
+
+ p = (MDB_page *)env->me_map;
+ env->me_metas[0] = METADATA(p);
+ env->me_metas[1] = (MDB_meta *)((char *)env->me_metas[0] + meta.mm_psize);
+
+ if ((i = mdb_env_read_meta(env, &toggle)) != 0)
+ return i;
+
+ DPRINTF("opened database version %u, pagesize %u",
+ env->me_metas[toggle]->mm_version, env->me_psize);
+ DPRINTF("depth: %u", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_depth);
+ DPRINTF("entries: %lu", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_entries);
+ DPRINTF("branch pages: %lu", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_branch_pages);
+ DPRINTF("leaf pages: %lu", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_leaf_pages);
+ DPRINTF("overflow pages: %lu", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_overflow_pages);
+ DPRINTF("root: %lu", env->me_metas[toggle]->mm_dbs[MAIN_DBI].md_root);
+
+ return MDB_SUCCESS;
+}
+
+#ifndef _WIN32
+/* Windows doesn't support destructor callbacks for thread-specific storage */
+static void
+mdb_env_reader_dest(void *ptr)
+{
+ MDB_reader *reader = ptr;
+
+ reader->mr_txnid = 0;
+ reader->mr_pid = 0;
+ reader->mr_tid = 0;
+}
+#endif
+
+/* downgrade the exclusive lock on the region back to shared */
+static void
+mdb_env_share_locks(MDB_env *env)
+{
+ int toggle = 0;
+
+ if (env->me_metas[0]->mm_txnid < env->me_metas[1]->mm_txnid)
+ toggle = 1;
+ env->me_txns->mti_me_toggle = toggle;
+ env->me_txns->mti_txnid = env->me_metas[toggle]->mm_txnid;
+
+#ifdef _WIN32
+ {
+ OVERLAPPED ov;
+ /* First acquire a shared lock. The Unlock will
+ * then release the existing exclusive lock.
+ */
+ memset(&ov, 0, sizeof(ov));
+ LockFileEx(env->me_lfd, 0, 0, 1, 0, &ov);
+ UnlockFile(env->me_lfd, 0, 0, 1, 0);
+ }
+#else
+ {
+ struct flock lock_info;
+ /* The shared lock replaces the existing lock */
+ memset((void *)&lock_info, 0, sizeof(lock_info));
+ lock_info.l_type = F_RDLCK;
+ lock_info.l_whence = SEEK_SET;
+ lock_info.l_start = 0;
+ lock_info.l_len = 1;
+ fcntl(env->me_lfd, F_SETLK, &lock_info);
+ }
+#endif
+}
+
+static int
+mdb_env_setup_locks(MDB_env *env, char *lpath, int mode, int *excl)
+{
+ int rc;
+ off_t size, rsize;
+
+ *excl = 0;
+
+#ifdef _WIN32
+ if ((env->me_lfd = CreateFile(lpath, GENERIC_READ|GENERIC_WRITE,
+ FILE_SHARE_READ|FILE_SHARE_WRITE, NULL, OPEN_ALWAYS,
+ FILE_ATTRIBUTE_NORMAL, NULL)) == INVALID_HANDLE_VALUE) {
+ rc = ErrCode();
+ return rc;
+ }
+ /* Try to get exclusive lock. If we succeed, then
+ * nobody is using the lock region and we should initialize it.
+ */
+ {
+ if (LockFile(env->me_lfd, 0, 0, 1, 0)) {
+ *excl = 1;
+ } else {
+ OVERLAPPED ov;
+ memset(&ov, 0, sizeof(ov));
+ if (!LockFileEx(env->me_lfd, 0, 0, 1, 0, &ov)) {
+ rc = ErrCode();
+ goto fail;
+ }
+ }
+ }
+ size = GetFileSize(env->me_lfd, NULL);
+#else
+ if ((env->me_lfd = open(lpath, O_RDWR|O_CREAT, mode)) == -1) {
+ rc = ErrCode();
+ return rc;
+ }
+ /* Try to get exclusive lock. If we succeed, then
+ * nobody is using the lock region and we should initialize it.
+ */
+ {
+ struct flock lock_info;
+ memset((void *)&lock_info, 0, sizeof(lock_info));
+ lock_info.l_type = F_WRLCK;
+ lock_info.l_whence = SEEK_SET;
+ lock_info.l_start = 0;
+ lock_info.l_len = 1;
+ rc = fcntl(env->me_lfd, F_SETLK, &lock_info);
+ if (rc == 0) {
+ *excl = 1;
+ } else {
+ lock_info.l_type = F_RDLCK;
+ rc = fcntl(env->me_lfd, F_SETLKW, &lock_info);
+ if (rc) {
+ rc = ErrCode();
+ goto fail;
+ }
+ }
+ }
+ size = lseek(env->me_lfd, 0, SEEK_END);
+#endif
+ rsize = (env->me_maxreaders-1) * sizeof(MDB_reader) + sizeof(MDB_txninfo);
+ if (size < rsize && *excl) {
+#ifdef _WIN32
+ SetFilePointer(env->me_lfd, rsize, NULL, 0);
+ if (!SetEndOfFile(env->me_lfd)) {
+ rc = ErrCode();
+ goto fail;
+ }
+#else
+ if (ftruncate(env->me_lfd, rsize) != 0) {
+ rc = ErrCode();
+ goto fail;
+ }
+#endif
+ } else {
+ rsize = size;
+ size = rsize - sizeof(MDB_txninfo);
+ env->me_maxreaders = size/sizeof(MDB_reader) + 1;
+ }
+#ifdef _WIN32
+ {
+ HANDLE mh;
+ mh = CreateFileMapping(env->me_lfd, NULL, PAGE_READWRITE,
+ 0, 0, NULL);
+ if (!mh) {
+ rc = ErrCode();
+ goto fail;
+ }
+ env->me_txns = MapViewOfFileEx(mh, FILE_MAP_WRITE, 0, 0, rsize, NULL);
+ CloseHandle(mh);
+ if (!env->me_txns) {
+ rc = ErrCode();
+ goto fail;
+ }
+ }
+#else
+ env->me_txns = mmap(0, rsize, PROT_READ|PROT_WRITE, MAP_SHARED,
+ env->me_lfd, 0);
+ if (env->me_txns == MAP_FAILED) {
+ rc = ErrCode();
+ goto fail;
+ }
+#endif
+ if (*excl) {
+#ifdef _WIN32
+ char *ptr;
+ if (!mdb_sec_inited) {
+ InitializeSecurityDescriptor(&mdb_null_sd,
+ SECURITY_DESCRIPTOR_REVISION);
+ SetSecurityDescriptorDacl(&mdb_null_sd, TRUE, 0, FALSE);
+ mdb_all_sa.nLength = sizeof(SECURITY_ATTRIBUTES);
+ mdb_all_sa.bInheritHandle = FALSE;
+ mdb_all_sa.lpSecurityDescriptor = &mdb_null_sd;
+ mdb_sec_inited = 1;
+ }
+ /* FIXME: only using up to 20 characters of the env path here,
+ * probably not enough to assure uniqueness...
+ */
+ sprintf(env->me_txns->mti_rmname, "Global\\MDBr%.20s", lpath);
+ ptr = env->me_txns->mti_rmname + sizeof("Global\\MDBr");
+ while ((ptr = strchr(ptr, '\\')))
+ *ptr++ = '/';
+ env->me_rmutex = CreateMutex(&mdb_all_sa, FALSE, env->me_txns->mti_rmname);
+ if (!env->me_rmutex) {
+ rc = ErrCode();
+ goto fail;
+ }
+ sprintf(env->me_txns->mti_rmname, "Global\\MDBw%.20s", lpath);
+ ptr = env->me_txns->mti_rmname + sizeof("Global\\MDBw");
+ while ((ptr = strchr(ptr, '\\')))
+ *ptr++ = '/';
+ env->me_wmutex = CreateMutex(&mdb_all_sa, FALSE, env->me_txns->mti_rmname);
+ if (!env->me_wmutex) {
+ rc = ErrCode();
+ goto fail;
+ }
+#else
+ pthread_mutexattr_t mattr;
+
+ pthread_mutexattr_init(&mattr);
+ rc = pthread_mutexattr_setpshared(&mattr, PTHREAD_PROCESS_SHARED);
+ if (rc) {
+ goto fail;
+ }
+ pthread_mutex_init(&env->me_txns->mti_mutex, &mattr);
+ pthread_mutex_init(&env->me_txns->mti_wmutex, &mattr);
+#endif
+ env->me_txns->mti_version = MDB_VERSION;
+ env->me_txns->mti_magic = MDB_MAGIC;
+ env->me_txns->mti_txnid = 0;
+ env->me_txns->mti_numreaders = 0;
+ env->me_txns->mti_me_toggle = 0;
+
+ } else {
+ if (env->me_txns->mti_magic != MDB_MAGIC) {
+ DPUTS("lock region has invalid magic");
+ rc = EINVAL;
+ goto fail;
+ }
+ if (env->me_txns->mti_version != MDB_VERSION) {
+ DPRINTF("lock region is version %u, expected version %u",
+ env->me_txns->mti_version, MDB_VERSION);
+ rc = MDB_VERSION_MISMATCH;
+ goto fail;
+ }
+ rc = ErrCode();
+ if (rc != EACCES && rc != EAGAIN) {
+ goto fail;
+ }
+#ifdef _WIN32
+ env->me_rmutex = OpenMutex(SYNCHRONIZE, FALSE, env->me_txns->mti_rmname);
+ if (!env->me_rmutex) {
+ rc = ErrCode();
+ goto fail;
+ }
+ env->me_wmutex = OpenMutex(SYNCHRONIZE, FALSE, env->me_txns->mti_wmname);
+ if (!env->me_wmutex) {
+ rc = ErrCode();
+ goto fail;
+ }
+#endif
+ }
+ return MDB_SUCCESS;
+
+fail:
+ close(env->me_lfd);
+ env->me_lfd = INVALID_HANDLE_VALUE;
+ return rc;
+
+}
+
+#define LOCKNAME "/lock.mdb"
+#define DATANAME "/data.mdb"
+int
+mdb_env_open(MDB_env *env, const char *path, unsigned int flags, mode_t mode)
+{
+ int oflags, rc, len, excl;
+ char *lpath, *dpath;
+
+ len = strlen(path);
+ lpath = malloc(len + sizeof(LOCKNAME) + len + sizeof(DATANAME));
+ if (!lpath)
+ return ENOMEM;
+ dpath = lpath + len + sizeof(LOCKNAME);
+ sprintf(lpath, "%s" LOCKNAME, path);
+ sprintf(dpath, "%s" DATANAME, path);
+
+ rc = mdb_env_setup_locks(env, lpath, mode, &excl);
+ if (rc)
+ goto leave;
+
+#ifdef _WIN32
+ if (F_ISSET(flags, MDB_RDONLY)) {
+ oflags = GENERIC_READ;
+ len = OPEN_EXISTING;
+ } else {
+ oflags = GENERIC_READ|GENERIC_WRITE;
+ len = OPEN_ALWAYS;
+ }
+ mode = FILE_ATTRIBUTE_NORMAL;
+ if ((env->me_fd = CreateFile(dpath, oflags, FILE_SHARE_READ|FILE_SHARE_WRITE,
+ NULL, len, mode, NULL)) == INVALID_HANDLE_VALUE) {
+ rc = ErrCode();
+ goto leave;
+ }
+#else
+ if (F_ISSET(flags, MDB_RDONLY))
+ oflags = O_RDONLY;
+ else
+ oflags = O_RDWR | O_CREAT;
+
+ if ((env->me_fd = open(dpath, oflags, mode)) == -1) {
+ rc = ErrCode();
+ goto leave;
+ }
+#endif
+
+ if ((rc = mdb_env_open2(env, flags)) == MDB_SUCCESS) {
+ /* synchronous fd for meta writes */
+#ifdef _WIN32
+ if (!(flags & (MDB_RDONLY|MDB_NOSYNC)))
+ mode |= FILE_FLAG_WRITE_THROUGH;
+ if ((env->me_mfd = CreateFile(dpath, oflags, FILE_SHARE_READ|FILE_SHARE_WRITE,
+ NULL, len, mode, NULL)) == INVALID_HANDLE_VALUE) {
+ rc = ErrCode();
+ goto leave;
+ }
+#else
+ if (!(flags & (MDB_RDONLY|MDB_NOSYNC)))
+ oflags |= MDB_DSYNC;
+ if ((env->me_mfd = open(dpath, oflags, mode)) == -1) {
+ rc = ErrCode();
+ goto leave;
+ }
+#endif
+ env->me_path = strdup(path);
+ DPRINTF("opened dbenv %p", (void *) env);
+ pthread_key_create(&env->me_txkey, mdb_env_reader_dest);
+ LAZY_RWLOCK_INIT(&env->me_dblock, NULL);
+ if (excl)
+ mdb_env_share_locks(env);
+ env->me_dbxs = calloc(env->me_maxdbs, sizeof(MDB_dbx));
+ env->me_dbs[0] = calloc(env->me_maxdbs, sizeof(MDB_db));
+ env->me_dbs[1] = calloc(env->me_maxdbs, sizeof(MDB_db));
+ env->me_numdbs = 2;
+ }
+
+leave:
+ if (rc) {
+ if (env->me_fd != INVALID_HANDLE_VALUE) {
+ close(env->me_fd);
+ env->me_fd = INVALID_HANDLE_VALUE;
+ }
+ if (env->me_lfd != INVALID_HANDLE_VALUE) {
+ close(env->me_lfd);
+ env->me_lfd = INVALID_HANDLE_VALUE;
+ }
+ }
+ free(lpath);
+ return rc;
+}
+
+void
+mdb_env_close(MDB_env *env)
+{
+ MDB_dpage *dp;
+
+ if (env == NULL)
+ return;
+
+ while (env->me_dpages) {
+ dp = env->me_dpages;
+ env->me_dpages = (MDB_dpage *)dp->h.md_parent;
+ free(dp);
+ }
+
+ free(env->me_dbs[1]);
+ free(env->me_dbs[0]);
+ free(env->me_dbxs);
+ free(env->me_path);
+
+ LAZY_RWLOCK_DESTROY(&env->me_dblock);
+ pthread_key_delete(env->me_txkey);
+
+ if (env->me_map) {
+ munmap(env->me_map, env->me_mapsize);
+ }
+ close(env->me_mfd);
+ close(env->me_fd);
+ if (env->me_txns) {
+ pid_t pid = getpid();
+ unsigned int i;
+ for (i=0; i<env->me_txns->mti_numreaders; i++)
+ if (env->me_txns->mti_readers[i].mr_pid == pid)
+ env->me_txns->mti_readers[i].mr_pid = 0;
+ munmap(env->me_txns, (env->me_maxreaders-1)*sizeof(MDB_reader)+sizeof(MDB_txninfo));
+ }
+ close(env->me_lfd);
+ free(env);
+}
+
+/* Search for key within a leaf page, using binary search.
+ * Returns the smallest entry larger or equal to the key.
+ * If exactp is non-null, stores whether the found entry was an exact match
+ * in *exactp (1 or 0).
+ * If kip is non-null, stores the index of the found entry in *kip.
+ * If no entry larger or equal to the key is found, returns NULL.
+ */
+static MDB_node *
+mdb_search_node(MDB_txn *txn, MDB_dbi dbi, MDB_page *mp, MDB_val *key,
+ int *exactp, unsigned int *kip)
+{
+ unsigned int i = 0;
+ int low, high;
+ int rc = 0;
+ MDB_node *node = NULL;
+ MDB_val nodekey;
+ DKBUF;
+
+ DPRINTF("searching %u keys in %s page %lu",
+ NUMKEYS(mp),
+ IS_LEAF(mp) ? "leaf" : "branch",
+ mp->mp_pgno);
+
+ assert(NUMKEYS(mp) > 0);
+
+ memset(&nodekey, 0, sizeof(nodekey));
+
+ low = IS_LEAF(mp) ? 0 : 1;
+ high = NUMKEYS(mp) - 1;
+ while (low <= high) {
+ i = (low + high) >> 1;
+
+ if (IS_LEAF2(mp)) {
+ nodekey.mv_size = txn->mt_dbs[dbi].md_pad;
+ nodekey.mv_data = LEAF2KEY(mp, i, nodekey.mv_size);
+ } else {
+ node = NODEPTR(mp, i);
+
+ nodekey.mv_size = node->mn_ksize;
+ nodekey.mv_data = NODEKEY(node);
+ }
+
+ rc = mdb_cmp(txn, dbi, key, &nodekey);
+
+ if (IS_LEAF(mp))
+ DPRINTF("found leaf index %u [%s], rc = %i",
+ i, DKEY(&nodekey), rc);
+ else
+ DPRINTF("found branch index %u [%s -> %lu], rc = %i",
+ i, DKEY(&nodekey), NODEPGNO(node), rc);
+
+ if (rc == 0)
+ break;
+ if (rc > 0)
+ low = i + 1;
+ else
+ high = i - 1;
+ }
+
+ if (rc > 0) { /* Found entry is less than the key. */
+ i++; /* Skip to get the smallest entry larger than key. */
+ if (i >= NUMKEYS(mp))
+ /* There is no entry larger or equal to the key. */
+ return NULL;
+ }
+ if (exactp)
+ *exactp = (rc == 0);
+ if (kip) /* Store the key index if requested. */
+ *kip = i;
+
+ /* nodeptr is fake for LEAF2 */
+ return IS_LEAF2(mp) ? NODEPTR(mp, 0) : NODEPTR(mp, i);
+}
+
+static void
+cursor_pop_page(MDB_cursor *cursor)
+{
+ MDB_ppage *top;
+
+ if (cursor->mc_snum) {
+ top = CURSOR_TOP(cursor);
+ cursor->mc_snum--;
+
+ DPRINTF("popped page %lu off db %u cursor %p", top->mp_page->mp_pgno,
+ cursor->mc_dbi, (void *) cursor);
+ }
+}
+
+static MDB_ppage *
+cursor_push_page(MDB_cursor *cursor, MDB_page *mp)
+{
+ MDB_ppage *ppage;
+
+ DPRINTF("pushing page %lu on db %u cursor %p", mp->mp_pgno,
+ cursor->mc_dbi, (void *) cursor);
+
+ assert(cursor->mc_snum < CURSOR_STACK);
+
+ ppage = &cursor->mc_stack[cursor->mc_snum++];
+ ppage->mp_page = mp;
+ ppage->mp_ki = 0;
+ return ppage;
+}
+
+static int
+mdb_get_page(MDB_txn *txn, pgno_t pgno, MDB_page **ret)
+{
+ MDB_page *p = NULL;
+
+ if (!F_ISSET(txn->mt_flags, MDB_TXN_RDONLY) && txn->mt_u.dirty_list[0].mid) {
+ MDB_dpage *dp;
+ MIDL2 id;
+ unsigned x;
+ id.mid = pgno;
+ x = mdb_midl2_search(txn->mt_u.dirty_list, &id);
+ if (x <= txn->mt_u.dirty_list[0].mid && txn->mt_u.dirty_list[x].mid == pgno) {
+ dp = txn->mt_u.dirty_list[x].mptr;
+ p = &dp->p;
+ }
+ }
+ if (!p) {
+ if (pgno <= txn->mt_env->me_metas[txn->mt_toggle]->mm_last_pg)
+ p = (MDB_page *)(txn->mt_env->me_map + txn->mt_env->me_psize * pgno);
+ }
+ *ret = p;
+ if (!p) {
+ DPRINTF("page %lu not found", pgno);
+ assert(p != NULL);
+ }
+ return (p != NULL) ? MDB_SUCCESS : MDB_PAGE_NOTFOUND;
+}
+
+static int
+mdb_search_page_root(MDB_txn *txn, MDB_dbi dbi, MDB_val *key,
+ MDB_cursor *cursor, int modify, MDB_pageparent *mpp)
+{
+ MDB_page *mp = mpp->mp_page;
+ DKBUF;
+ int rc;
+
+ if (cursor && cursor_push_page(cursor, mp) == NULL)
+ return ENOMEM;
+
+ while (IS_BRANCH(mp)) {
+ unsigned int i = 0;
+ MDB_node *node;
+
+ DPRINTF("branch page %lu has %u keys", mp->mp_pgno, NUMKEYS(mp));
+ assert(NUMKEYS(mp) > 1);
+ DPRINTF("found index 0 to page %lu", NODEPGNO(NODEPTR(mp, 0)));
+
+ if (key == NULL) /* Initialize cursor to first page. */
+ i = 0;
+ else if (key->mv_size > MAXKEYSIZE && key->mv_data == NULL) {
+ /* cursor to last page */
+ i = NUMKEYS(mp)-1;
+ } else {
+ int exact;
+ node = mdb_search_node(txn, dbi, mp, key, &exact, &i);
+ if (node == NULL)
+ i = NUMKEYS(mp) - 1;
+ else if (!exact) {
+ assert(i > 0);
+ i--;
+ }
+ }
+
+ if (key)
+ DPRINTF("following index %u for key [%s]",
+ i, DKEY(key));
+ assert(i < NUMKEYS(mp));
+ node = NODEPTR(mp, i);
+
+ if (cursor)
+ CURSOR_TOP(cursor)->mp_ki = i;
+
+ mpp->mp_parent = mp;
+ if ((rc = mdb_get_page(txn, NODEPGNO(node), &mp)))
+ return rc;
+ mpp->mp_pi = i;
+ mpp->mp_page = mp;
+
+ if (cursor && cursor_push_page(cursor, mp) == NULL)
+ return ENOMEM;
+
+ if (modify) {
+ MDB_dhead *dh = ((MDB_dhead *)mp)-1;
+ if ((rc = mdb_touch(txn, dbi, mpp)) != 0)
+ return rc;
+ dh = ((MDB_dhead *)mpp->mp_page)-1;
+ dh->md_parent = mpp->mp_parent;
+ dh->md_pi = mpp->mp_pi;
+ }
+
+ mp = mpp->mp_page;
+ }
+
+ if (!IS_LEAF(mp)) {
+ DPRINTF("internal error, index points to a %02X page!?",
+ mp->mp_flags);
+ return MDB_CORRUPTED;
+ }
+
+ DPRINTF("found leaf page %lu for key [%s]", mp->mp_pgno,
+ key ? DKEY(key) : NULL);
+
+ return MDB_SUCCESS;
+}
+
+/* Search for the page a given key should be in.
+ * Stores a pointer to the found page in *mpp.
+ * If key is NULL, search for the lowest page (used by mdb_cursor_first).
+ * If cursor is non-null, pushes parent pages on the cursor stack.
+ * If modify is true, visited pages are updated with new page numbers.
+ */
+static int
+mdb_search_page(MDB_txn *txn, MDB_dbi dbi, MDB_val *key,
+ MDB_cursor *cursor, int modify, MDB_pageparent *mpp)
+{
+ int rc;
+ pgno_t root;
+
+ /* Choose which root page to start with. If a transaction is given
+ * use the root page from the transaction, otherwise read the last
+ * committed root page.
+ */
+ if (F_ISSET(txn->mt_flags, MDB_TXN_ERROR)) {
+ DPUTS("transaction has failed, must abort");
+ return EINVAL;
+ } else
+ root = txn->mt_dbs[dbi].md_root;
+
+ if (root == P_INVALID) { /* Tree is empty. */
+ DPUTS("tree is empty");
+ return MDB_NOTFOUND;
+ }
+
+ if ((rc = mdb_get_page(txn, root, &mpp->mp_page)))
+ return rc;
+
+ DPRINTF("db %u root page %lu has flags 0x%X",
+ dbi, root, mpp->mp_page->mp_flags);
+
+ if (modify) {
+ /* For sub-databases, update main root first */
+ if (dbi > MAIN_DBI && !txn->mt_dbxs[dbi].md_dirty) {
+ MDB_pageparent mp2;
+ rc = mdb_search_page(txn, MAIN_DBI, &txn->mt_dbxs[dbi].md_name,
+ NULL, 1, &mp2);
+ if (rc)
+ return rc;
+ txn->mt_dbxs[dbi].md_dirty = 1;
+ }
+ if (!F_ISSET(mpp->mp_page->mp_flags, P_DIRTY)) {
+ mpp->mp_parent = NULL;
+ mpp->mp_pi = 0;
+ if ((rc = mdb_touch(txn, dbi, mpp)))
+ return rc;
+ txn->mt_dbs[dbi].md_root = mpp->mp_page->mp_pgno;
+ }
+ }
+
+ return mdb_search_page_root(txn, dbi, key, cursor, modify, mpp);
+}
+
+static int
+mdb_read_data(MDB_txn *txn, MDB_node *leaf, MDB_val *data)
+{
+ MDB_page *omp; /* overflow mpage */
+ pgno_t pgno;
+ int rc;
+
+ if (!F_ISSET(leaf->mn_flags, F_BIGDATA)) {
+ data->mv_size = leaf->mn_dsize;
+ data->mv_data = NODEDATA(leaf);
+ return MDB_SUCCESS;
+ }
+
+ /* Read overflow data.
+ */
+ data->mv_size = leaf->mn_dsize;
+ memcpy(&pgno, NODEDATA(leaf), sizeof(pgno));
+ if ((rc = mdb_get_page(txn, pgno, &omp))) {
+ DPRINTF("read overflow page %lu failed", pgno);
+ return rc;
+ }
+ data->mv_data = METADATA(omp);
+
+ return MDB_SUCCESS;
+}
+
+int
+mdb_get(MDB_txn *txn, MDB_dbi dbi,
+ MDB_val *key, MDB_val *data)
+{
+ int rc, exact;
+ MDB_node *leaf;
+ MDB_pageparent mpp;
+ DKBUF;
+
+ assert(key);
+ assert(data);
+ DPRINTF("===> get db %u key [%s]", dbi, DKEY(key));
+
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ if (key->mv_size == 0 || key->mv_size > MAXKEYSIZE) {
+ return EINVAL;
+ }
+
+ if ((rc = mdb_search_page(txn, dbi, key, NULL, 0, &mpp)) != MDB_SUCCESS)
+ return rc;
+
+ leaf = mdb_search_node(txn, dbi, mpp.mp_page, key, &exact, NULL);
+ if (leaf && exact) {
+ /* Return first duplicate data item */
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ MDB_xcursor mx;
+
+ mdb_xcursor_init0(txn, dbi, &mx);
+ mdb_xcursor_init1(txn, dbi, &mx, mpp.mp_page, leaf);
+ rc = mdb_search_page(&mx.mx_txn, mx.mx_cursor.mc_dbi, NULL, NULL, 0, &mpp);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ if (IS_LEAF2(mpp.mp_page)) {
+ data->mv_size = txn->mt_dbs[dbi].md_pad;
+ data->mv_data = LEAF2KEY(mpp.mp_page, 0, data->mv_size);
+ } else {
+ leaf = NODEPTR(mpp.mp_page, 0);
+ data->mv_size = NODEKSZ(leaf);
+ data->mv_data = NODEKEY(leaf);
+ }
+ } else {
+ rc = mdb_read_data(txn, leaf, data);
+ }
+ } else {
+ rc = MDB_NOTFOUND;
+ }
+
+ return rc;
+}
+
+static int
+mdb_sibling(MDB_cursor *cursor, int move_right)
+{
+ int rc;
+ MDB_node *indx;
+ MDB_ppage *parent;
+ MDB_page *mp;
+
+ if (cursor->mc_snum < 2) {
+ return MDB_NOTFOUND; /* root has no siblings */
+ }
+ parent = CURSOR_PARENT(cursor);
+
+ DPRINTF("parent page is page %lu, index %u",
+ parent->mp_page->mp_pgno, parent->mp_ki);
+
+ cursor_pop_page(cursor);
+ if (move_right ? (parent->mp_ki + 1 >= NUMKEYS(parent->mp_page))
+ : (parent->mp_ki == 0)) {
+ DPRINTF("no more keys left, moving to %s sibling",
+ move_right ? "right" : "left");
+ if ((rc = mdb_sibling(cursor, move_right)) != MDB_SUCCESS)
+ return rc;
+ parent = CURSOR_TOP(cursor);
+ } else {
+ if (move_right)
+ parent->mp_ki++;
+ else
+ parent->mp_ki--;
+ DPRINTF("just moving to %s index key %u",
+ move_right ? "right" : "left", parent->mp_ki);
+ }
+ assert(IS_BRANCH(parent->mp_page));
+
+ indx = NODEPTR(parent->mp_page, parent->mp_ki);
+ if ((rc = mdb_get_page(cursor->mc_txn, NODEPGNO(indx), &mp)))
+ return rc;;
+#if 0
+ mp->parent = parent->mp_page;
+ mp->parent_index = parent->mp_ki;
+#endif
+
+ cursor_push_page(cursor, mp);
+
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_cursor_next(MDB_cursor *cursor, MDB_val *key, MDB_val *data, MDB_cursor_op op)
+{
+ MDB_ppage *top;
+ MDB_page *mp;
+ MDB_node *leaf;
+ int rc;
+
+ if (cursor->mc_eof) {
+ return MDB_NOTFOUND;
+ }
+
+ assert(cursor->mc_initialized);
+
+ top = CURSOR_TOP(cursor);
+ mp = top->mp_page;
+
+ if (cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_flags & MDB_DUPSORT) {
+ leaf = NODEPTR(mp, top->mp_ki);
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ if (op == MDB_NEXT || op == MDB_NEXT_DUP) {
+ rc = mdb_cursor_next(&cursor->mc_xcursor->mx_cursor, data, NULL, MDB_NEXT);
+ if (op != MDB_NEXT || rc == MDB_SUCCESS)
+ return rc;
+ }
+ } else {
+ cursor->mc_xcursor->mx_cursor.mc_initialized = 0;
+ if (op == MDB_NEXT_DUP)
+ return MDB_NOTFOUND;
+ }
+ }
+
+ DPRINTF("cursor_next: top page is %lu in cursor %p", mp->mp_pgno, (void *) cursor);
+
+ if (top->mp_ki + 1 >= NUMKEYS(mp)) {
+ DPUTS("=====> move to next sibling page");
+ if (mdb_sibling(cursor, 1) != MDB_SUCCESS) {
+ cursor->mc_eof = 1;
+ return MDB_NOTFOUND;
+ }
+ top = CURSOR_TOP(cursor);
+ mp = top->mp_page;
+ DPRINTF("next page is %lu, key index %u", mp->mp_pgno, top->mp_ki);
+ } else
+ top->mp_ki++;
+
+ DPRINTF("==> cursor points to page %lu with %u keys, key index %u",
+ mp->mp_pgno, NUMKEYS(mp), top->mp_ki);
+
+ if (IS_LEAF2(mp)) {
+ key->mv_size = cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_pad;
+ key->mv_data = LEAF2KEY(mp, top->mp_ki, key->mv_size);
+ return MDB_SUCCESS;
+ }
+
+ assert(IS_LEAF(mp));
+ leaf = NODEPTR(mp, top->mp_ki);
+
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ mdb_xcursor_init1(cursor->mc_txn, cursor->mc_dbi, cursor->mc_xcursor, mp, leaf);
+ }
+ if (data) {
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, data) != MDB_SUCCESS))
+ return rc;
+
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ rc = mdb_cursor_first(&cursor->mc_xcursor->mx_cursor, data, NULL);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ }
+ }
+
+ MDB_SET_KEY(leaf, key);
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_cursor_prev(MDB_cursor *cursor, MDB_val *key, MDB_val *data, MDB_cursor_op op)
+{
+ MDB_ppage *top;
+ MDB_page *mp;
+ MDB_node *leaf;
+ int rc;
+
+ assert(cursor->mc_initialized);
+
+ top = CURSOR_TOP(cursor);
+ mp = top->mp_page;
+
+ if (cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_flags & MDB_DUPSORT) {
+ leaf = NODEPTR(mp, top->mp_ki);
+ if (op == MDB_PREV || op == MDB_PREV_DUP) {
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ rc = mdb_cursor_prev(&cursor->mc_xcursor->mx_cursor, data, NULL, MDB_PREV);
+ if (op != MDB_PREV || rc == MDB_SUCCESS)
+ return rc;
+ } else {
+ cursor->mc_xcursor->mx_cursor.mc_initialized = 0;
+ if (op == MDB_PREV_DUP)
+ return MDB_NOTFOUND;
+ }
+ }
+ }
+
+ DPRINTF("cursor_prev: top page is %lu in cursor %p", mp->mp_pgno, (void *) cursor);
+
+ if (top->mp_ki == 0) {
+ DPUTS("=====> move to prev sibling page");
+ if (mdb_sibling(cursor, 0) != MDB_SUCCESS) {
+ cursor->mc_initialized = 0;
+ return MDB_NOTFOUND;
+ }
+ top = CURSOR_TOP(cursor);
+ mp = top->mp_page;
+ top->mp_ki = NUMKEYS(mp) - 1;
+ DPRINTF("prev page is %lu, key index %u", mp->mp_pgno, top->mp_ki);
+ } else
+ top->mp_ki--;
+
+ cursor->mc_eof = 0;
+
+ DPRINTF("==> cursor points to page %lu with %u keys, key index %u",
+ mp->mp_pgno, NUMKEYS(mp), top->mp_ki);
+
+ if (IS_LEAF2(mp)) {
+ key->mv_size = cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_pad;
+ key->mv_data = LEAF2KEY(mp, top->mp_ki, key->mv_size);
+ return MDB_SUCCESS;
+ }
+
+ assert(IS_LEAF(mp));
+ leaf = NODEPTR(mp, top->mp_ki);
+
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ mdb_xcursor_init1(cursor->mc_txn, cursor->mc_dbi, cursor->mc_xcursor, mp, leaf);
+ }
+ if (data) {
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, data) != MDB_SUCCESS))
+ return rc;
+
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ rc = mdb_cursor_last(&cursor->mc_xcursor->mx_cursor, data, NULL);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ }
+ }
+
+ MDB_SET_KEY(leaf, key);
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_cursor_set(MDB_cursor *cursor, MDB_val *key, MDB_val *data,
+ MDB_cursor_op op, int *exactp)
+{
+ int rc;
+ MDB_node *leaf;
+ MDB_ppage *top;
+ MDB_pageparent mpp;
+ DKBUF;
+
+ assert(cursor);
+ assert(key);
+ assert(key->mv_size > 0);
+
+ /* See if we're already on the right page */
+ if (cursor->mc_initialized) {
+ MDB_val nodekey;
+ top = CURSOR_TOP(cursor);
+ /* Don't try this for LEAF2 pages. Maybe support that later. */
+ if ((top->mp_page->mp_flags & (P_LEAF|P_LEAF2)) == P_LEAF) {
+ leaf = NODEPTR(top->mp_page, 0);
+ MDB_SET_KEY(leaf, &nodekey);
+ rc = mdb_cmp(cursor->mc_txn, cursor->mc_dbi, key, &nodekey);
+ if (rc >= 0) {
+ leaf = NODEPTR(top->mp_page, NUMKEYS(top->mp_page)-1);
+ MDB_SET_KEY(leaf, &nodekey);
+ rc = mdb_cmp(cursor->mc_txn, cursor->mc_dbi, key, &nodekey);
+ if (rc <= 0) {
+ /* we're already on the right page */
+ mpp.mp_page = top->mp_page;
+ rc = 0;
+ goto set2;
+ }
+ }
+ }
+ }
+ cursor->mc_snum = 0;
+
+ rc = mdb_search_page(cursor->mc_txn, cursor->mc_dbi, key, cursor, 0, &mpp);
+ if (rc != MDB_SUCCESS)
+ return rc;
+
+ assert(IS_LEAF(mpp.mp_page));
+
+ top = CURSOR_TOP(cursor);
+set2:
+ leaf = mdb_search_node(cursor->mc_txn, cursor->mc_dbi, mpp.mp_page, key, exactp, &top->mp_ki);
+ if (exactp != NULL && !*exactp) {
+ /* MDB_SET specified and not an exact match. */
+ return MDB_NOTFOUND;
+ }
+
+ if (leaf == NULL) {
+ DPUTS("===> inexact leaf not found, goto sibling");
+ if ((rc = mdb_sibling(cursor, 1)) != MDB_SUCCESS)
+ return rc; /* no entries matched */
+ top = CURSOR_TOP(cursor);
+ top->mp_ki = 0;
+ mpp.mp_page = top->mp_page;
+ assert(IS_LEAF(mpp.mp_page));
+ leaf = NODEPTR(mpp.mp_page, 0);
+ }
+
+ cursor->mc_initialized = 1;
+ cursor->mc_eof = 0;
+
+ if (IS_LEAF2(mpp.mp_page)) {
+ key->mv_size = cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_pad;
+ key->mv_data = LEAF2KEY(mpp.mp_page, top->mp_ki, key->mv_size);
+ return MDB_SUCCESS;
+ }
+
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ mdb_xcursor_init1(cursor->mc_txn, cursor->mc_dbi, cursor->mc_xcursor, mpp.mp_page, leaf);
+ }
+ if (data) {
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ if (op == MDB_SET || op == MDB_SET_RANGE) {
+ rc = mdb_cursor_first(&cursor->mc_xcursor->mx_cursor, data, NULL);
+ } else {
+ int ex2, *ex2p;
+ if (op == MDB_GET_BOTH) {
+ ex2p = &ex2;
+ } else {
+ ex2p = NULL;
+ }
+ rc = mdb_cursor_set(&cursor->mc_xcursor->mx_cursor, data, NULL, MDB_SET_RANGE, ex2p);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ }
+ } else if (op == MDB_GET_BOTH || op == MDB_GET_BOTH_RANGE) {
+ MDB_val d2;
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, &d2)) != MDB_SUCCESS)
+ return rc;
+ rc = mdb_dcmp(cursor->mc_txn, cursor->mc_dbi, data, &d2);
+ if (rc) {
+ if (op == MDB_GET_BOTH || rc > 0)
+ return MDB_NOTFOUND;
+ }
+
+ } else {
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, data)) != MDB_SUCCESS)
+ return rc;
+ }
+ }
+
+ /* The key already matches in all other cases */
+ if (op == MDB_SET_RANGE)
+ MDB_SET_KEY(leaf, key);
+ DPRINTF("==> cursor placed on key [%s]", DKEY(key));
+
+ return rc;
+}
+
+static int
+mdb_cursor_first(MDB_cursor *cursor, MDB_val *key, MDB_val *data)
+{
+ int rc;
+ MDB_pageparent mpp;
+ MDB_node *leaf;
+
+ cursor->mc_snum = 0;
+
+ rc = mdb_search_page(cursor->mc_txn, cursor->mc_dbi, NULL, cursor, 0, &mpp);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ assert(IS_LEAF(mpp.mp_page));
+
+ leaf = NODEPTR(mpp.mp_page, 0);
+ cursor->mc_initialized = 1;
+ cursor->mc_eof = 0;
+
+ if (IS_LEAF2(mpp.mp_page)) {
+ key->mv_size = cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_pad;
+ key->mv_data = LEAF2KEY(mpp.mp_page, 0, key->mv_size);
+ return MDB_SUCCESS;
+ }
+
+ if (data) {
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ mdb_xcursor_init1(cursor->mc_txn, cursor->mc_dbi, cursor->mc_xcursor, mpp.mp_page, leaf);
+ rc = mdb_cursor_first(&cursor->mc_xcursor->mx_cursor, data, NULL);
+ if (rc)
+ return rc;
+ } else {
+ if (cursor->mc_xcursor)
+ cursor->mc_xcursor->mx_cursor.mc_initialized = 0;
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, data)) != MDB_SUCCESS)
+ return rc;
+ }
+ }
+ MDB_SET_KEY(leaf, key);
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_cursor_last(MDB_cursor *cursor, MDB_val *key, MDB_val *data)
+{
+ int rc;
+ MDB_ppage *top;
+ MDB_pageparent mpp;
+ MDB_node *leaf;
+ MDB_val lkey;
+
+ cursor->mc_snum = 0;
+
+ lkey.mv_size = MAXKEYSIZE+1;
+ lkey.mv_data = NULL;
+
+ rc = mdb_search_page(cursor->mc_txn, cursor->mc_dbi, &lkey, cursor, 0, &mpp);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ assert(IS_LEAF(mpp.mp_page));
+
+ leaf = NODEPTR(mpp.mp_page, NUMKEYS(mpp.mp_page)-1);
+ cursor->mc_initialized = 1;
+ cursor->mc_eof = 0;
+
+ top = CURSOR_TOP(cursor);
+ top->mp_ki = NUMKEYS(top->mp_page) - 1;
+
+ if (IS_LEAF2(mpp.mp_page)) {
+ key->mv_size = cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_pad;
+ key->mv_data = LEAF2KEY(mpp.mp_page, top->mp_ki, key->mv_size);
+ return MDB_SUCCESS;
+ }
+
+ if (data) {
+ if (F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ mdb_xcursor_init1(cursor->mc_txn, cursor->mc_dbi, cursor->mc_xcursor, mpp.mp_page, leaf);
+ rc = mdb_cursor_last(&cursor->mc_xcursor->mx_cursor, data, NULL);
+ if (rc)
+ return rc;
+ } else {
+ if ((rc = mdb_read_data(cursor->mc_txn, leaf, data)) != MDB_SUCCESS)
+ return rc;
+ }
+ }
+
+ MDB_SET_KEY(leaf, key);
+ return MDB_SUCCESS;
+}
+
+int
+mdb_cursor_get(MDB_cursor *cursor, MDB_val *key, MDB_val *data,
+ MDB_cursor_op op)
+{
+ int rc;
+ int exact = 0;
+
+ assert(cursor);
+
+ switch (op) {
+ case MDB_GET_BOTH:
+ case MDB_GET_BOTH_RANGE:
+ if (data == NULL || cursor->mc_xcursor == NULL) {
+ rc = EINVAL;
+ break;
+ }
+ /* FALLTHRU */
+ case MDB_SET:
+ case MDB_SET_RANGE:
+ if (key == NULL || key->mv_size == 0 || key->mv_size > MAXKEYSIZE) {
+ rc = EINVAL;
+ } else if (op == MDB_SET_RANGE)
+ rc = mdb_cursor_set(cursor, key, data, op, NULL);
+ else
+ rc = mdb_cursor_set(cursor, key, data, op, &exact);
+ break;
+ case MDB_GET_MULTIPLE:
+ if (data == NULL ||
+ !(cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_flags & MDB_DUPFIXED) ||
+ !cursor->mc_initialized) {
+ rc = EINVAL;
+ break;
+ }
+ rc = MDB_SUCCESS;
+ if (!cursor->mc_xcursor->mx_cursor.mc_initialized || cursor->mc_xcursor->mx_cursor.mc_eof)
+ break;
+ goto fetchm;
+ case MDB_NEXT_MULTIPLE:
+ if (data == NULL ||
+ !(cursor->mc_txn->mt_dbs[cursor->mc_dbi].md_flags & MDB_DUPFIXED)) {
+ rc = EINVAL;
+ break;
+ }
+ if (!cursor->mc_initialized)
+ rc = mdb_cursor_first(cursor, key, data);
+ else
+ rc = mdb_cursor_next(cursor, key, data, MDB_NEXT_DUP);
+ if (rc == MDB_SUCCESS) {
+ if (cursor->mc_xcursor->mx_cursor.mc_initialized) {
+ MDB_ppage *top;
+fetchm:
+ top = CURSOR_TOP(&cursor->mc_xcursor->mx_cursor);
+ data->mv_size = NUMKEYS(top->mp_page) *
+ cursor->mc_xcursor->mx_txn.mt_dbs[cursor->mc_xcursor->mx_cursor.mc_dbi].md_pad;
+ data->mv_data = METADATA(top->mp_page);
+ top->mp_ki = NUMKEYS(top->mp_page)-1;
+ } else {
+ rc = MDB_NOTFOUND;
+ }
+ }
+ break;
+ case MDB_NEXT:
+ case MDB_NEXT_DUP:
+ case MDB_NEXT_NODUP:
+ if (!cursor->mc_initialized)
+ rc = mdb_cursor_first(cursor, key, data);
+ else
+ rc = mdb_cursor_next(cursor, key, data, op);
+ break;
+ case MDB_PREV:
+ case MDB_PREV_DUP:
+ case MDB_PREV_NODUP:
+ if (!cursor->mc_initialized || cursor->mc_eof)
+ rc = mdb_cursor_last(cursor, key, data);
+ else
+ rc = mdb_cursor_prev(cursor, key, data, op);
+ break;
+ case MDB_FIRST:
+ rc = mdb_cursor_first(cursor, key, data);
+ break;
+ case MDB_LAST:
+ rc = mdb_cursor_last(cursor, key, data);
+ break;
+ default:
+ DPRINTF("unhandled/unimplemented cursor operation %u", op);
+ rc = EINVAL;
+ break;
+ }
+
+ return rc;
+}
+
+/* Allocate a page and initialize it
+ */
+static MDB_dpage *
+mdb_new_page(MDB_txn *txn, MDB_dbi dbi, uint32_t flags, int num)
+{
+ MDB_dpage *dp;
+
+ if ((dp = mdb_alloc_page(txn, dbi, NULL, 0, num)) == NULL)
+ return NULL;
+ DPRINTF("allocated new mpage %lu, page size %u",
+ dp->p.mp_pgno, txn->mt_env->me_psize);
+ dp->p.mp_flags = flags | P_DIRTY;
+ dp->p.mp_lower = PAGEHDRSZ;
+ dp->p.mp_upper = txn->mt_env->me_psize;
+
+ if (IS_BRANCH(&dp->p))
+ txn->mt_dbs[dbi].md_branch_pages++;
+ else if (IS_LEAF(&dp->p))
+ txn->mt_dbs[dbi].md_leaf_pages++;
+ else if (IS_OVERFLOW(&dp->p)) {
+ txn->mt_dbs[dbi].md_overflow_pages += num;
+ dp->p.mp_pages = num;
+ }
+
+ return dp;
+}
+
+static size_t
+mdb_leaf_size(MDB_env *env, MDB_val *key, MDB_val *data)
+{
+ size_t sz;
+
+ sz = LEAFSIZE(key, data);
+ if (data->mv_size >= env->me_psize / MDB_MINKEYS) {
+ /* put on overflow page */
+ sz -= data->mv_size - sizeof(pgno_t);
+ }
+
+ return sz + sizeof(indx_t);
+}
+
+static size_t
+mdb_branch_size(MDB_env *env, MDB_val *key)
+{
+ size_t sz;
+
+ sz = INDXSIZE(key);
+ if (sz >= env->me_psize / MDB_MINKEYS) {
+ /* put on overflow page */
+ /* not implemented */
+ /* sz -= key->size - sizeof(pgno_t); */
+ }
+
+ return sz + sizeof(indx_t);
+}
+
+static int
+mdb_add_node(MDB_txn *txn, MDB_dbi dbi, MDB_page *mp, indx_t indx,
+ MDB_val *key, MDB_val *data, pgno_t pgno, uint8_t flags)
+{
+ unsigned int i;
+ size_t node_size = NODESIZE;
+ indx_t ofs;
+ MDB_node *node;
+ MDB_dpage *ofp = NULL; /* overflow page */
+ DKBUF;
+
+ assert(mp->mp_upper >= mp->mp_lower);
+
+ DPRINTF("add to %s page %lu index %i, data size %zu key size %zu [%s]",
+ IS_LEAF(mp) ? "leaf" : "branch",
+ mp->mp_pgno, indx, data ? data->mv_size : 0,
+ key ? key->mv_size : 0, key ? DKEY(key) : NULL);
+
+ if (IS_LEAF2(mp)) {
+ /* Move higher keys up one slot. */
+ int ksize = txn->mt_dbs[dbi].md_pad, dif;
+ char *ptr = LEAF2KEY(mp, indx, ksize);
+ dif = NUMKEYS(mp) - indx;
+ if (dif > 0)
+ memmove(ptr+ksize, ptr, dif*ksize);
+ /* insert new key */
+ memcpy(ptr, key->mv_data, ksize);
+
+ /* Just using these for counting */
+ mp->mp_lower += sizeof(indx_t);
+ mp->mp_upper -= ksize - sizeof(indx_t);
+ return MDB_SUCCESS;
+ }
+
+ if (key != NULL)
+ node_size += key->mv_size;
+
+ if (IS_LEAF(mp)) {
+ assert(data);
+ if (F_ISSET(flags, F_BIGDATA)) {
+ /* Data already on overflow page. */
+ node_size += sizeof(pgno_t);
+ } else if (data->mv_size >= txn->mt_env->me_psize / MDB_MINKEYS) {
+ int ovpages = OVPAGES(data->mv_size, txn->mt_env->me_psize);
+ /* Put data on overflow page. */
+ DPRINTF("data size is %zu, put on overflow page",
+ data->mv_size);
+ node_size += sizeof(pgno_t);
+ if ((ofp = mdb_new_page(txn, dbi, P_OVERFLOW, ovpages)) == NULL)
+ return ENOMEM;
+ DPRINTF("allocated overflow page %lu", ofp->p.mp_pgno);
+ flags |= F_BIGDATA;
+ } else {
+ node_size += data->mv_size;
+ }
+ }
+
+ if (node_size + sizeof(indx_t) > SIZELEFT(mp)) {
+ DPRINTF("not enough room in page %lu, got %u ptrs",
+ mp->mp_pgno, NUMKEYS(mp));
+ DPRINTF("upper - lower = %u - %u = %u", mp->mp_upper, mp->mp_lower,
+ mp->mp_upper - mp->mp_lower);
+ DPRINTF("node size = %zu", node_size);
+ return ENOSPC;
+ }
+
+ /* Move higher pointers up one slot. */
+ for (i = NUMKEYS(mp); i > indx; i--)
+ mp->mp_ptrs[i] = mp->mp_ptrs[i - 1];
+
+ /* Adjust free space offsets. */
+ ofs = mp->mp_upper - node_size;
+ assert(ofs >= mp->mp_lower + sizeof(indx_t));
+ mp->mp_ptrs[indx] = ofs;
+ mp->mp_upper = ofs;
+ mp->mp_lower += sizeof(indx_t);
+
+ /* Write the node data. */
+ node = NODEPTR(mp, indx);
+ node->mn_ksize = (key == NULL) ? 0 : key->mv_size;
+ node->mn_flags = flags;
+ if (IS_LEAF(mp))
+ node->mn_dsize = data->mv_size;
+ else
+ NODEPGNO(node) = pgno;
+
+ if (key)
+ memcpy(NODEKEY(node), key->mv_data, key->mv_size);
+
+ if (IS_LEAF(mp)) {
+ assert(key);
+ if (ofp == NULL) {
+ if (F_ISSET(flags, F_BIGDATA))
+ memcpy(node->mn_data + key->mv_size, data->mv_data,
+ sizeof(pgno_t));
+ else
+ memcpy(node->mn_data + key->mv_size, data->mv_data,
+ data->mv_size);
+ } else {
+ memcpy(node->mn_data + key->mv_size, &ofp->p.mp_pgno,
+ sizeof(pgno_t));
+ memcpy(METADATA(&ofp->p), data->mv_data, data->mv_size);
+ }
+ }
+
+ return MDB_SUCCESS;
+}
+
+static void
+mdb_del_node(MDB_page *mp, indx_t indx, int ksize)
+{
+ unsigned int sz;
+ indx_t i, j, numkeys, ptr;
+ MDB_node *node;
+ char *base;
+
+ DPRINTF("delete node %u on %s page %lu", indx,
+ IS_LEAF(mp) ? "leaf" : "branch", mp->mp_pgno);
+ assert(indx < NUMKEYS(mp));
+
+ if (IS_LEAF2(mp)) {
+ int x = NUMKEYS(mp) - 1 - indx;
+ base = LEAF2KEY(mp, indx, ksize);
+ if (x)
+ memmove(base, base + ksize, x * ksize);
+ mp->mp_lower -= sizeof(indx_t);
+ mp->mp_upper += ksize - sizeof(indx_t);
+ return;
+ }
+
+ node = NODEPTR(mp, indx);
+ sz = NODESIZE + node->mn_ksize;
+ if (IS_LEAF(mp)) {
+ if (F_ISSET(node->mn_flags, F_BIGDATA))
+ sz += sizeof(pgno_t);
+ else
+ sz += NODEDSZ(node);
+ }
+
+ ptr = mp->mp_ptrs[indx];
+ numkeys = NUMKEYS(mp);
+ for (i = j = 0; i < numkeys; i++) {
+ if (i != indx) {
+ mp->mp_ptrs[j] = mp->mp_ptrs[i];
+ if (mp->mp_ptrs[i] < ptr)
+ mp->mp_ptrs[j] += sz;
+ j++;
+ }
+ }
+
+ base = (char *)mp + mp->mp_upper;
+ memmove(base + sz, base, ptr - mp->mp_upper);
+
+ mp->mp_lower -= sizeof(indx_t);
+ mp->mp_upper += sz;
+}
+
+static void
+mdb_xcursor_init0(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx)
+{
+ MDB_dbi dbn;
+
+ mx->mx_txn = *txn;
+ mx->mx_txn.mt_dbxs = mx->mx_dbxs;
+ mx->mx_txn.mt_dbs = mx->mx_dbs;
+ mx->mx_dbxs[0] = txn->mt_dbxs[0];
+ mx->mx_dbxs[1] = txn->mt_dbxs[1];
+ if (dbi > 1) {
+ mx->mx_dbxs[2] = txn->mt_dbxs[dbi];
+ dbn = 2;
+ } else {
+ dbn = 1;
+ }
+ mx->mx_dbxs[dbn+1].md_parent = dbn;
+ mx->mx_dbxs[dbn+1].md_cmp = mx->mx_dbxs[dbn].md_dcmp;
+ mx->mx_dbxs[dbn+1].md_rel = mx->mx_dbxs[dbn].md_rel;
+ mx->mx_dbxs[dbn+1].md_dirty = 0;
+ mx->mx_txn.mt_numdbs = dbn+2;
+
+ mx->mx_cursor.mc_snum = 0;
+ mx->mx_cursor.mc_txn = &mx->mx_txn;
+ mx->mx_cursor.mc_dbi = dbn+1;
+}
+
+static void
+mdb_xcursor_init1(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx, MDB_page *mp, MDB_node *node)
+{
+ MDB_db *db = NODEDATA(node);
+ MDB_dbi dbn;
+ mx->mx_dbs[0] = txn->mt_dbs[0];
+ mx->mx_dbs[1] = txn->mt_dbs[1];
+ if (dbi > 1) {
+ mx->mx_dbs[2] = txn->mt_dbs[dbi];
+ dbn = 3;
+ } else {
+ dbn = 2;
+ }
+ DPRINTF("Sub-db %u for db %u root page %lu", dbn, dbi, db->md_root);
+ mx->mx_dbs[dbn] = *db;
+ if (F_ISSET(mp->mp_flags, P_DIRTY))
+ mx->mx_dbxs[dbn].md_dirty = 1;
+ mx->mx_dbxs[dbn].md_name.mv_data = NODEKEY(node);
+ mx->mx_dbxs[dbn].md_name.mv_size = node->mn_ksize;
+ mx->mx_txn.mt_next_pgno = txn->mt_next_pgno;
+ mx->mx_txn.mt_u = txn->mt_u;
+ mx->mx_cursor.mc_initialized = 0;
+ mx->mx_cursor.mc_eof = 0;
+}
+
+static void
+mdb_xcursor_fini(MDB_txn *txn, MDB_dbi dbi, MDB_xcursor *mx)
+{
+ txn->mt_next_pgno = mx->mx_txn.mt_next_pgno;
+ txn->mt_u = mx->mx_txn.mt_u;
+ txn->mt_dbs[0] = mx->mx_dbs[0];
+ txn->mt_dbs[1] = mx->mx_dbs[1];
+ txn->mt_dbxs[0].md_dirty = mx->mx_dbxs[0].md_dirty;
+ txn->mt_dbxs[1].md_dirty = mx->mx_dbxs[1].md_dirty;
+ if (dbi > 1) {
+ txn->mt_dbs[dbi] = mx->mx_dbs[2];
+ txn->mt_dbxs[dbi].md_dirty = mx->mx_dbxs[2].md_dirty;
+ }
+}
+
+int
+mdb_cursor_open(MDB_txn *txn, MDB_dbi dbi, MDB_cursor **ret)
+{
+ MDB_cursor *cursor;
+ size_t size = sizeof(MDB_cursor);
+
+ if (txn == NULL || ret == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ if (txn->mt_dbs[dbi].md_flags & MDB_DUPSORT)
+ size += sizeof(MDB_xcursor);
+
+ if ((cursor = calloc(1, size)) != NULL) {
+ cursor->mc_dbi = dbi;
+ cursor->mc_txn = txn;
+ if (txn->mt_dbs[dbi].md_flags & MDB_DUPSORT) {
+ MDB_xcursor *mx = (MDB_xcursor *)(cursor + 1);
+ cursor->mc_xcursor = mx;
+ mdb_xcursor_init0(txn, dbi, mx);
+ }
+ } else {
+ return ENOMEM;
+ }
+
+ *ret = cursor;
+
+ return MDB_SUCCESS;
+}
+
+/* Return the count of duplicate data items for the current key */
+int
+mdb_cursor_count(MDB_cursor *mc, unsigned long *countp)
+{
+ MDB_ppage *top;
+ MDB_node *leaf;
+
+ if (mc == NULL || countp == NULL)
+ return EINVAL;
+
+ if (!(mc->mc_txn->mt_dbs[mc->mc_dbi].md_flags & MDB_DUPSORT))
+ return EINVAL;
+
+ top = CURSOR_TOP(mc);
+ leaf = NODEPTR(top->mp_page, top->mp_ki);
+ if (!F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ *countp = 1;
+ } else {
+ if (!mc->mc_xcursor->mx_cursor.mc_initialized)
+ return EINVAL;
+
+ *countp = mc->mc_xcursor->mx_txn.mt_dbs[mc->mc_xcursor->mx_cursor.mc_dbi].md_entries;
+ }
+ return MDB_SUCCESS;
+}
+
+void
+mdb_cursor_close(MDB_cursor *cursor)
+{
+ if (cursor != NULL) {
+ free(cursor);
+ }
+}
+
+static int
+mdb_update_key(MDB_page *mp, indx_t indx, MDB_val *key)
+{
+ indx_t ptr, i, numkeys;
+ int delta;
+ size_t len;
+ MDB_node *node;
+ char *base;
+ DKBUF;
+
+ node = NODEPTR(mp, indx);
+ ptr = mp->mp_ptrs[indx];
+ DPRINTF("update key %u (ofs %u) [%.*s] to [%s] on page %lu",
+ indx, ptr,
+ (int)node->mn_ksize, (char *)NODEKEY(node),
+ DKEY(key),
+ mp->mp_pgno);
+
+ delta = key->mv_size - node->mn_ksize;
+ if (delta) {
+ if (delta > 0 && SIZELEFT(mp) < delta) {
+ DPRINTF("OUCH! Not enough room, delta = %d", delta);
+ return ENOSPC;
+ }
+
+ numkeys = NUMKEYS(mp);
+ for (i = 0; i < numkeys; i++) {
+ if (mp->mp_ptrs[i] <= ptr)
+ mp->mp_ptrs[i] -= delta;
+ }
+
+ base = (char *)mp + mp->mp_upper;
+ len = ptr - mp->mp_upper + NODESIZE;
+ memmove(base - delta, base, len);
+ mp->mp_upper -= delta;
+
+ node = NODEPTR(mp, indx);
+ node->mn_ksize = key->mv_size;
+ }
+
+ memcpy(NODEKEY(node), key->mv_data, key->mv_size);
+
+ return MDB_SUCCESS;
+}
+
+/* Move a node from src to dst.
+ */
+static int
+mdb_move_node(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *src, indx_t srcindx,
+ MDB_pageparent *dst, indx_t dstindx)
+{
+ int rc;
+ MDB_node *srcnode;
+ MDB_val key, data;
+ DKBUF;
+
+ /* Mark src and dst as dirty. */
+ if ((rc = mdb_touch(txn, dbi, src)) ||
+ (rc = mdb_touch(txn, dbi, dst)))
+ return rc;;
+
+ if (IS_LEAF2(src->mp_page)) {
+ srcnode = NODEPTR(src->mp_page, 0); /* fake */
+ key.mv_size = txn->mt_dbs[dbi].md_pad;
+ key.mv_data = LEAF2KEY(src->mp_page, srcindx, key.mv_size);
+ data.mv_size = 0;
+ data.mv_data = NULL;
+ } else {
+ srcnode = NODEPTR(src->mp_page, srcindx);
+ key.mv_size = NODEKSZ(srcnode);
+ key.mv_data = NODEKEY(srcnode);
+ data.mv_size = NODEDSZ(srcnode);
+ data.mv_data = NODEDATA(srcnode);
+ }
+ DPRINTF("moving %s node %u [%s] on page %lu to node %u on page %lu",
+ IS_LEAF(src->mp_page) ? "leaf" : "branch",
+ srcindx,
+ DKEY(&key),
+ src->mp_page->mp_pgno,
+ dstindx, dst->mp_page->mp_pgno);
+
+ /* Add the node to the destination page.
+ */
+ rc = mdb_add_node(txn, dbi, dst->mp_page, dstindx, &key, &data, NODEPGNO(srcnode),
+ srcnode->mn_flags);
+ if (rc != MDB_SUCCESS)
+ return rc;
+
+ /* Delete the node from the source page.
+ */
+ mdb_del_node(src->mp_page, srcindx, key.mv_size);
+
+ /* The key value just changed due to del_node, find it again.
+ */
+ if (!IS_LEAF2(src->mp_page)) {
+ srcnode = NODEPTR(src->mp_page, srcindx);
+ key.mv_data = NODEKEY(srcnode);
+ }
+
+ /* Update the parent separators.
+ */
+ if (srcindx == 0) {
+ if (src->mp_pi != 0) {
+ DPRINTF("update separator for source page %lu to [%s]",
+ src->mp_page->mp_pgno, DKEY(&key));
+ if ((rc = mdb_update_key(src->mp_parent, src->mp_pi,
+ &key)) != MDB_SUCCESS)
+ return rc;
+ }
+ if (IS_BRANCH(src->mp_page)) {
+ MDB_val nullkey;
+ nullkey.mv_size = 0;
+ assert(mdb_update_key(src->mp_page, 0, &nullkey) == MDB_SUCCESS);
+ }
+ }
+
+ if (dstindx == 0) {
+ if (dst->mp_pi != 0) {
+ DPRINTF("update separator for destination page %lu to [%s]",
+ dst->mp_page->mp_pgno, DKEY(&key));
+ if ((rc = mdb_update_key(dst->mp_parent, dst->mp_pi,
+ &key)) != MDB_SUCCESS)
+ return rc;
+ }
+ if (IS_BRANCH(dst->mp_page)) {
+ MDB_val nullkey;
+ nullkey.mv_size = 0;
+ assert(mdb_update_key(dst->mp_page, 0, &nullkey) == MDB_SUCCESS);
+ }
+ }
+
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_merge(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *src, MDB_pageparent *dst)
+{
+ int rc;
+ indx_t i;
+ MDB_node *srcnode;
+ MDB_val key, data;
+ MDB_pageparent mpp;
+ MDB_dhead *dh;
+
+ DPRINTF("merging page %lu and %lu", src->mp_page->mp_pgno, dst->mp_page->mp_pgno);
+
+ assert(txn != NULL);
+ assert(src->mp_parent); /* can't merge root page */
+ assert(dst->mp_parent);
+
+ /* Mark src and dst as dirty. */
+ if ((rc = mdb_touch(txn, dbi, src)) ||
+ (rc = mdb_touch(txn, dbi, dst)))
+ return rc;
+
+ /* Move all nodes from src to dst.
+ */
+ if (IS_LEAF2(src->mp_page)) {
+ key.mv_size = txn->mt_dbs[dbi].md_pad;
+ key.mv_data = METADATA(src->mp_page);
+ for (i = 0; i < NUMKEYS(src->mp_page); i++) {
+ rc = mdb_add_node(txn, dbi, dst->mp_page, NUMKEYS(dst->mp_page), &key,
+ NULL, 0, 0);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ key.mv_data = (char *)key.mv_data + key.mv_size;
+ }
+ } else {
+ for (i = 0; i < NUMKEYS(src->mp_page); i++) {
+ srcnode = NODEPTR(src->mp_page, i);
+
+ key.mv_size = srcnode->mn_ksize;
+ key.mv_data = NODEKEY(srcnode);
+ data.mv_size = NODEDSZ(srcnode);
+ data.mv_data = NODEDATA(srcnode);
+ rc = mdb_add_node(txn, dbi, dst->mp_page, NUMKEYS(dst->mp_page), &key,
+ &data, NODEPGNO(srcnode), srcnode->mn_flags);
+ if (rc != MDB_SUCCESS)
+ return rc;
+ }
+ }
+
+ DPRINTF("dst page %lu now has %u keys (%.1f%% filled)",
+ dst->mp_page->mp_pgno, NUMKEYS(dst->mp_page), (float)PAGEFILL(txn->mt_env, dst->mp_page) / 10);
+
+ /* Unlink the src page from parent.
+ */
+ mdb_del_node(src->mp_parent, src->mp_pi, 0);
+ if (src->mp_pi == 0) {
+ key.mv_size = 0;
+ if ((rc = mdb_update_key(src->mp_parent, 0, &key)) != MDB_SUCCESS)
+ return rc;
+ }
+
+ if (IS_LEAF(src->mp_page))
+ txn->mt_dbs[dbi].md_leaf_pages--;
+ else
+ txn->mt_dbs[dbi].md_branch_pages--;
+
+ mpp.mp_page = src->mp_parent;
+ dh = (MDB_dhead *)src->mp_parent;
+ dh--;
+ mpp.mp_parent = dh->md_parent;
+ mpp.mp_pi = dh->md_pi;
+
+ return mdb_rebalance(txn, dbi, &mpp);
+}
+
+#define FILL_THRESHOLD 250
+
+static int
+mdb_rebalance(MDB_txn *txn, MDB_dbi dbi, MDB_pageparent *mpp)
+{
+ MDB_node *node;
+ MDB_page *root;
+ MDB_pageparent npp;
+ indx_t si = 0, di = 0;
+ int rc;
+
+ assert(txn != NULL);
+ assert(mpp != NULL);
+
+ DPRINTF("rebalancing %s page %lu (has %u keys, %.1f%% full)",
+ IS_LEAF(mpp->mp_page) ? "leaf" : "branch",
+ mpp->mp_page->mp_pgno, NUMKEYS(mpp->mp_page), (float)PAGEFILL(txn->mt_env, mpp->mp_page) / 10);
+
+ if (PAGEFILL(txn->mt_env, mpp->mp_page) >= FILL_THRESHOLD) {
+ DPRINTF("no need to rebalance page %lu, above fill threshold",
+ mpp->mp_page->mp_pgno);
+ return MDB_SUCCESS;
+ }
+
+ if (mpp->mp_parent == NULL) {
+ if (NUMKEYS(mpp->mp_page) == 0) {
+ DPUTS("tree is completely empty");
+ txn->mt_dbs[dbi].md_root = P_INVALID;
+ txn->mt_dbs[dbi].md_depth = 0;
+ txn->mt_dbs[dbi].md_leaf_pages = 0;
+ } else if (IS_BRANCH(mpp->mp_page) && NUMKEYS(mpp->mp_page) == 1) {
+ DPUTS("collapsing root page!");
+ txn->mt_dbs[dbi].md_root = NODEPGNO(NODEPTR(mpp->mp_page, 0));
+ if ((rc = mdb_get_page(txn, txn->mt_dbs[dbi].md_root, &root)))
+ return rc;
+ txn->mt_dbs[dbi].md_depth--;
+ txn->mt_dbs[dbi].md_branch_pages--;
+ } else
+ DPUTS("root page doesn't need rebalancing");
+ return MDB_SUCCESS;
+ }
+
+ /* The parent (branch page) must have at least 2 pointers,
+ * otherwise the tree is invalid.
+ */
+ assert(NUMKEYS(mpp->mp_parent) > 1);
+
+ /* Leaf page fill factor is below the threshold.
+ * Try to move keys from left or right neighbor, or
+ * merge with a neighbor page.
+ */
+
+ /* Find neighbors.
+ */
+ if (mpp->mp_pi == 0) {
+ /* We're the leftmost leaf in our parent.
+ */
+ DPUTS("reading right neighbor");
+ node = NODEPTR(mpp->mp_parent, mpp->mp_pi + 1);
+ if ((rc = mdb_get_page(txn, NODEPGNO(node), &npp.mp_page)))
+ return rc;
+ npp.mp_pi = mpp->mp_pi + 1;
+ si = 0;
+ di = NUMKEYS(mpp->mp_page);
+ } else {
+ /* There is at least one neighbor to the left.
+ */
+ DPUTS("reading left neighbor");
+ node = NODEPTR(mpp->mp_parent, mpp->mp_pi - 1);
+ if ((rc = mdb_get_page(txn, NODEPGNO(node), &npp.mp_page)))
+ return rc;
+ npp.mp_pi = mpp->mp_pi - 1;
+ si = NUMKEYS(npp.mp_page) - 1;
+ di = 0;
+ }
+ npp.mp_parent = mpp->mp_parent;
+
+ DPRINTF("found neighbor page %lu (%u keys, %.1f%% full)",
+ npp.mp_page->mp_pgno, NUMKEYS(npp.mp_page), (float)PAGEFILL(txn->mt_env, npp.mp_page) / 10);
+
+ /* If the neighbor page is above threshold and has at least two
+ * keys, move one key from it.
+ *
+ * Otherwise we should try to merge them.
+ */
+ if (PAGEFILL(txn->mt_env, npp.mp_page) >= FILL_THRESHOLD && NUMKEYS(npp.mp_page) >= 2)
+ return mdb_move_node(txn, dbi, &npp, si, mpp, di);
+ else { /* FIXME: if (has_enough_room()) */
+ if (mpp->mp_pi == 0)
+ return mdb_merge(txn, dbi, &npp, mpp);
+ else
+ return mdb_merge(txn, dbi, mpp, &npp);
+ }
+}
+
+static int
+mdb_del0(MDB_txn *txn, MDB_dbi dbi, unsigned int ki, MDB_pageparent *mpp, MDB_node *leaf)
+{
+ int rc;
+
+ /* add overflow pages to free list */
+ if (!IS_LEAF2(mpp->mp_page) && F_ISSET(leaf->mn_flags, F_BIGDATA)) {
+ int i, ovpages;
+ pgno_t pg;
+
+ memcpy(&pg, NODEDATA(leaf), sizeof(pg));
+ ovpages = OVPAGES(NODEDSZ(leaf), txn->mt_env->me_psize);
+ for (i=0; i<ovpages; i++) {
+ DPRINTF("freed ov page %lu", pg);
+ mdb_midl_insert(txn->mt_free_pgs, pg);
+ pg++;
+ }
+ }
+ mdb_del_node(mpp->mp_page, ki, txn->mt_dbs[dbi].md_pad);
+ txn->mt_dbs[dbi].md_entries--;
+ rc = mdb_rebalance(txn, dbi, mpp);
+ if (rc != MDB_SUCCESS)
+ txn->mt_flags |= MDB_TXN_ERROR;
+
+ return rc;
+}
+
+int
+mdb_del(MDB_txn *txn, MDB_dbi dbi,
+ MDB_val *key, MDB_val *data)
+{
+ int rc, exact;
+ unsigned int ki;
+ MDB_node *leaf;
+ MDB_pageparent mpp;
+ DKBUF;
+
+ assert(key != NULL);
+
+ DPRINTF("====> delete db %u key [%s]", dbi, DKEY(key));
+
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ if (F_ISSET(txn->mt_flags, MDB_TXN_RDONLY)) {
+ return EACCES;
+ }
+
+ if (key->mv_size == 0 || key->mv_size > MAXKEYSIZE) {
+ return EINVAL;
+ }
+
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ if ((rc = mdb_search_page(txn, dbi, key, NULL, 1, &mpp)) != MDB_SUCCESS)
+ return rc;
+
+ leaf = mdb_search_node(txn, dbi, mpp.mp_page, key, &exact, &ki);
+ if (leaf == NULL || !exact) {
+ return MDB_NOTFOUND;
+ }
+
+ if (!IS_LEAF2(mpp.mp_page) && F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ MDB_xcursor mx;
+ MDB_pageparent mp2;
+
+ mdb_xcursor_init0(txn, dbi, &mx);
+ mdb_xcursor_init1(txn, dbi, &mx, mpp.mp_page, leaf);
+ if (data) {
+ rc = mdb_del(&mx.mx_txn, mx.mx_cursor.mc_dbi, data, NULL);
+ mdb_xcursor_fini(txn, dbi, &mx);
+ /* If sub-DB still has entries, we're done */
+ if (mx.mx_txn.mt_dbs[mx.mx_cursor.mc_dbi].md_root != P_INVALID) {
+ memcpy(NODEDATA(leaf), &mx.mx_txn.mt_dbs[mx.mx_cursor.mc_dbi],
+ sizeof(MDB_db));
+ txn->mt_dbs[dbi].md_entries--;
+ return rc;
+ }
+ /* otherwise fall thru and delete the sub-DB */
+ } else {
+ /* add all the child DB's pages to the free list */
+ rc = mdb_search_page(&mx.mx_txn, mx.mx_cursor.mc_dbi,
+ NULL, &mx.mx_cursor, 0, &mp2);
+ if (rc == MDB_SUCCESS) {
+ MDB_ppage *top, *parent;
+ MDB_node *ni;
+ unsigned int i;
+
+ cursor_pop_page(&mx.mx_cursor);
+ if (mx.mx_cursor.mc_snum) {
+ top = CURSOR_TOP(&mx.mx_cursor);
+ while (mx.mx_cursor.mc_snum > 1) {
+ parent = CURSOR_PARENT(&mx.mx_cursor);
+ for (i=0; i<NUMKEYS(top->mp_page); i++) {
+ ni = NODEPTR(top->mp_page, i);
+ mdb_midl_insert(txn->mt_free_pgs, NODEPGNO(ni));
+ }
+ parent->mp_ki++;
+ if (parent->mp_ki >= NUMKEYS(parent->mp_page)) {
+ cursor_pop_page(&mx.mx_cursor);
+ top = parent;
+ } else {
+ ni = NODEPTR(parent->mp_page, parent->mp_ki);
+ rc = mdb_get_page(&mx.mx_txn, NODEPGNO(ni), &top->mp_page);
+ }
+ }
+ }
+ mdb_midl_insert(txn->mt_free_pgs, mx.mx_txn.mt_dbs[mx.mx_cursor.mc_dbi].md_root);
+ }
+ }
+ }
+
+ return mdb_del0(txn, dbi, ki, &mpp, leaf);
+}
+
+/* Split page <*mpp>, and insert <key,(data|newpgno)> in either left or
+ * right sibling, at index <*newindxp> (as if unsplit). Updates *mpp and
+ * *newindxp with the actual values after split, ie if *mpp and *newindxp
+ * refer to a node in the new right sibling page.
+ */
+static int
+mdb_split(MDB_txn *txn, MDB_dbi dbi, MDB_page **mpp, unsigned int *newindxp,
+ MDB_val *newkey, MDB_val *newdata, pgno_t newpgno)
+{
+ uint8_t flags;
+ int rc = MDB_SUCCESS, ins_new = 0;
+ indx_t newindx;
+ pgno_t pgno = 0;
+ unsigned int i, j, split_indx, nkeys, pmax;
+ MDB_node *node;
+ MDB_val sepkey, rkey, rdata;
+ MDB_page *copy, *cptr;
+ MDB_dpage *mdp, *rdp, *pdp;
+ MDB_dhead *dh;
+ DKBUF;
+
+ assert(txn != NULL);
+
+ dh = ((MDB_dhead *)*mpp) - 1;
+ mdp = (MDB_dpage *)dh;
+ newindx = *newindxp;
+
+ DPRINTF("-----> splitting %s page %lu and adding [%s] at index %i",
+ IS_LEAF(&mdp->p) ? "leaf" : "branch", mdp->p.mp_pgno,
+ DKEY(newkey), *newindxp);
+
+ if (mdp->h.md_parent == NULL) {
+ if ((pdp = mdb_new_page(txn, dbi, P_BRANCH, 1)) == NULL)
+ return ENOMEM;
+ mdp->h.md_pi = 0;
+ mdp->h.md_parent = &pdp->p;
+ txn->mt_dbs[dbi].md_root = pdp->p.mp_pgno;
+ DPRINTF("root split! new root = %lu", pdp->p.mp_pgno);
+ txn->mt_dbs[dbi].md_depth++;
+
+ /* Add left (implicit) pointer. */
+ if ((rc = mdb_add_node(txn, dbi, &pdp->p, 0, NULL, NULL,
+ mdp->p.mp_pgno, 0)) != MDB_SUCCESS)
+ return rc;
+ } else {
+ DPRINTF("parent branch page is %lu", mdp->h.md_parent->mp_pgno);
+ }
+
+ /* Create a right sibling. */
+ if ((rdp = mdb_new_page(txn, dbi, mdp->p.mp_flags, 1)) == NULL)
+ return ENOMEM;
+ rdp->h.md_parent = mdp->h.md_parent;
+ rdp->h.md_pi = mdp->h.md_pi + 1;
+ DPRINTF("new right sibling: page %lu", rdp->p.mp_pgno);
+
+ nkeys = NUMKEYS(&mdp->p);
+ split_indx = nkeys / 2 + 1;
+
+ if (IS_LEAF2(&rdp->p)) {
+ char *split, *ins;
+ int x;
+ unsigned int lsize, rsize, ksize;
+ /* Move half of the keys to the right sibling */
+ copy = NULL;
+ x = *newindxp - split_indx;
+ ksize = txn->mt_dbs[dbi].md_pad;
+ split = LEAF2KEY(&mdp->p, split_indx, ksize);
+ rsize = (nkeys - split_indx) * ksize;
+ lsize = (nkeys - split_indx) * sizeof(indx_t);
+ mdp->p.mp_lower -= lsize;
+ rdp->p.mp_lower += lsize;
+ mdp->p.mp_upper += rsize - lsize;
+ rdp->p.mp_upper -= rsize - lsize;
+ sepkey.mv_size = ksize;
+ if (newindx == split_indx) {
+ sepkey.mv_data = newkey->mv_data;
+ } else {
+ sepkey.mv_data = split;
+ }
+ if (x<0) {
+ ins = LEAF2KEY(&mdp->p, *newindxp, ksize);
+ memcpy(&rdp->p.mp_ptrs, split, rsize);
+ sepkey.mv_data = &rdp->p.mp_ptrs;
+ memmove(ins+ksize, ins, (split_indx - *newindxp) * ksize);
+ memcpy(ins, newkey->mv_data, ksize);
+ mdp->p.mp_lower += sizeof(indx_t);
+ mdp->p.mp_upper -= ksize - sizeof(indx_t);
+ } else {
+ if (x)
+ memcpy(&rdp->p.mp_ptrs, split, x * ksize);
+ ins = LEAF2KEY(&rdp->p, x, ksize);
+ memcpy(ins, newkey->mv_data, ksize);
+ memcpy(ins+ksize, split + x * ksize, rsize - x * ksize);
+ rdp->p.mp_lower += sizeof(indx_t);
+ rdp->p.mp_upper -= ksize - sizeof(indx_t);
+ *newindxp = x;
+ *mpp = &rdp->p;
+ }
+ goto newsep;
+ }
+
+ /* For leaf pages, check the split point based on what
+ * fits where, since otherwise add_node can fail.
+ */
+ if (IS_LEAF(&mdp->p)) {
+ unsigned int psize, nsize;
+ /* Maximum free space in an empty page */
+ pmax = txn->mt_env->me_psize - PAGEHDRSZ;
+ nsize = mdb_leaf_size(txn->mt_env, newkey, newdata);
+ if (newindx <= split_indx) {
+split1:
+ psize = nsize;
+ for (i=0; i<split_indx; i++) {
+ node = NODEPTR(&mdp->p, i);
+ psize += NODESIZE + NODEKSZ(node);
+ if (F_ISSET(node->mn_flags, F_BIGDATA))
+ psize += sizeof(pgno_t);
+ else
+ psize += NODEDSZ(node);
+ if (psize > pmax) {
+ split_indx--;
+ goto split1;
+ }
+ }
+ } else {
+split2:
+ psize = nsize;
+ for (i=split_indx; i<nkeys; i++) {
+ node = NODEPTR(&mdp->p, i);
+ psize += NODESIZE + NODEKSZ(node);
+ if (F_ISSET(node->mn_flags, F_BIGDATA))
+ psize += sizeof(pgno_t);
+ else
+ psize += NODEDSZ(node);
+ if (psize > pmax) {
+ split_indx++;
+ goto split2;
+ }
+ }
+ }
+ }
+
+ /* First find the separating key between the split pages.
+ */
+ memset(&sepkey, 0, sizeof(sepkey));
+ if (newindx == split_indx) {
+ sepkey.mv_size = newkey->mv_size;
+ sepkey.mv_data = newkey->mv_data;
+ } else {
+ node = NODEPTR(&mdp->p, split_indx);
+ sepkey.mv_size = node->mn_ksize;
+ sepkey.mv_data = NODEKEY(node);
+ }
+
+newsep:
+ DPRINTF("separator is [%s]", DKEY(&sepkey));
+
+ /* Copy separator key to the parent.
+ */
+ if (SIZELEFT(rdp->h.md_parent) < mdb_branch_size(txn->mt_env, &sepkey)) {
+ rc = mdb_split(txn, dbi, &rdp->h.md_parent, &rdp->h.md_pi,
+ &sepkey, NULL, rdp->p.mp_pgno);
+
+ /* Right page might now have changed parent.
+ * Check if left page also changed parent.
+ */
+ if (rdp->h.md_parent != mdp->h.md_parent &&
+ mdp->h.md_pi >= NUMKEYS(mdp->h.md_parent)) {
+ mdp->h.md_parent = rdp->h.md_parent;
+ mdp->h.md_pi = rdp->h.md_pi - 1;
+ }
+ } else {
+ rc = mdb_add_node(txn, dbi, rdp->h.md_parent, rdp->h.md_pi,
+ &sepkey, NULL, rdp->p.mp_pgno, 0);
+ }
+ if (IS_LEAF2(&rdp->p)) {
+ return rc;
+ }
+ if (rc != MDB_SUCCESS) {
+ return rc;
+ }
+
+ /* Move half of the keys to the right sibling. */
+ if ((copy = malloc(txn->mt_env->me_psize)) == NULL)
+ return ENOMEM;
+
+ copy->mp_pgno = mdp->p.mp_pgno;
+ copy->mp_flags = mdp->p.mp_flags;
+ copy->mp_lower = PAGEHDRSZ;
+ copy->mp_upper = txn->mt_env->me_psize;
+ cptr = copy;
+ for (i = j = 0; i <= nkeys; j++) {
+ if (i == split_indx) {
+ /* Insert in right sibling. */
+ /* Reset insert index for right sibling. */
+ j = (i == newindx && ins_new);
+ cptr = &rdp->p;
+ }
+
+ if (i == newindx && !ins_new) {
+ /* Insert the original entry that caused the split. */
+ rkey.mv_data = newkey->mv_data;
+ rkey.mv_size = newkey->mv_size;
+ if (IS_LEAF(&mdp->p)) {
+ rdata.mv_data = newdata->mv_data;
+ rdata.mv_size = newdata->mv_size;
+ } else
+ pgno = newpgno;
+ flags = 0;
+
+ ins_new = 1;
+
+ /* Update page and index for the new key. */
+ *newindxp = j;
+ if (cptr == &rdp->p)
+ *mpp = cptr;
+ } else if (i == nkeys) {
+ break;
+ } else {
+ node = NODEPTR(&mdp->p, i);
+ rkey.mv_data = NODEKEY(node);
+ rkey.mv_size = node->mn_ksize;
+ if (IS_LEAF(&mdp->p)) {
+ rdata.mv_data = NODEDATA(node);
+ rdata.mv_size = node->mn_dsize;
+ } else
+ pgno = NODEPGNO(node);
+ flags = node->mn_flags;
+
+ i++;
+ }
+
+ if (!IS_LEAF(&mdp->p) && j == 0) {
+ /* First branch index doesn't need key data. */
+ rkey.mv_size = 0;
+ }
+
+ rc = mdb_add_node(txn, dbi, cptr, j, &rkey, &rdata, pgno, flags);
+ }
+ nkeys = NUMKEYS(copy);
+ for (i=0; i<nkeys; i++)
+ mdp->p.mp_ptrs[i] = copy->mp_ptrs[i];
+ mdp->p.mp_lower = copy->mp_lower;
+ mdp->p.mp_upper = copy->mp_upper;
+ memcpy(NODEPTR(&mdp->p, nkeys-1), NODEPTR(copy, nkeys-1),
+ txn->mt_env->me_psize - copy->mp_upper);
+
+ free(copy);
+ return rc;
+}
+
+static int
+mdb_put0(MDB_txn *txn, MDB_dbi dbi,
+ MDB_val *key, MDB_val *data, unsigned int flags)
+{
+ int rc = MDB_SUCCESS, exact;
+ unsigned int ki;
+ MDB_node *leaf;
+ MDB_pageparent mpp;
+ MDB_val xdata, *rdata, dkey;
+ MDB_db dummy;
+ char dbuf[PAGESIZE];
+ int do_sub = 0;
+ size_t nsize;
+ DKBUF;
+
+ DPRINTF("==> put db %u key [%s], size %zu, data size %zu",
+ dbi, DKEY(key), key->mv_size, data->mv_size);
+
+ dkey.mv_size = 0;
+ mpp.mp_parent = NULL;
+ mpp.mp_pi = 0;
+ rc = mdb_search_page(txn, dbi, key, NULL, 1, &mpp);
+ if (rc == MDB_SUCCESS) {
+ leaf = mdb_search_node(txn, dbi, mpp.mp_page, key, &exact, &ki);
+ if (leaf && exact) {
+ if (flags == MDB_NOOVERWRITE) {
+ DPRINTF("duplicate key [%s]", DKEY(key));
+ return MDB_KEYEXIST;
+ }
+ /* there's only a key anyway, so this is a no-op */
+ if (IS_LEAF2(mpp.mp_page))
+ return MDB_SUCCESS;
+
+ if (F_ISSET(txn->mt_dbs[dbi].md_flags, MDB_DUPSORT)) {
+ /* Was a single item before, must convert now */
+ if (!F_ISSET(leaf->mn_flags, F_DUPDATA)) {
+ dkey.mv_size = NODEDSZ(leaf);
+ dkey.mv_data = dbuf;
+ memcpy(dbuf, NODEDATA(leaf), dkey.mv_size);
+ /* data matches, ignore it */
+ if (!mdb_dcmp(txn, dbi, data, &dkey))
+ return (flags == MDB_NODUPDATA) ? MDB_KEYEXIST : MDB_SUCCESS;
+ memset(&dummy, 0, sizeof(dummy));
+ if (txn->mt_dbs[dbi].md_flags & MDB_DUPFIXED) {
+ dummy.md_pad = data->mv_size;
+ dummy.md_flags = MDB_DUPFIXED;
+ if (txn->mt_dbs[dbi].md_flags & MDB_INTEGERDUP)
+ dummy.md_flags |= MDB_INTEGERKEY;
+ }
+ dummy.md_root = P_INVALID;
+ if (dkey.mv_size == sizeof(MDB_db)) {
+ memcpy(NODEDATA(leaf), &dummy, sizeof(dummy));
+ goto put_sub;
+ }
+ mdb_del_node(mpp.mp_page, ki, 0);
+ do_sub = 1;
+ rdata = &xdata;
+ xdata.mv_size = sizeof(MDB_db);
+ xdata.mv_data = &dummy;
+ goto new_sub;
+ }
+ goto put_sub;
+ }
+ /* same size, just replace it */
+ if (!F_ISSET(leaf->mn_flags, F_BIGDATA) &&
+ NODEDSZ(leaf) == data->mv_size) {
+ memcpy(NODEDATA(leaf), data->mv_data, data->mv_size);
+ goto done;
+ }
+ mdb_del_node(mpp.mp_page, ki, 0);
+ }
+ if (leaf == NULL) { /* append if not found */
+ ki = NUMKEYS(mpp.mp_page);
+ DPRINTF("appending key at index %i", ki);
+ }
+ } else if (rc == MDB_NOTFOUND) {
+ MDB_dpage *dp;
+ /* new file, just write a root leaf page */
+ DPUTS("allocating new root leaf page");
+ if ((dp = mdb_new_page(txn, dbi, P_LEAF, 1)) == NULL) {
+ return ENOMEM;
+ }
+ mpp.mp_page = &dp->p;
+ txn->mt_dbs[dbi].md_root = mpp.mp_page->mp_pgno;
+ txn->mt_dbs[dbi].md_depth++;
+ txn->mt_dbxs[dbi].md_dirty = 1;
+ if ((txn->mt_dbs[dbi].md_flags & (MDB_DUPSORT|MDB_DUPFIXED)) == MDB_DUPFIXED)
+ mpp.mp_page->mp_flags |= P_LEAF2;
+ ki = 0;
+ }
+ else
+ goto done;
+
+ assert(IS_LEAF(mpp.mp_page));
+ DPRINTF("there are %u keys, should insert new key at index %i",
+ NUMKEYS(mpp.mp_page), ki);
+
+ rdata = data;
+
+new_sub:
+ nsize = IS_LEAF2(mpp.mp_page) ? key->mv_size : mdb_leaf_size(txn->mt_env, key, rdata);
+ if (SIZELEFT(mpp.mp_page) < nsize) {
+ rc = mdb_split(txn, dbi, &mpp.mp_page, &ki, key, rdata, P_INVALID);
+ } else {
+ /* There is room already in this leaf page. */
+ rc = mdb_add_node(txn, dbi, mpp.mp_page, ki, key, rdata, 0, 0);
+ }
+
+ if (rc != MDB_SUCCESS)
+ txn->mt_flags |= MDB_TXN_ERROR;
+ else {
+ /* Remember if we just added a subdatabase */
+ if (flags & F_SUBDATA) {
+ leaf = NODEPTR(mpp.mp_page, ki);
+ leaf->mn_flags |= F_SUBDATA;
+ }
+
+ /* Now store the actual data in the child DB. Note that we're
+ * storing the user data in the keys field, so there are strict
+ * size limits on dupdata. The actual data fields of the child
+ * DB are all zero size.
+ */
+ if (do_sub) {
+ MDB_xcursor mx;
+
+ leaf = NODEPTR(mpp.mp_page, ki);
+put_sub:
+ mdb_xcursor_init0(txn, dbi, &mx);
+ mdb_xcursor_init1(txn, dbi, &mx, mpp.mp_page, leaf);
+ xdata.mv_size = 0;
+ xdata.mv_data = "";
+ if (flags == MDB_NODUPDATA)
+ flags = MDB_NOOVERWRITE;
+ /* converted, write the original data first */
+ if (dkey.mv_size) {
+ rc = mdb_put0(&mx.mx_txn, mx.mx_cursor.mc_dbi, &dkey, &xdata, flags);
+ if (rc) return rc;
+ leaf->mn_flags |= F_DUPDATA;
+ }
+ rc = mdb_put0(&mx.mx_txn, mx.mx_cursor.mc_dbi, data, &xdata, flags);
+ mdb_xcursor_fini(txn, dbi, &mx);
+ memcpy(NODEDATA(leaf), &mx.mx_txn.mt_dbs[mx.mx_cursor.mc_dbi],
+ sizeof(MDB_db));
+ }
+ txn->mt_dbs[dbi].md_entries++;
+ }
+
+done:
+ return rc;
+}
+
+int
+mdb_put(MDB_txn *txn, MDB_dbi dbi,
+ MDB_val *key, MDB_val *data, unsigned int flags)
+{
+ assert(key != NULL);
+ assert(data != NULL);
+
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ if (F_ISSET(txn->mt_flags, MDB_TXN_RDONLY)) {
+ return EACCES;
+ }
+
+ if (key->mv_size == 0 || key->mv_size > MAXKEYSIZE) {
+ return EINVAL;
+ }
+
+ if ((flags & (MDB_NOOVERWRITE|MDB_NODUPDATA)) != flags)
+ return EINVAL;
+
+ return mdb_put0(txn, dbi, key, data, flags);
+}
+
+int
+mdb_env_set_flags(MDB_env *env, unsigned int flag, int onoff)
+{
+#define CHANGEABLE (MDB_NOSYNC)
+ if ((flag & CHANGEABLE) != flag)
+ return EINVAL;
+ if (onoff)
+ env->me_flags |= flag;
+ else
+ env->me_flags &= ~flag;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_get_flags(MDB_env *env, unsigned int *arg)
+{
+ if (!env || !arg)
+ return EINVAL;
+
+ *arg = env->me_flags;
+ return MDB_SUCCESS;
+}
+
+int
+mdb_env_get_path(MDB_env *env, const char **arg)
+{
+ if (!env || !arg)
+ return EINVAL;
+
+ *arg = env->me_path;
+ return MDB_SUCCESS;
+}
+
+static int
+mdb_stat0(MDB_env *env, MDB_db *db, MDB_stat *arg)
+{
+ arg->ms_psize = env->me_psize;
+ arg->ms_depth = db->md_depth;
+ arg->ms_branch_pages = db->md_branch_pages;
+ arg->ms_leaf_pages = db->md_leaf_pages;
+ arg->ms_overflow_pages = db->md_overflow_pages;
+ arg->ms_entries = db->md_entries;
+
+ return MDB_SUCCESS;
+}
+int
+mdb_env_stat(MDB_env *env, MDB_stat *arg)
+{
+ int toggle;
+
+ if (env == NULL || arg == NULL)
+ return EINVAL;
+
+ mdb_env_read_meta(env, &toggle);
+
+ return mdb_stat0(env, &env->me_metas[toggle]->mm_dbs[MAIN_DBI], arg);
+}
+
+int mdb_open(MDB_txn *txn, const char *name, unsigned int flags, MDB_dbi *dbi)
+{
+ MDB_val key, data;
+ MDB_dbi i;
+ int rc, dirty = 0;
+ size_t len;
+
+ /* main DB? */
+ if (!name) {
+ *dbi = MAIN_DBI;
+ if (flags & (MDB_DUPSORT|MDB_REVERSEKEY|MDB_INTEGERKEY))
+ txn->mt_dbs[MAIN_DBI].md_flags |= (flags & (MDB_DUPSORT|MDB_REVERSEKEY|MDB_INTEGERKEY));
+ return MDB_SUCCESS;
+ }
+
+ /* Is the DB already open? */
+ len = strlen(name);
+ for (i=2; i<txn->mt_numdbs; i++) {
+ if (len == txn->mt_dbxs[i].md_name.mv_size &&
+ !strncmp(name, txn->mt_dbxs[i].md_name.mv_data, len)) {
+ *dbi = i;
+ return MDB_SUCCESS;
+ }
+ }
+
+ if (txn->mt_numdbs >= txn->mt_env->me_maxdbs - 1)
+ return ENFILE;
+
+ /* Find the DB info */
+ key.mv_size = len;
+ key.mv_data = (void *)name;
+ rc = mdb_get(txn, MAIN_DBI, &key, &data);
+
+ /* Create if requested */
+ if (rc == MDB_NOTFOUND && (flags & MDB_CREATE)) {
+ MDB_db dummy;
+ data.mv_size = sizeof(MDB_db);
+ data.mv_data = &dummy;
+ memset(&dummy, 0, sizeof(dummy));
+ dummy.md_root = P_INVALID;
+ dummy.md_flags = flags & 0xffff;
+ rc = mdb_put0(txn, MAIN_DBI, &key, &data, F_SUBDATA);
+ dirty = 1;
+ }
+
+ /* OK, got info, add to table */
+ if (rc == MDB_SUCCESS) {
+ txn->mt_dbxs[txn->mt_numdbs].md_name.mv_data = strdup(name);
+ txn->mt_dbxs[txn->mt_numdbs].md_name.mv_size = len;
+ txn->mt_dbxs[txn->mt_numdbs].md_cmp = NULL;
+ txn->mt_dbxs[txn->mt_numdbs].md_dcmp = NULL;
+ txn->mt_dbxs[txn->mt_numdbs].md_rel = NULL;
+ txn->mt_dbxs[txn->mt_numdbs].md_parent = MAIN_DBI;
+ txn->mt_dbxs[txn->mt_numdbs].md_dirty = dirty;
+ memcpy(&txn->mt_dbs[txn->mt_numdbs], data.mv_data, sizeof(MDB_db));
+ *dbi = txn->mt_numdbs;
+ txn->mt_env->me_dbs[0][txn->mt_numdbs] = txn->mt_dbs[txn->mt_numdbs];
+ txn->mt_env->me_dbs[1][txn->mt_numdbs] = txn->mt_dbs[txn->mt_numdbs];
+ txn->mt_numdbs++;
+ }
+
+ return rc;
+}
+
+int mdb_stat(MDB_txn *txn, MDB_dbi dbi, MDB_stat *arg)
+{
+ if (txn == NULL || arg == NULL || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ return mdb_stat0(txn->mt_env, &txn->mt_dbs[dbi], arg);
+}
+
+void mdb_close(MDB_txn *txn, MDB_dbi dbi)
+{
+ char *ptr;
+ if (dbi <= MAIN_DBI || dbi >= txn->mt_numdbs)
+ return;
+ ptr = txn->mt_dbxs[dbi].md_name.mv_data;
+ txn->mt_dbxs[dbi].md_name.mv_data = NULL;
+ txn->mt_dbxs[dbi].md_name.mv_size = 0;
+ free(ptr);
+}
+
+int mdb_set_compare(MDB_txn *txn, MDB_dbi dbi, MDB_cmp_func *cmp)
+{
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ txn->mt_dbxs[dbi].md_cmp = cmp;
+ return MDB_SUCCESS;
+}
+
+int mdb_set_dupsort(MDB_txn *txn, MDB_dbi dbi, MDB_cmp_func *cmp)
+{
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ txn->mt_dbxs[dbi].md_dcmp = cmp;
+ return MDB_SUCCESS;
+}
+
+int mdb_set_relfunc(MDB_txn *txn, MDB_dbi dbi, MDB_rel_func *rel)
+{
+ if (txn == NULL || !dbi || dbi >= txn->mt_numdbs)
+ return EINVAL;
+
+ txn->mt_dbxs[dbi].md_rel = rel;
+ return MDB_SUCCESS;
+}
--- /dev/null
+/** @file mdb.h
+ * @brief memory-mapped database library
+ *
+ * @mainpage MDB Memory-Mapped Database Manager
+ * MDB is a Btree-based database management library modeled loosely on the
+ * BerkeleyDB API, but much simplified. The entire database is exposed
+ * in a read-only memory map, and all data fetches return data directly
+ * from the mapped memory, so no malloc's or memcpy's occur during
+ * data fetches. As such, the library is extremely simple because it
+ * requires no page caching layer of its own, and it is extremely high
+ * performance and memory-efficient. It is also fully transactional with
+ * full ACID semantics, and because the memory map is read-only, the
+ * database integrity cannot be corrupted by stray pointer writes from
+ * application code.
+ *
+ * The library is fully thread-aware and supports concurrent read/write
+ * access from multiple processes and threads. Data pages use a copy-on-
+ * write strategy so no active data pages are ever overwritten, which
+ * also provides resistance to corruption and eliminates the need of any
+ * special recovery procedures after a system crash. Writes are fully
+ * serialized; only one write transaction may be active at a time, which
+ * guarantees that writers can never deadlock. The database structure is
+ * multi-versioned so readers run with no locks; writers cannot block
+ * readers, and readers don't block writers.
+ *
+ * Unlike other well-known database mechanisms which use either write-ahead
+ * transaction logs or append-only data writes, MDB requires no maintenance
+ * during operation. Both write-ahead loggers and append-only databases
+ * require periodic checkpointing and/or compaction of their log or database
+ * files otherwise they grow without bound. MDB tracks free pages within
+ * the database and re-uses them for new write operations, so the database
+ * size does not grow without bound in normal use.
+ *
+ * @author Howard Chu, Symas Corporation.
+ *
+ * @copyright Copyright 2011 Howard Chu, Symas Corp. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ *
+ * @par Derived From:
+ * This code is derived from btree.c written by Martin Hedenfalk.
+ *
+ * Copyright (c) 2009, 2010 Martin Hedenfalk <martin@bzero.se>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+#ifndef _MDB_H_
+#define _MDB_H_
+
+#include <sys/types.h>
+
+/** @defgroup Version Version Macros
+ * @{
+ */
+/** Library major version */
+#define MDB_VERSION_MAJOR 0
+/** Library minor version */
+#define MDB_VERSION_MINOR 8
+/** Library patch version */
+#define MDB_VERSION_PATCH 0
+
+/** Combine args a,b,c into a single integer for easy version comparisons */
+#define MDB_VERINT(a,b,c) (((a) << 24) | ((b) << 16) | (c))
+
+/** The full library version as a single integer */
+#define MDB_VERSION_FULL \
+ MDB_VERINT(MDB_VERSION_MAJOR,MDB_VERSION_MINOR,MDB_VERSION_PATCH)
+
+/** The release date of this library version */
+#define MDB_VERSION_DATE "August 11, 2011"
+
+/** A stringifier for the version info */
+#define MDB_VERSTR(a,b,c,d) "MDB " #a "." #b "." #c ": (" #d ")"
+
+/** A helper for the stringifier macro */
+#define MDB_VERFOO(a,b,c,d) MDB_VERSTR(a,b,c,d)
+
+/** The full library version as a C string */
+#define MDB_VERSION_STRING \
+ MDB_VERFOO(MDB_VERSION_MAJOR,MDB_VERSION_MINOR,MDB_VERSION_PATCH,MDB_VERSION_DATE)
+/** @} */
+
+/** Opaque structure for navigating through a database */
+typedef struct MDB_cursor MDB_cursor;
+
+/** Opaque structure for a transaction handle. All database operations require
+ * a transaction handle. Transactions may be read-only or read-write.
+ */
+typedef struct MDB_txn MDB_txn;
+
+/** Opaque structure for a database environment. A DB environment supports multiple
+ * databases, all residing in the same shared-memory map.
+ */
+typedef struct MDB_env MDB_env;
+
+/** A handle for an individual database in the DB environment. */
+typedef unsigned int MDB_dbi;
+
+/** Generic structure used for passing keys and data in and out of the database. */
+typedef struct MDB_val {
+ size_t mv_size; /**< size of the data item */
+ void *mv_data; /**< address of the data item */
+} MDB_val;
+
+/** A callback function used to compare two keys in a database */
+typedef int (MDB_cmp_func)(const MDB_val *a, const MDB_val *b);
+
+/** A callback function used to relocate a position-dependent data item
+ * in a fixed-address database. This feature is currently unimplemented.
+ */
+typedef void (MDB_rel_func)(void *ptr, void *oldptr);
+
+/** @defgroup mdb_put mdb_put flags
+ * @{
+ */
+/** For mdb_put: don't write if the key already exists. */
+#define MDB_NOOVERWRITE 0x10
+/** For mdb_put: don't write if the key and data pair already exist.
+ * Only for #MDB_DUPSORT
+ */
+#define MDB_NODUPDATA 0x20
+/* @} */
+
+/** Cursor operations */
+typedef enum MDB_cursor_op {
+ MDB_FIRST, /**< Position at first key/data item */
+ MDB_GET_BOTH, /**< Position at key/data pair. Only for #MDB_DUPSORT */
+ MDB_GET_BOTH_RANGE, /**< position at key, nearest data. Only for #MDB_DUPSORT */
+ MDB_GET_MULTIPLE, /**< Return all the duplicate data items at the current
+ cursor position. Only for #MDB_DUPFIXED */
+ MDB_LAST, /**< Position at last key/data item */
+ MDB_NEXT, /**< Position at next data item */
+ MDB_NEXT_DUP, /**< Position at next data item of current key.
+ Only for #MDB_DUPSORT */
+ MDB_NEXT_MULTIPLE, /**< Return all duplicate data items at the next
+ cursor position. Only for #MDB_DUPFIXED */
+ MDB_NEXT_NODUP, /**< Position at first data item of next key.
+ Only for #MDB_DUPSORT */
+ MDB_PREV, /**< Position at previous data item */
+ MDB_PREV_DUP, /**< Position at previous data item of current key.
+ Only for #MDB_DUPSORT */
+ MDB_PREV_NODUP, /**< Position at last data item of previous key.
+ only for #MDB_DUPSORT */
+ MDB_SET, /**< Position at specified key */
+ MDB_SET_RANGE /**< Position at first key greater than or equal to specified key. */
+} MDB_cursor_op;
+
+/** @defgroup errors Return Codes
+ *
+ * BerkeleyDB uses -30800 to -30999, we'll go under them
+ * @{
+ */
+ /** Successful result */
+#define MDB_SUCCESS 0
+ /** key/data pair already exists */
+#define MDB_KEYEXIST (-30799)
+ /** key/data pair not found (EOF) */
+#define MDB_NOTFOUND (-30798)
+ /** Requested page not found - this usually indicates corruption */
+#define MDB_PAGE_NOTFOUND (-30797)
+ /** Located page was wrong type */
+#define MDB_CORRUPTED (-30796)
+ /** Update of meta page failed, probably I/O error */
+#define MDB_PANIC (-30795)
+ /** Environment version mismatch */
+#define MDB_VERSION_MISMATCH (-30794)
+/** @} */
+
+/** @defgroup mdb_open mdb_open flags
+ * @{
+ */
+ /** use reverse string keys */
+#define MDB_REVERSEKEY 0x02
+ /** use sorted duplicates */
+#define MDB_DUPSORT 0x04
+ /** numeric keys in native byte order */
+#define MDB_INTEGERKEY 0x08
+ /** with #MDB_DUPSORT, sorted dup items have fixed size */
+#define MDB_DUPFIXED 0x10
+ /** with #MDB_DUPSORT, dups are numeric in native byte order */
+#define MDB_INTEGERDUP 0x20
+/** @} */
+
+/** @defgroup mdb_env environment flags
+ * @{
+ */
+ /** mmap at a fixed address */
+#define MDB_FIXEDMAP 0x01
+ /** don't fsync after commit */
+#define MDB_NOSYNC 0x10000
+ /** read only */
+#define MDB_RDONLY 0x20000
+/** @} */
+
+ /** create env or DB if not already existing */
+#define MDB_CREATE 0x40000
+
+/** Statistics for a database in the environment */
+typedef struct MDB_stat {
+ unsigned int ms_psize; /**< Size of a database page.
+ This is currently the same for all databases. */
+ unsigned int ms_depth; /**< Depth (height) of the B-tree */
+ unsigned long ms_branch_pages; /**< Number of internal (non-leaf) pages */
+ unsigned long ms_leaf_pages; /**< Number of leaf pages */
+ unsigned long ms_overflow_pages; /**< Number of overflow pages */
+ unsigned long ms_entries; /**< Number of data items */
+} MDB_stat;
+
+ /** Return the mdb library version information.
+ * @param[out] major if non-NULL, the library major version number is copied here
+ * @param[out] minor if non-NULL, the library minor version number is copied here
+ * @param[out] patch if non-NULL, the library patch version number is copied here
+ * @retval "version string" The library version as a string
+ */
+char *mdb_version(int *major, int *minor, int *patch);
+
+ /** Return a string describing a given error code.
+ * This function is a superset of the ANSI C X3.159-1989 (ANSI C) strerror(3)
+ * function. If the error code is greater than or equal to 0, then the string
+ * returned by the system function strerror(3) is returned. If the error code
+ * is less than 0, an error string corresponding to the MDB library error is
+ * returned. See @ref errors for a list of MDB-specific error codes.
+ * @param[in] err The error code
+ * @retval "error message" The description of the error
+ */
+char *mdb_strerror(int err);
+
+ /** Create an MDB environment handle.
+ * This function allocates memory for a #MDB_env structure. To release
+ * the allocated memory and discard the handle, call #mdb_env_close().
+ * Before the handle may be used, it must be opened using #mdb_env_open().
+ * Various other options may also need to be set before opening the handle,
+ * e.g. #mdb_env_set_mapsize(), #mdb_env_set_maxreaders(), #mdb_env_set_maxdbs(),
+ * depending on usage requirements.
+ * @param[out] env The address where the new handle will be stored
+ * @return A non-zero error value on failure and 0 on success.
+ */
+int mdb_env_create(MDB_env **env);
+
+ /** Open an environment handle.
+ * If this function fails, #mdb_env_close() must be called to discard the #MDB_env handle.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] path The directory in which the database files reside
+ * @param[in] flags Special options for this environment. This parameter
+ * must be set to 0 or by bitwise OR'ing together one or more of the
+ * values described here.
+ * <ul>
+ * <li>#MDB_FIXEDMAP
+ * use a fixed address for the mmap region. This flag must be specified
+ * when creating the environment, and is stored persistently in the environment.
+ * If successful, the memory map will always reside at the same virtual address
+ * and pointers used to reference data items in the database will be constant
+ * across multiple invocations. This option may not always work, depending on
+ * how the operating system has allocated memory to shared libraries and other uses.
+ * The feature is highly experimental.
+ * <li>#MDB_NOSYNC
+ * Don't perform a synchronous flush after committing a transaction. This means
+ * transactions will exhibit the ACI (atomicity, consistency, and isolation)
+ * properties, but not D (durability); that is database integrity will be
+ * maintained but it is possible some number of the most recently committed
+ * transactions may be undone after a system crash. The number of transactions
+ * at risk is governed by how often the system flushes dirty buffers to disk
+ * and how often #mdb_env_sync() is called. This flag may be changed
+ * at any time using #mdb_env_set_flags().
+ * <li>#MDB_RDONLY
+ * Open the environment in read-only mode. No write operations will be allowed.
+ * <li>#MDB_CREATE
+ * Create the environment if it doesn't already exist.
+ * </ul>
+ * @param[in] mode The UNIX permissions to set on created files.
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_VERSION_MISMATCH - the version of the MDB library doesn't match the
+ * version that created the database environment.
+ * <li>EINVAL - the environment file headers are corrupted.
+ * <li>ENOENT - the directory specified by the path parameter doesn't exist.
+ * <li>EACCES - the user didn't have permission to access the environment files.
+ * <li>EAGAIN - the environment was locked by another process.
+ * </ul>
+ */
+int mdb_env_open(MDB_env *env, const char *path, unsigned int flags, mode_t mode);
+
+ /** Return statistics about the MDB environment.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[out] stat The address of an #MDB_stat structure
+ * where the statistics will be copied
+ */
+int mdb_env_stat(MDB_env *env, MDB_stat *stat);
+
+ /** Flush the data buffers to disk.
+ * Data is always written to disk when #mdb_txn_commit() is called,
+ * but the operating system may keep it buffered. MDB always flushes
+ * the OS buffers upon commit as well, unless the environment was
+ * opened with #MDB_NOSYNC.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] force If non-zero, force the flush to occur. Otherwise
+ * if the environment has the #MDB_NOSYNC flag set the flushes
+ * will be omitted.
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * <li>EIO - an error occurred during synchronization.
+ * </ul>
+ */
+int mdb_env_sync(MDB_env *env, int force);
+
+ /** Close the environment and release the memory map.
+ * Only a single thread may call this function. All transactions, databases,
+ * and cursors must already be closed before calling this function. Attempts to
+ * use any such handles after calling this function will cause a SIGSEGV.
+ * The environment handle will be freed and must not be used again after this call.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ */
+void mdb_env_close(MDB_env *env);
+
+ /** Set environment flags.
+ * This may be used to set some flags that weren't already set during
+ * #mdb_env_open(), or to unset these flags. Currently only the
+ * #MDB_NOSYNC flag setting may be changed with this function.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] flags The flags to change, bitwise OR'ed together
+ * @param[in] onoff A non-zero value sets the flags, zero clears them.
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_env_set_flags(MDB_env *env, unsigned int flags, int onoff);
+
+ /** Get environment flags.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[out] flags The address of an integer to store the flags
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_env_get_flags(MDB_env *env, unsigned int *flags);
+
+ /** Return the path that was used in #mdb_env_open().
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[out] path Address of a string pointer to contain the path. This
+ * is the actual string in the environment, not a copy. It should not be
+ * altered in any way.
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_env_get_path(MDB_env *env, const char **path);
+
+ /** Set the size of the memory map to use for this environment.
+ * The size should be a multiple of the OS page size. The default is
+ * 10485760 bytes. The value should be chosen as large as possible,
+ * to accomodate future growth of the database.
+ * This function may only be called after #mdb_env_create() and before #mdb_env_open().
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] size The size in bytes
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified, or the environment is already open.
+ * </ul>
+ */
+int mdb_env_set_mapsize(MDB_env *env, size_t size);
+
+ /** Set the maximum number of threads for the environment.
+ * This defines the number of slots in the lock table that is used to track readers in the
+ * the environment. The default is 126.
+ * This function may only be called after #mdb_env_create() and before #mdb_env_open().
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] readers The maximum number of threads
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified, or the environment is already open.
+ * </ul>
+ */
+int mdb_env_set_maxreaders(MDB_env *env, int readers);
+
+ /** Get the maximum number of threads for the environment.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[out] readers Address of an integer to store the number of readers
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_env_get_maxreaders(MDB_env *env, int *readers);
+
+ /** Set the maximum number of databases for the environment.
+ * This function is only needed if multiple databases will be used in the
+ * environment. Simpler applications that only use a single database can ignore
+ * this option.
+ * This function may only be called after #mdb_env_create() and before #mdb_env_open().
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] dbs The maximum number of databases
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified, or the environment is already open.
+ * </ul>
+ */
+int mdb_env_set_maxdbs(MDB_env *env, int dbs);
+
+ /** Create a transaction for use with the environment.
+ * The transaction handle may be discarded using #mdb_txn_abort() or #mdb_txn_commit().
+ * @note Transactions may not span threads, a transaction must only be used by a
+ * single thread.
+ * @note Cursors may not span transactions; each cursor must be opened and closed
+ * within a single transaction.
+ * @param[in] env An environment handle returned by #mdb_env_create()
+ * @param[in] flags Special options for this transaction. This parameter
+ * must be set to 0 or by bitwise OR'ing together one or more of the
+ * values described here.
+ * <ul>
+ * <li>#MDB_RDONLY
+ * This transaction will not perform any write operations.
+ * </ul>
+ * @param[out] txn Address where the new #MDB_txn handle will be stored
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_PANIC - a fatal error occurred earlier and the environment
+ * must be shut down.
+ * <li>ENOMEM - out of memory, or a read-only transaction was requested and
+ * the reader lock table is full. See #mdb_env_set_maxreaders().
+ * </ul>
+ */
+int mdb_txn_begin(MDB_env *env, unsigned int flags, MDB_txn **txn);
+
+ /** Commit all the operations of a transaction into the database.
+ * All cursors opened within the transaction must be closed before the transaction
+ * is committed.
+ * The transaction handle will be freed and must not be used again after this call.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * <li>ENOSPC - no more disk space.
+ * <li>EIO - a low-level I/O error occurred while writing.
+ * </ul>
+ */
+int mdb_txn_commit(MDB_txn *txn);
+
+ /** Abandon all the operations of the transaction instead of saving them.
+ * All cursors opened within the transaction must be closed before the transaction
+ * is aborted.
+ * The transaction handle will be freed and must not be used again after this call.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ */
+void mdb_txn_abort(MDB_txn *txn);
+
+ /** Reset a read-only transaction.
+ * This releases the current reader lock but doesn't free the
+ * transaction handle, allowing it to be used again later by #mdb_txn_renew().
+ * It otherwise has the same affect as #mdb_txn_abort() but saves some memory
+ * allocation/deallocation overhead if a thread is going to start a new
+ * read-only transaction again soon.
+ * All cursors opened within the transaction must be closed before the transaction
+ * is reset.
+ * Reader locks generally don't interfere with writers, but they keep old
+ * versions of database pages allocated. Thus they prevent the old pages
+ * from being reused when writers commit new data, and so under heavy load
+ * the database size may grow much more rapidly than otherwise.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ */
+void mdb_txn_reset(MDB_txn *txn);
+
+ /** Renew a read-only transaction.
+ * This acquires a new reader lock for a transaction handle that had been
+ * released by #mdb_txn_reset(). It must be called before a reset transaction
+ * may be used again.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_PANIC - a fatal error occurred earlier and the environment
+ * must be shut down.
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_txn_renew(MDB_txn *txn);
+
+ /** Open a database in the environment.
+ * The database handle may be discarded by calling #mdb_close().
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] name The name of the database to open. If only a single
+ * database is needed in the enviroment, this value may be NULL.
+ * @param[in] flags Special options for this database. This parameter
+ * must be set to 0 or by bitwise OR'ing together one or more of the
+ * values described here.
+ * <ul>
+ * <li>#MDB_REVERSEKEY
+ * Keys are strings to be compared in reverse order, from the end
+ * of the strings to the beginning. By default, Keys are treated as strings and
+ * compared from beginning to end.
+ * <li>#MDB_DUPSORT
+ * Duplicate keys may be used in the database. (Or, from another perspective,
+ * keys may have multiple data items, stored in sorted order.) By default
+ * keys must be unique and may have only a single data item.
+ * <li>#MDB_INTEGERKEY
+ * Keys are binary integers in native byte order. On Big-Endian systems
+ * this flag has no effect. On Little-Endian systems this flag behaves
+ * the same as #MDB_REVERSEKEY. This flag is simply provided as a
+ * convenience so that applications don't need to detect Endianness themselves
+ * when using integers as keys.
+ * <li>#MDB_DUPFIXED
+ * This flag may only be used in combination with #MDB_DUPSORT. This option
+ * tells the library that the data items for this database are all the same
+ * size, which allows further optimizations in storage and retrieval. When
+ * all data items are the same size, the #MDB_GET_MULTIPLE and #MDB_NEXT_MULTIPLE
+ * cursor operations may be used to retrieve multiple items at once.
+ * <li>#MDB_INTEGERDUP
+ * This option specifies that duplicate data items are also integers, and
+ * should be sorted as such.
+ * <li>#MDB_CREATE
+ * Create the named database if it doesn't exist. This option is not
+ * allowed in a read-only transaction or a read-only environment.
+ * </ul>
+ * @param[out] dbi Address where the new #MDB_dbi handle will be stored
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_NOTFOUND - the specified database doesn't exist in the environment
+ * and #MDB_CREATE was not specified.
+ * <li>ENFILE - too many databases have been opened. See #mdb_env_set_maxdbs().
+ * </ul>
+ */
+int mdb_open(MDB_txn *txn, const char *name, unsigned int flags, MDB_dbi *dbi);
+
+ /** Retrieve statistics for a database.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[out] stat The address of an #MDB_stat structure
+ * where the statistics will be copied
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_stat(MDB_txn *txn, MDB_dbi dbi, MDB_stat *stat);
+
+ /** Close a database handle.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ */
+void mdb_close(MDB_txn *txn, MDB_dbi dbi);
+
+ /** Set a custom key comparison function for a database.
+ * The comparison function is called whenever it is necessary to compare a
+ * key specified by the application with a key currently stored in the database.
+ * If no comparison function is specified, and no special key flags were specified
+ * with #mdb_open(), the keys are compared lexically, with shorter keys collating
+ * before longer keys.
+ * @warning This function must be called before any data access functions are used,
+ * otherwise data corruption may occur. The same function must be used by every
+ * process accessing the database, every time the database is used.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] cmp A #MDB_cmp_func function
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_set_compare(MDB_txn *txn, MDB_dbi dbi, MDB_cmp_func *cmp);
+
+ /** Set a custom data comparison function for a #MDB_DUPSORT database.
+ * This comparison function is called whenever it is necessary to compare a data
+ * item specified by the application with a data item currently stored in the database.
+ * This function only takes effect if the database was opened with the #MDB_DUPSORT
+ * flag.
+ * If no comparison function is specified, and no special key flags were specified
+ * with #mdb_open(), the data items are compared lexically, with shorter items collating
+ * before longer items.
+ * @warning This function must be called before any data access functions are used,
+ * otherwise data corruption may occur. The same function must be used by every
+ * process accessing the database, every time the database is used.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] cmp A #MDB_cmp_func function
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_set_dupsort(MDB_txn *txn, MDB_dbi dbi, MDB_cmp_func *cmp);
+
+ /** Set a relocation function for a #MDB_FIXEDMAP database.
+ * @todo The relocation function is called whenever it is necessary to move the data
+ * of an item to a different position in the database (e.g. through tree
+ * balancing operations, shifts as a result of adds or deletes, etc.). It is
+ * intended to allow address/position-dependent data items to be stored in
+ * a database in an environment opened with the #MDB_FIXEDMAP option.
+ * Currently the relocation feature is unimplemented and setting
+ * this function has no effect.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] rel A #MDB_rel_func function
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_set_relfunc(MDB_txn *txn, MDB_dbi dbi, MDB_rel_func *rel);
+
+ /** Get items from a database.
+ * This function retrieves key/data pairs from the database. The address
+ * and length of the data associated with the specified \b key are returned
+ * in the structure to which \b data refers.
+ * If the database supports duplicate keys (#MDB_DUPSORT) then the
+ * first data item for the key will be returned. Retrieval of other
+ * items requires the use of #mdb_cursor_get().
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] key The key to search for in the database
+ * @param[out] data The data corresponding to the key
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_NOTFOUND - the key was not in the database.
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_get(MDB_txn *txn, MDB_dbi dbi, MDB_val *key, MDB_val *data);
+
+ /** Store items into a database.
+ * This function stores key/data pairs in the database. The default behavior
+ * is to enter the new key/data pair, replacing any previously existing key
+ * if duplicates are disallowed, or adding a duplicate data item if
+ * duplicates are allowed (#MDB_DUPSORT).
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] key The key to store in the database
+ * @param[in] data The data to store
+ * @param[in] flags Special options for this operation. This parameter
+ * must be set to 0 or by bitwise OR'ing together one or more of the
+ * values described here.
+ * <ul>
+ * <li>#MDB_NODUPDATA - enter the new key/data pair only if it does not
+ * already appear in the database. This flag may only be specified
+ * if the database was opened with #MDB_DUPSORT. The function will
+ * return #MDB_KEYEXIST if the key/data pair already appears in the
+ * database.
+ * <li>#MDB_NOOVERWRITE - enter the new key/data pair only if the key
+ * does not already appear in the database. The function will return
+ * #MDB_KEYEXIST if the key already appears in the database, even if
+ * the database supports duplicates (#MDB_DUPSORT).
+ * </ul>
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EACCESS - an attempt was made to write in a read-only transaction.
+ * <li>EINVAL - an invalid parameter was specified.
+ * <li>ENOMEM - the database is full, see #mdb_env_set_mapsize().
+ * </ul>
+ */
+int mdb_put(MDB_txn *txn, MDB_dbi dbi, MDB_val *key, MDB_val *data,
+ unsigned int flags);
+
+ /** Delete items from a database.
+ * This function removes key/data pairs from the database.
+ * If the database does not support sorted duplicate data items
+ * (#MDB_DUPSORT) the data parameter is ignored.
+ * If the database supports sorted duplicates and the data parameter
+ * is NULL, all of the duplicate data items for the key will be
+ * deleted. Otherwise, if the data parameter is non-NULL
+ * only the matching data item will be deleted.
+ * This function will return #MDB_NOTFOUND if the specified key/data
+ * pair is not in the database.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] key The key to delete from the database
+ * @param[in] data The data to delete
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EACCESS - an attempt was made to write in a read-only transaction.
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_del(MDB_txn *txn, MDB_dbi dbi, MDB_val *key, MDB_val *data);
+
+ /** Create a cursor handle.
+ * Cursors are associated with a specific transaction and database and
+ * may not span threads.
+ * @todo Cursors only support read operations. Support for cursor_put() and
+ * cursor_del() needs to be added.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[out] cursor Address where the new #MDB_cursor handle will be stored
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_cursor_open(MDB_txn *txn, MDB_dbi dbi, MDB_cursor **cursor);
+
+ /** Close a cursor handle.
+ * The cursor handle will be freed and must not be used again after this call.
+ * @param[in] cursor A cursor handle returned by #mdb_cursor_open()
+ */
+void mdb_cursor_close(MDB_cursor *cursor);
+
+ /** Retrieve by cursor.
+ * This function retrieves key/data pairs from the database. The address and length
+ * of the key are returned in the object to which \b key refers (except for the
+ * case of the #MDB_SET option, in which the \b key object is unchanged), and
+ * the address and length of the data are returned in the object to which \b data
+ * refers.
+ * @bug Cursors are not coordinated with write operations. If a cursor in a
+ * write transaction is performing a sequential scan while records are being
+ * inserted or deleted in the same transaction, the cursor will be corrupted.
+ * @param[in] cursor A cursor handle returned by #mdb_cursor_open()
+ * @param[in,out] key The key for a retrieved item
+ * @param[in,out] data The data of a retrieved item
+ * @param[in] op A cursor operation #MDB_cursor_op
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>#MDB_NOTFOUND - no matching key found.
+ * <li>EINVAL - an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_cursor_get(MDB_cursor *cursor, MDB_val *key, MDB_val *data,
+ MDB_cursor_op op);
+
+ /** Return count of duplicates for current key.
+ * This call is only valid on databases that support sorted duplicate
+ * data items #MDB_DUPSORT.
+ * @param[in] cursor A cursor handle returned by #mdb_cursor_open()
+ * @param[out] countp Address where the count will be stored
+ * @return A non-zero error value on failure and 0 on success. Some possible
+ * errors are:
+ * <ul>
+ * <li>EINVAL - cursor is not initialized, or an invalid parameter was specified.
+ * </ul>
+ */
+int mdb_cursor_count(MDB_cursor *cursor, unsigned long *countp);
+
+ /** Compare two data items according to a particular database.
+ * This returns a comparison as if the two data items were keys in the
+ * specified database.
+ * @param[in] txn A transaction handle returned by #mdb_txn_begin()
+ * @param[in] dbi A database handle returned by #mdb_open()
+ * @param[in] a The first item to compare
+ * @param[in] b The second item to compare
+ * @return < 0 if a < b, 0 if a == b, > 0 if a > b
+ */
+int mdb_cmp(MDB_txn *txn, MDB_dbi dbi, const MDB_val *a, const MDB_val *b);
+
+#endif /* _MDB_H_ */
--- /dev/null
+/* mdb_stat.c - memory-mapped database status tool */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+#include <stdio.h>
+#include <stdlib.h>
+#include <time.h>
+#include "mdb.h"
+
+int main(int argc,char * argv[])
+{
+ int rc;
+ MDB_env *env;
+ MDB_txn *txn;
+ MDB_dbi dbi;
+ MDB_stat mst;
+ char *envname = argv[1];
+ char *subname = NULL;
+
+ rc = mdb_env_create(&env);
+
+ if (argc > 2) {
+ mdb_env_set_maxdbs(env, 4);
+ subname = argv[2];
+ }
+
+ rc = mdb_env_open(env, envname, MDB_RDONLY, 0);
+ if (rc) {
+ printf("mdb_env_open failed, error %d\n", rc);
+ exit(1);
+ }
+ rc = mdb_txn_begin(env, 1, &txn);
+ if (rc) {
+ printf("mdb_txn_begin failed, error %d\n", rc);
+ exit(1);
+ }
+ rc = mdb_open(txn, subname, 0, &dbi);
+ if (rc) {
+ printf("mdb_open failed, error %d\n", rc);
+ exit(1);
+ }
+
+ rc = mdb_stat(txn, dbi, &mst);
+ printf("Page size: %u\n", mst.ms_psize);
+ printf("Tree depth: %u\n", mst.ms_depth);
+ printf("Branch pages: %lu\n", mst.ms_branch_pages);
+ printf("Leaf pages: %lu\n", mst.ms_leaf_pages);
+ printf("Overflow pages: %lu\n", mst.ms_overflow_pages);
+ printf("Entries: %lu\n", mst.ms_entries);
+ mdb_close(txn, dbi);
+ mdb_txn_abort(txn);
+ mdb_env_close(env);
+
+ return 0;
+}
--- /dev/null
+/* idl.c - ldap bdb back-end ID list functions */
+/* $OpenLDAP$ */
+/* This work is part of OpenLDAP Software <http://www.openldap.org/>.
+ *
+ * Copyright 2000-2011 The OpenLDAP Foundation.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+
+#include <string.h>
+#include <sys/types.h>
+#include <assert.h>
+#include "midl.h"
+
+typedef unsigned long pgno_t;
+
+#define CMP(x,y) ( (x) > (y) ? -1 : (x) < (y) )
+
+unsigned mdb_midl_search( ID *ids, ID id )
+{
+ /*
+ * binary search of id in ids
+ * if found, returns position of id
+ * if not found, returns first position greater than id
+ */
+ unsigned base = 0;
+ unsigned cursor = 0;
+ int val = 0;
+ unsigned n = ids[0];
+
+ while( 0 < n ) {
+ int pivot = n >> 1;
+ cursor = base + pivot;
+ val = CMP( id, ids[cursor + 1] );
+
+ if( val < 0 ) {
+ n = pivot;
+
+ } else if ( val > 0 ) {
+ base = cursor + 1;
+ n -= pivot + 1;
+
+ } else {
+ return cursor + 1;
+ }
+ }
+
+ if( val > 0 ) {
+ return cursor + 2;
+ } else {
+ return cursor + 1;
+ }
+}
+
+int mdb_midl_insert( ID *ids, ID id )
+{
+ unsigned x, i;
+
+ if (MDB_IDL_IS_RANGE( ids )) {
+ /* if already in range, treat as a dup */
+ if (id >= MDB_IDL_FIRST(ids) && id <= MDB_IDL_LAST(ids))
+ return -1;
+ if (id < MDB_IDL_FIRST(ids))
+ ids[1] = id;
+ else if (id > MDB_IDL_LAST(ids))
+ ids[2] = id;
+ return 0;
+ }
+
+ x = mdb_midl_search( ids, id );
+ assert( x > 0 );
+
+ if( x < 1 ) {
+ /* internal error */
+ return -2;
+ }
+
+ if ( x <= ids[0] && ids[x] == id ) {
+ /* duplicate */
+ return -1;
+ }
+
+ if ( ++ids[0] >= MDB_IDL_DB_MAX ) {
+ if( id < ids[1] ) {
+ ids[1] = id;
+ ids[2] = ids[ids[0]-1];
+ } else if ( ids[ids[0]-1] < id ) {
+ ids[2] = id;
+ } else {
+ ids[2] = ids[ids[0]-1];
+ }
+ ids[0] = NOID;
+
+ } else {
+ /* insert id */
+ for (i=ids[0]; i>x; i--)
+ ids[i] = ids[i-1];
+ ids[x] = id;
+ }
+
+ return 0;
+}
+
+unsigned mdb_midl2_search( MIDL2 *ids, MIDL2 *id )
+{
+ /*
+ * binary search of id in ids
+ * if found, returns position of id
+ * if not found, returns first position greater than id
+ */
+ unsigned base = 0;
+ unsigned cursor = 0;
+ int val = 0;
+ unsigned n = ids[0].mid;
+
+ while( 0 < n ) {
+ int pivot = n >> 1;
+ cursor = base + pivot;
+ val = CMP( ids[cursor + 1].mid, id->mid );
+
+ if( val < 0 ) {
+ n = pivot;
+
+ } else if ( val > 0 ) {
+ base = cursor + 1;
+ n -= pivot + 1;
+
+ } else {
+ return cursor + 1;
+ }
+ }
+
+ if( val > 0 ) {
+ return cursor + 2;
+ } else {
+ return cursor + 1;
+ }
+}
+
+int mdb_midl2_insert( MIDL2 *ids, MIDL2 *id )
+{
+ unsigned x, i;
+
+ x = mdb_midl2_search( ids, id );
+ assert( x > 0 );
+
+ if( x < 1 ) {
+ /* internal error */
+ return -2;
+ }
+
+ if ( x <= ids[0].mid && ids[x].mid == id->mid ) {
+ /* duplicate */
+ return -1;
+ }
+
+ if ( ids[0].mid >= MDB_IDL_DB_MAX ) {
+ /* too big */
+ return -2;
+
+ } else {
+ /* insert id */
+ ids[0].mid++;
+ for (i=ids[0].mid; i>x; i--)
+ ids[i] = ids[i-1];
+ ids[x] = *id;
+ }
+
+ return 0;
+}
--- /dev/null
+/* idl.h - ldap bdb back-end ID list header file */
+/* $OpenLDAP$ */
+/* This work is part of OpenLDAP Software <http://www.openldap.org/>.
+ *
+ * Copyright 2000-2011 The OpenLDAP Foundation.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+
+#ifndef _MDB_MIDL_H_
+#define _MDB_MIDL_H_
+
+#define ID unsigned long
+#define NOID (~(ID)0)
+
+/* IDL sizes - likely should be even bigger
+ * limiting factors: sizeof(ID), thread stack size
+ */
+#define MDB_IDL_LOGN 16 /* DB_SIZE is 2^16, UM_SIZE is 2^17 */
+#define MDB_IDL_DB_SIZE (1<<MDB_IDL_LOGN)
+#define MDB_IDL_UM_SIZE (1<<(MDB_IDL_LOGN+1))
+#define MDB_IDL_UM_SIZEOF (MDB_IDL_UM_SIZE * sizeof(ID))
+
+#define MDB_IDL_DB_MAX (MDB_IDL_DB_SIZE-1)
+
+#define MDB_IDL_UM_MAX (MDB_IDL_UM_SIZE-1)
+
+#define MDB_IDL_IS_RANGE(ids) ((ids)[0] == NOID)
+#define MDB_IDL_RANGE_SIZE (3)
+#define MDB_IDL_RANGE_SIZEOF (MDB_IDL_RANGE_SIZE * sizeof(ID))
+#define MDB_IDL_SIZEOF(ids) ((MDB_IDL_IS_RANGE(ids) \
+ ? MDB_IDL_RANGE_SIZE : ((ids)[0]+1)) * sizeof(ID))
+
+#define MDB_IDL_RANGE_FIRST(ids) ((ids)[1])
+#define MDB_IDL_RANGE_LAST(ids) ((ids)[2])
+
+#define MDB_IDL_RANGE( ids, f, l ) \
+ do { \
+ (ids)[0] = NOID; \
+ (ids)[1] = (f); \
+ (ids)[2] = (l); \
+ } while(0)
+
+#define MDB_IDL_ZERO(ids) \
+ do { \
+ (ids)[0] = 0; \
+ (ids)[1] = 0; \
+ (ids)[2] = 0; \
+ } while(0)
+
+#define MDB_IDL_IS_ZERO(ids) ( (ids)[0] == 0 )
+#define MDB_IDL_IS_ALL( range, ids ) ( (ids)[0] == NOID \
+ && (ids)[1] <= (range)[1] && (range)[2] <= (ids)[2] )
+
+#define MDB_IDL_CPY( dst, src ) (memcpy( dst, src, MDB_IDL_SIZEOF( src ) ))
+
+#define MDB_IDL_ID( bdb, ids, id ) MDB_IDL_RANGE( ids, id, ((bdb)->bi_lastid) )
+#define MDB_IDL_ALL( bdb, ids ) MDB_IDL_RANGE( ids, 1, ((bdb)->bi_lastid) )
+
+#define MDB_IDL_FIRST( ids ) ( (ids)[1] )
+#define MDB_IDL_LAST( ids ) ( MDB_IDL_IS_RANGE(ids) \
+ ? (ids)[2] : (ids)[(ids)[0]] )
+
+#define MDB_IDL_N( ids ) ( MDB_IDL_IS_RANGE(ids) \
+ ? ((ids)[2]-(ids)[1])+1 : (ids)[0] )
+
+int mdb_midl_insert( ID *ids, ID id );
+
+typedef struct MIDL2 {
+ ID mid;
+ void *mptr;
+} MIDL2;
+
+unsigned mdb_midl2_search( MIDL2 *ids, MIDL2 *id );
+int mdb_midl2_insert( MIDL2 *ids, MIDL2 *id );
+
+#endif /* _MDB_MIDL_H_ */
--- /dev/null
+/* mtest.c - memory-mapped database tester/toy */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+#define _XOPEN_SOURCE 500 /* srandom(), random() */
+#include <stdio.h>
+#include <stdlib.h>
+#include <time.h>
+#include "mdb.h"
+
+int main(int argc,char * argv[])
+{
+ int i = 0, j = 0, rc;
+ MDB_env *env;
+ MDB_dbi dbi;
+ MDB_val key, data;
+ MDB_txn *txn;
+ MDB_stat mst;
+ MDB_cursor *cursor;
+ int count;
+ int *values;
+ char sval[32];
+
+ srandom(time(NULL));
+
+ count = (random()%384) + 64;
+ values = (int *)malloc(count*sizeof(int));
+
+ for(i = 0;i<count;i++) {
+ values[i] = random()%1024;
+ }
+
+ rc = mdb_env_create(&env);
+ rc = mdb_env_set_mapsize(env, 10485760);
+ rc = mdb_env_open(env, "./testdb", MDB_FIXEDMAP /*|MDB_NOSYNC*/, 0664);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_open(txn, NULL, 0, &dbi);
+
+ key.mv_size = sizeof(int);
+ key.mv_data = sval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+
+ printf("Adding %d values\n", count);
+ for (i=0;i<count;i++) {
+ sprintf(sval, "%03x %d foo bar", values[i], values[i]);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NOOVERWRITE);
+ if (rc) j++;
+ }
+ if (j) printf("%d duplicates skipped\n", j);
+ rc = mdb_txn_commit(txn);
+ rc = mdb_env_stat(env, &mst);
+
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_txn_abort(txn);
+
+ j=0;
+ key.mv_data = sval;
+ for (i= count - 1; i > -1; i-= (random()%5)) {
+ j++;
+ txn=NULL;
+ rc = mdb_txn_begin(env, 0, &txn);
+ sprintf(sval, "%03x ", values[i]);
+ rc = mdb_del(txn, dbi, &key, NULL);
+ if (rc) {
+ j--;
+ mdb_txn_abort(txn);
+ } else {
+ rc = mdb_txn_commit(txn);
+ }
+ }
+ free(values);
+ printf("Deleted %d values\n", j);
+
+ rc = mdb_env_stat(env, &mst);
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ printf("Cursor next\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ printf("Cursor prev\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_PREV)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+#if 0
+ /* write ops aren't coordinated with cursors,
+ * this stuff all breaks
+ */
+ printf("Deleting with cursor\n");
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_cursor_open(db, txn, &cur2);
+ for (i=0; i<50; i++) {
+ rc = mdb_cursor_get(cur2, &key, &data, MDB_NEXT);
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ rc = mdb_del(db, txn, &key, NULL);
+ }
+
+ printf("Restarting cursor in txn\n");
+ rc = mdb_cursor_get(cur2, &key, &data, MDB_FIRST);
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ for (i=0; i<32; i++) {
+ rc = mdb_cursor_get(cur2, &key, &data, MDB_NEXT);
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cur2);
+ rc = mdb_txn_commit(txn);
+ mdb_cursor_close(cursor);
+ printf("Restarting cursor outside txn\n");
+ rc = mdb_cursor_open(db, NULL, &cursor);
+ rc = mdb_cursor_get(cursor, &key, &data, MDB_FIRST);
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ for (i=0; i<32; i++) {
+ rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT);
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+#endif
+ mdb_cursor_close(cursor);
+ mdb_close(txn, dbi);
+
+ mdb_txn_abort(txn);
+ mdb_env_close(env);
+
+ return 0;
+}
--- /dev/null
+/* mtest2.c - memory-mapped database tester/toy */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+
+/* Just like mtest.c, but using a subDB instead of the main DB */
+
+#define _XOPEN_SOURCE 500 /* srandom(), random() */
+#include <stdio.h>
+#include <stdlib.h>
+#include <time.h>
+#include "mdb.h"
+
+int main(int argc,char * argv[])
+{
+ int i = 0, j = 0, rc;
+ MDB_env *env;
+ MDB_dbi dbi;
+ MDB_val key, data;
+ MDB_txn *txn;
+ MDB_stat mst;
+ MDB_cursor *cursor;
+ int count;
+ int *values;
+ char sval[32];
+
+ srandom(time(NULL));
+
+ count = (random()%384) + 64;
+ values = (int *)malloc(count*sizeof(int));
+
+ for(i = 0;i<count;i++) {
+ values[i] = random()%1024;
+ }
+
+ rc = mdb_env_create(&env);
+ rc = mdb_env_set_mapsize(env, 10485760);
+ rc = mdb_env_set_maxdbs(env, 4);
+ rc = mdb_env_open(env, "./testdb", MDB_FIXEDMAP|MDB_NOSYNC, 0664);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_open(txn, "id1", MDB_CREATE, &dbi);
+
+ key.mv_size = sizeof(int);
+ key.mv_data = sval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+
+ printf("Adding %d values\n", count);
+ for (i=0;i<count;i++) {
+ sprintf(sval, "%03x %d foo bar", values[i], values[i]);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NOOVERWRITE);
+ if (rc) j++;
+ }
+ if (j) printf("%d duplicates skipped\n", j);
+ rc = mdb_txn_commit(txn);
+ rc = mdb_env_stat(env, &mst);
+
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_txn_abort(txn);
+
+ j=0;
+ key.mv_data = sval;
+ for (i= count - 1; i > -1; i-= (random()%5)) {
+ j++;
+ txn=NULL;
+ rc = mdb_txn_begin(env, 0, &txn);
+ sprintf(sval, "%03x ", values[i]);
+ rc = mdb_del(txn, dbi, &key, NULL);
+ if (rc) {
+ j--;
+ mdb_txn_abort(txn);
+ } else {
+ rc = mdb_txn_commit(txn);
+ }
+ }
+ free(values);
+ printf("Deleted %d values\n", j);
+
+ rc = mdb_env_stat(env, &mst);
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ printf("Cursor next\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ printf("Cursor prev\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_PREV)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_close(txn, dbi);
+
+ mdb_txn_abort(txn);
+ mdb_env_close(env);
+
+ return 0;
+}
--- /dev/null
+/* mtest3.c - memory-mapped database tester/toy */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+
+/* Tests for sorted duplicate DBs */
+#define _XOPEN_SOURCE 500 /* srandom(), random() */
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <time.h>
+#include "mdb.h"
+
+int main(int argc,char * argv[])
+{
+ int i = 0, j = 0, rc;
+ MDB_env *env;
+ MDB_dbi dbi;
+ MDB_val key, data;
+ MDB_txn *txn;
+ MDB_stat mst;
+ MDB_cursor *cursor;
+ int count;
+ int *values;
+ char sval[32];
+ char kval[sizeof(int)];
+
+ srandom(time(NULL));
+
+ memset(sval, 0, sizeof(sval));
+
+ count = (random()%384) + 64;
+ values = (int *)malloc(count*sizeof(int));
+
+ for(i = 0;i<count;i++) {
+ values[i] = random()%1024;
+ }
+
+ rc = mdb_env_create(&env);
+ rc = mdb_env_set_mapsize(env, 10485760);
+ rc = mdb_env_set_maxdbs(env, 4);
+ rc = mdb_env_open(env, "./testdb", MDB_FIXEDMAP|MDB_NOSYNC, 0664);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_open(txn, "id2", MDB_CREATE|MDB_DUPSORT, &dbi);
+
+ key.mv_size = sizeof(int);
+ key.mv_data = kval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+
+ printf("Adding %d values\n", count);
+ for (i=0;i<count;i++) {
+ if (!(i & 0x0f))
+ sprintf(kval, "%03x", values[i]);
+ sprintf(sval, "%03x %d foo bar", values[i], values[i]);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NODUPDATA);
+ if (rc) j++;
+ }
+ if (j) printf("%d duplicates skipped\n", j);
+ rc = mdb_txn_commit(txn);
+ rc = mdb_env_stat(env, &mst);
+
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_txn_abort(txn);
+
+ j=0;
+
+ for (i= count - 1; i > -1; i-= (random()%5)) {
+ j++;
+ txn=NULL;
+ rc = mdb_txn_begin(env, 0, &txn);
+ sprintf(kval, "%03x", values[i & ~0x0f]);
+ sprintf(sval, "%03x %d foo bar", values[i], values[i]);
+ key.mv_size = sizeof(int);
+ key.mv_data = kval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+ rc = mdb_del(txn, dbi, &key, &data);
+ if (rc) {
+ j--;
+ mdb_txn_abort(txn);
+ } else {
+ rc = mdb_txn_commit(txn);
+ }
+ }
+ free(values);
+ printf("Deleted %d values\n", j);
+
+ rc = mdb_env_stat(env, &mst);
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ printf("Cursor next\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ printf("Cursor prev\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_PREV)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_close(txn, dbi);
+
+ mdb_txn_abort(txn);
+ mdb_env_close(env);
+
+ return 0;
+}
--- /dev/null
+/* mtest4.c - memory-mapped database tester/toy */
+/*
+ * Copyright 2011 Howard Chu, Symas Corp.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted only as authorized by the OpenLDAP
+ * Public License.
+ *
+ * A copy of this license is available in the file LICENSE in the
+ * top-level directory of the distribution or, alternatively, at
+ * <http://www.OpenLDAP.org/license.html>.
+ */
+
+/* Tests for sorted duplicate DBs with fixed-size keys */
+#define _XOPEN_SOURCE 500 /* srandom(), random() */
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <time.h>
+#include "mdb.h"
+
+int main(int argc,char * argv[])
+{
+ int i = 0, j = 0, rc;
+ MDB_env *env;
+ MDB_dbi dbi;
+ MDB_val key, data;
+ MDB_txn *txn;
+ MDB_stat mst;
+ MDB_cursor *cursor;
+ int count;
+ int *values;
+ char sval[8];
+ char kval[sizeof(int)];
+
+ memset(sval, 0, sizeof(sval));
+
+ count = 510;
+ values = (int *)malloc(count*sizeof(int));
+
+ for(i = 0;i<count;i++) {
+ values[i] = i*5;
+ }
+
+ rc = mdb_env_create(&env);
+ rc = mdb_env_set_mapsize(env, 10485760);
+ rc = mdb_env_set_maxdbs(env, 4);
+ rc = mdb_env_open(env, "./testdb", MDB_FIXEDMAP|MDB_NOSYNC, 0664);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_open(txn, "id2", MDB_CREATE|MDB_DUPSORT|MDB_DUPFIXED, &dbi);
+
+ key.mv_size = sizeof(int);
+ key.mv_data = kval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+
+ printf("Adding %d values\n", count);
+ strcpy(kval, "001");
+ for (i=0;i<count;i++) {
+ sprintf(sval, "%07x", values[i]);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NODUPDATA);
+ if (rc) j++;
+ }
+ if (j) printf("%d duplicates skipped\n", j);
+ rc = mdb_txn_commit(txn);
+ rc = mdb_env_stat(env, &mst);
+
+ /* there should be one full page of dups now.
+ */
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %p %.*s, data: %p %.*s\n",
+ key.mv_data, (int) key.mv_size, (char *) key.mv_data,
+ data.mv_data, (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_txn_abort(txn);
+
+ /* test all 3 branches of split code:
+ * 1: new key in lower half
+ * 2: new key at split point
+ * 3: new key in upper half
+ */
+
+ key.mv_size = sizeof(int);
+ key.mv_data = kval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+
+ sprintf(sval, "%07x", values[3]+1);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NODUPDATA);
+ mdb_txn_abort(txn);
+
+ sprintf(sval, "%07x", values[255]+1);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NODUPDATA);
+ mdb_txn_abort(txn);
+
+ sprintf(sval, "%07x", values[500]+1);
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_put(txn, dbi, &key, &data, MDB_NODUPDATA);
+ rc = mdb_txn_commit(txn);
+
+ /* Try MDB_NEXT_MULTIPLE */
+ rc = mdb_txn_begin(env, 0, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT_MULTIPLE)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_txn_abort(txn);
+ j=0;
+
+ for (i= count - 1; i > -1; i-= (random()%3)) {
+ j++;
+ txn=NULL;
+ rc = mdb_txn_begin(env, 0, &txn);
+ sprintf(sval, "%07x", values[i]);
+ key.mv_size = sizeof(int);
+ key.mv_data = kval;
+ data.mv_size = sizeof(sval);
+ data.mv_data = sval;
+ rc = mdb_del(txn, dbi, &key, &data);
+ if (rc) {
+ j--;
+ mdb_txn_abort(txn);
+ } else {
+ rc = mdb_txn_commit(txn);
+ }
+ }
+ free(values);
+ printf("Deleted %d values\n", j);
+
+ rc = mdb_env_stat(env, &mst);
+ rc = mdb_txn_begin(env, 1, &txn);
+ rc = mdb_cursor_open(txn, dbi, &cursor);
+ printf("Cursor next\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_NEXT)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ printf("Cursor prev\n");
+ while ((rc = mdb_cursor_get(cursor, &key, &data, MDB_PREV)) == 0) {
+ printf("key: %.*s, data: %.*s\n",
+ (int) key.mv_size, (char *) key.mv_data,
+ (int) data.mv_size, (char *) data.mv_data);
+ }
+ mdb_cursor_close(cursor);
+ mdb_close(txn, dbi);
+
+ mdb_txn_abort(txn);
+ mdb_env_close(env);
+
+ return 0;
+}