]> git.sur5r.net Git - bacula/docs/blobdiff - docs/manuals/de/old/utility/fix_tex.pl
Reset everything to English
[bacula/docs] / docs / manuals / de / old / utility / fix_tex.pl
diff --git a/docs/manuals/de/old/utility/fix_tex.pl b/docs/manuals/de/old/utility/fix_tex.pl
new file mode 100755 (executable)
index 0000000..9865757
--- /dev/null
@@ -0,0 +1,184 @@
+#!/usr/bin/perl -w
+# Fixes various things within tex files.
+
+use strict;
+
+my %args;
+
+
+sub get_includes {
+       # Get a list of include files from the top-level tex file.
+       my (@list,$file);
+       
+       foreach my $filename (@_) {
+               $filename or next;
+               # Start with the top-level latex file so it gets checked too.
+               push (@list,$filename);
+
+               # Get a list of all the html files in the directory.
+               open IF,"<$filename" or die "Cannot open input file $filename";
+               while (<IF>) {
+                       chomp;
+                       push @list,"$1.tex" if (/\\include\{(.*?)\}/);
+               }
+
+               close IF;
+       }
+       return @list;
+}
+
+sub convert_files {
+       my (@files) = @_;
+       my ($linecnt,$filedata,$output,$itemcnt,$indentcnt,$cnt);
+       
+       $cnt = 0;
+       foreach my $file (@files) {
+               # Open the file and load the whole thing into $filedata. A bit wasteful but
+               #   easier to deal with, and we don't have a problem with speed here.
+               $filedata = "";
+               open IF,"<$file" or die "Cannot open input file $file";
+               while (<IF>) {
+                       $filedata .= $_;
+               }
+               close IF;
+               
+               # We look for a line that starts with \item, and indent the two next lines (if not blank)
+               #  by three spaces.
+               my $linecnt = 3;
+               $indentcnt = 0;
+               $output = "";
+               # Process a line at a time.
+               foreach (split(/\n/,$filedata)) {
+                       $_ .= "\n"; # Put back the return.
+                       # If this line is less than the third line past the \item command,
+                       #  and the line isn't blank and doesn't start with whitespace
+                       #  add three spaces to the start of the line. Keep track of the number
+                       #  of lines changed.
+                       if ($linecnt < 3 and !/^\\item/) {
+                               if (/^[^\n\s]/) {
+                                       $output .= "   " . $_;
+                                       $indentcnt++;
+                               } else {
+                                       $output .= $_;
+                               }
+                               $linecnt++;
+                       } else {
+                               $linecnt = 3;
+                               $output .= $_;
+                       }
+                       /^\\item / and $linecnt = 1;
+               }
+
+               
+               # This is an item line.  We need to process it too. If inside a \begin{description} environment, convert 
+               #  \item {\bf xxx} to \item [xxx] or \item [{xxx}] (if xxx contains '[' or ']'.
+               $itemcnt = 0;
+               $filedata = $output;
+               $output = "";
+               my ($before,$descrip,$this,$between);
+
+               # Find any \begin{description} environment
+               while ($filedata =~ /(\\begin[\s\n]*\{[\s\n]*description[\s\n]*\})(.*?)(\\end[\s\n]*\{[\s\n]*description[\s\n]*\})/s) {
+                       $output .= $` . $1;
+                       $filedata = $3 . $';
+                       $descrip = $2;
+
+                       # Search for \item {\bf xxx}
+                       while ($descrip =~ /\\item[\s\n]*\{[\s\n]*\\bf[\s\n]*/s) {
+                               $descrip = $';
+                               $output .= $`;
+                               ($between,$descrip) = find_matching_brace($descrip);
+                               if (!$descrip) {
+                                       $linecnt = $output =~ tr/\n/\n/;
+                                       print STDERR "Missing matching curly brace at line $linecnt in $file\n" if (!$descrip);
+                               }
+
+                               # Now do the replacement.
+                               $between = '{' . $between . '}' if ($between =~ /\[|\]/);
+                               $output .= "\\item \[$between\]";       
+                               $itemcnt++;
+                       }
+                       $output .= $descrip;
+               }
+               $output .= $filedata;
+       
+               # If any hyphens or \item commnads were converted, save the file.
+               if ($indentcnt or $itemcnt) {
+                       open OF,">$file" or die "Cannot open output file $file";
+                       print OF $output;
+                       close OF;
+                       print "$indentcnt indent", ($indentcnt == 1) ? "" : "s"," added in $file\n";
+                       print "$itemcnt item", ($itemcnt == 1) ? "" : "s"," Changed in $file\n";
+               }
+
+               $cnt += $indentcnt + $itemcnt;
+       }
+       return $cnt;
+}
+
+sub find_matching_brace {
+       # Finds text up to the next matching brace.  Assumes that the input text doesn't contain
+       #  the opening brace, but we want to find text up to a matching closing one.
+       # Returns the text between the matching braces, followed by the rest of the text following
+       #   (which does not include the matching brace).
+       # 
+       my $str = shift;
+       my ($this,$temp);
+       my $cnt = 1;
+
+       while ($cnt) {
+               # Ignore verbatim constructs involving curly braces, or if the character preceding
+               #  the curly brace is a backslash.
+               if ($str =~ /\\verb\*?\{.*?\{|\\verb\*?\}.*?\}|\{|\}/s) {
+                       $this .= $`;
+                       $str = $';
+                       $temp = $&;
+
+                       if ((substr($this,-1,1) eq '\\') or 
+                               $temp =~ /^\\verb/) {
+                                       $this .= $temp;
+                                       next;
+                       }
+                               
+                       $cnt +=  ($temp eq '{') ? 1 : -1;
+                       # If this isn't the matching curly brace ($cnt > 0), include the brace.
+                       $this .= $temp if ($cnt);
+               } else {
+                       # No matching curly brace found.
+                       return ($this . $str,'');
+               }
+       }
+       return ($this,$str);
+}
+
+sub check_arguments {
+       # Checks command-line arguments for ones starting with --  puts them into
+       #   a hash called %args and removes them from @ARGV.
+       my $args = shift;
+       my $i;
+
+       for ($i = 0; $i < $#ARGV; $i++) {
+               $ARGV[$i] =~ /^\-+/ or next;
+               $ARGV[$i] =~ s/^\-+//;
+               $args{$ARGV[$i]} = "";
+               delete ($ARGV[$i]);
+               
+       }
+}
+
+##################################################################
+#                       MAIN                                  ####
+##################################################################
+
+my @includes;
+my $cnt;
+
+check_arguments(\%args);
+die "No Files given to Check\n" if ($#ARGV < 0);
+
+# Examine the file pointed to by the first argument to get a list of 
+#  includes to test.
+@includes = get_includes(@ARGV);
+
+$cnt = convert_files(@includes);
+print "No lines changed\n" unless $cnt;