texdirflatten (27jun17)

git-svn-id: svn://tug.org/texlive/trunk@44713 c570f23f-e606-0410-a88d-b1316a301751

1 files changed, 156 insertions, 70 deletions

diff --git a/Build/source/texk/texlive/linked_scripts/texdirflatten/texdirflatten b/Build/source/texk/texlive/linked_scripts/texdirflatten/texdirflatten
index c43b489f18e..f67adc77dc6 100644
--- a/Build/source/texk/texlive/linked_scripts/texdirflatten/texdirflatten
+++ b/Build/source/texk/texlive/linked_scripts/texdirflatten/texdirflatten
@@ -1,4 +1,9 @@
-#! /usr/bin/perl
+#! /usr/bin/env perl
+
+# Use:
+# $ perldoc texdirflatten
+# to see embedded documentation. Alternatively, you can create manual
+# or HTML pages using pod2man and pod2html.
 
 =pod
 
@@ -9,7 +14,7 @@ single output directory -- i.e., flattens its hierarchy.
 
 =head1 SYNOPSIS
 
-texdirflatten -f input.tex [-o outputdir]
+texdirflatten [-1|--onetex] [-f|--file input.tex] [-o outputdir] [-?|--help]
 
 =head1 DESCRIPTION
 
@@ -24,16 +29,25 @@ Run without parameters to see usage.
 
 =over
 
-=item B<-f> I<input.tex>
+=item B<--file>, B<-f> I<input.tex>
 
 Specifies input (La)TeX file.
 
-=item B<-o> I<outputdir>
+=item B<--onetex>, B<-1>
+
+If specified, produces a single TeX file by expanding all \input and
+\include commands in place.
+
+=item B<--output>, B<-o> I<outputdir>
 
 Directory to collect all files. B<texdirflatten> will copy each source
 file, graphics and bibliography file to this directory. It will be
 created if it is unexistent. If unspecified, it defaults to C<flat/>.
 
+=item B<--help>, B<-?>
+
+Show this manual page.
+
 =back
 
 =head1 EXAMPLES
@@ -48,6 +62,10 @@ and gathers it and all its components in the C<submit_01/> directory:
 Please take backups before running this command. No warranties
 whatsoever provided.
 
+You may need to run C<epstopdf> on EPS files if you are using C<pdflatex>:
+
+ $ for i in *.eps; do epstopdf $i; done
+
 =head1 BUGS
 
 Bug reports and patches are welcome.
@@ -58,15 +76,13 @@ Cengiz Gunay <cengique<AT>users.sf.net>
 
 =head1 COPYRIGHT AND LICENSE
 
-Copyleft 2003-2009, Cengiz Gunay
+Copyleft 2003-2017, Cengiz Gunay
 
 This library is free software; you may redistribute it and/or modify
 it under the same terms as Perl itself.
 
 =cut
 
-# $Id: texdirflatten,v 1.1 2009/04/17 14:15:39 cengiz Exp $
-
 # TODO:
 #
 # - parse BIBINPUTS environment variable and search figures and such
@@ -77,10 +93,13 @@ package texdirflatten;
 use Getopt::Long qw(GetOptions);
 use Pod::Usage;
 use strict;
+use warnings;
+use re 'eval';
 
 # Global
-my $outputdir   = "flat";
-my $file;
+my $outputdir   = "flat";	# Output directory
+my $file = "";			# File to read
+my $onetex = 0;			# Indicates all input and include commands be expanded
 my $help;
 
 # parse helpers
@@ -88,35 +107,41 @@ my %texfiles;
 
 # GetOpt::Long::
 my $result = GetOptions ("output|o=s" => \$outputdir,
-				       "file|f=s" => \$file,
-				       "help|?" => \$help );
+			 "onetex|1" => \$onetex,
+			 "file|f=s" => \$file,
+			 "help|?" => \$help );
 
+# regexp for none or an even number of backslashes (\):
+my $unescaped =
+      '(?<! \\\\)(?(?= \\\\) (?: \\\\\\\\)* )?';
 
-#print "help: $help, file: $file\n";
+# counter for files translated (like LyX does)
+$::filecount = 0;
+my $DEBUG = 1;
 
-#print "poop: " . popfile('slkjs/lkjlsjd/saa.abc');
-#{
-#  print ;
-#}
+pod2usage( -verbose => 2 ) if ($help);
 
-pod2usage( -section => "SYNOPSIS") if ($file eq "" || $help);
+pod2usage( -section => "SYNOPSIS") if ($file eq "");
 
 system "mkdir $outputdir" if (not -d $outputdir);
 
 # start recursing
-parseTeX($file, "");
+parseTeX($file, "", popfile($file), 0);
 
 sub parseTeX {
   my $file = shift;
   my $inputdir = shift;
+  my $outfile = "$outputdir/" . shift;
+  my $level = shift;
+
+  # Add a proper TeX suffix if it's missing
+  $outfile .= '.tex' if ! ($outfile =~ /\.tex\s*$/i );
 
   my @flats;
   my @longs;
   my $popped;
 
   open my $FILE, $file or die "Cannot find file to read $file\n";
-  my $outfile = "$outputdir/" . popfile($file);
-  open my $COPY, ">$outfile" or die "Cannot write file $outfile\n";
 
   # Read to whole file first, and then scan for regexps
   my $contents = "";
@@ -125,71 +150,70 @@ sub parseTeX {
   }
   close $FILE;
 
+  ## First, remove comments so that input commands inside are ignored.
+  # Also remove trailing newline and all whitespace at the beginning
+  # of the following line like TeX does it. Only for %'s following
+  # the proper backslash pattern.
+  $contents =~ s/( $unescaped )\%.*?\n[ \t]*/$1/gx;
+
   # Look for graphics include statement
-  if ($contents =~ /\\input\@path{{([^}]*)}}/) {
+  if ($contents =~ /\\input\@path\{\{([^}]*)\}\}/) {
     $inputdir = $1;
     print "Found input directory: $inputdir\n";
   }
 
+  print "Parsing '$file' in directory '$inputdir'\n";
+
   # Default value
   #$inputdir = './' if (-z $inputdir);
 
   # three cases: graphics, inputs and bibs
 
   # an \includegraphics statement
-  @flats = ();
-  @longs = ();
+  @::flats = ();
+  @::longs = ();
+  #$::popped = "";
 
-  $contents =~ 
-    s/\\includegraphics(\[[^\]]*\])?\{([^}]*)\}/"\\includegraphics" . $1 .
-  "{" . ($popped = popfile($2) and push @longs, $2 and push @flats,$popped and $popped) . "}"/egm;
+  $contents =~
+    s/\\includegraphics(\[[^\]]*\])?\{([^}]*)\}
+      (?{ $::popped = flattenfilename($2);
+	push @::longs, $2; push @::flats, $::popped })
+     /\\includegraphics$1\{$::popped\}/gmx;
 
-  if ($#flats > -1) {
+  if ($#::flats > -1) {
     my ($long, $flat);
 
-    foreach $long (@longs) {
-      $flat = shift @flats;
-      print "Found graphics: '$long'\n";
+    foreach $long (@::longs) {
+      $flat = shift @::flats;
 
       # convert LyX directory dots
       $long =~ s/\\lyxdot /./g;
+      $flat =~ s/\\lyxdot /./g;
 
-      my $epsfile = $inputdir . $long;
+      print "Looking for graphics: '$long'\n";
 
-      if (system("cp $epsfile.eps $outputdir/$flat.eps") != 0) {
-	print "Cannot find $epsfile.eps!\n";
-	# if from figures dir, copy the EPS file as well
-	if ($long =~ /figures\//) {
-	  $epsfile = $inputdir . $long;
-	} else {
-	  $epsfile = $inputdir . '../figures/' . $long;
-	}
+      my @exts = ("", ".eps", ".pdf", ".pstex");
+      my @dirs = ("", "$inputdir", $inputdir . '../figures/');
 
-	if (system("cp $epsfile.eps $outputdir/$flat.eps") != 0) {
-	  print "Cannot find $epsfile.eps!\n";
-	  if (system("cp $epsfile.pstex $outputdir/$flat.eps") != 0) {
-	    die "Cannot find figure file '$epsfile'";
-	  } else {
-	    print "Found: '$epsfile.pstex'\n";
-	  }
-	}
-      } else {
-	print "Found: '$epsfile.pstex'\n";
-      }
+      seekfile($long, $flat, ["", ".eps", ".pdf", ".pstex"], 
+	       ["", "$inputdir", $inputdir . '../figures/']);
 
     }
-    my $long = $2;
-
   }
 
+  # Must do bibs before input commands because, when expanded, they
+  # add new bibs inside.
+  replacebibs(\$contents, $inputdir);
+
   # an \input or \include statement
   @flats = ();
   @longs = ();
-  $contents =~ 
-    s/\\(input|include){([^}]*)}/"\\${1}{" .
-  ($popped = popfile($2) and push @longs, $2 and push @flats,$popped and $popped)  . "}"/egm;
+  $contents =~
+    s/\\(input|include)\{([^}]*)\}/"\\${1}\{" .
+      ($popped = flattenfilename($2) and push @longs, $2 and push @flats,$popped and $popped)
+	. "\}"/egm;
 
-  if ($#flats > 0) {
+  if ($#flats > -1) {
     #print "Found " . scalar @longs . " items on line: '@flats'.\n";
 
     my ($long, $flat);
@@ -208,24 +232,60 @@ sub parseTeX {
 	  }
 	}
 	$texfiles{$flat} = $long;
-	parseTeX($long, $inputdir); #later
-	
+	my $newcontents = parseTeX($long, basedir($long), $flat, $level + 1); #later
+	if ( $onetex ) {
+	  # If specified, expand inclusion directives
+	  die "Failed to find input/include to expand for '$flat'!\n"
+	    if (! ( $contents =~
+		s/\\(input|include) \{ \Q$flat\E \} 
+                  (?{ print "===Found \\$1\{$flat\} to expand.\n" if $DEBUG; })
+		  / ($1 eq 'include' ? "\\clearpage\n" : '') . "$newcontents"/gxem ) );
+	}
       }
     }
   }
 
-  replacebibs(\$contents, $inputdir);
   #print "Writing to \"$outfile\"\n" .
   #  "----------------------------------------\n" .
   #  "$contents\n" .
   #  "----------------------------------------\n";
 
-  # write to flat copy
-  print $COPY $contents;
+  if (! $onetex || $level == 0) {  # do output at zero recurse level
+    # open flat output file
+    open my $COPY, ">$outfile" or die "Cannot write file $outfile\n";
 
+    # write to flat copy
+    print $COPY $contents;
 
-  close $COPY;
+    close $COPY;
+  } else {
+    return $contents;
+  }
+}
 
+# Look for file to copy
+sub seekfile {
+    my ($file, $dest, $exts, $dirs) = @_;
+    my @exts = @$exts;
+    my @dirs = @$dirs;
+    
+  OUT: foreach my $dir (@dirs) {
+      foreach my $ext (@exts) {
+	  my $tryfile = "$dir$file$ext";
+	  if (-r $tryfile) {
+	      # add the extension to $dest only if it hasn't any dots
+	      my $dext = '';
+	      $dext = $ext if (! $dest =~ /\./); 
+	      
+	      if (system("cp $tryfile $outputdir/$dest$dext") != 0) {
+		  die "Cannot copy $tryfile!\n";
+	      }
+	      # If we found it, then get out of the loop
+	      print "Found: '$tryfile'\n";
+	      last OUT;
+	  }
+      }
+  }
 }
 
 # Return everything after the last /
@@ -238,6 +298,35 @@ sub popfile {
   pop @a;
 }
 
+# Make a new file by flattening directory names
+sub flattenfilename {
+  local $_ = shift;
+
+  # convert slashes to pluses, dots to Xs
+  tr|/|+|;
+  tr|\.|X|;
+
+  # add a unique number at the beginning so files don't start with a
+  # dot
+  $_ = sprintf("%03d_", ++$::filecount) . "$_";
+
+  print "===Flattened file to '$_'\n" if $DEBUG;
+
+  return $_;
+}
+
+sub basedir {
+  my $file = shift (@_);
+
+  my @a = split /\//, $file;
+
+  #print "Popping $file... split: @a ($#a), pop: " . pop(@a) . "\n";
+  pop @a;			# remove file part
+  my $basedir = join('/', @a);		# combine the rest with /'s
+  $basedir .= '/' if (! $basedir =~ /^$/);		# add a trailing / if not empty
+  $basedir;
+}
+
 # get the bibliography files
 sub replacebibs {
   my ($contentsref, $inputdir) = @_;
@@ -253,13 +342,13 @@ sub replacebibs {
   $" = ',';			# set list item separator
   # search for \bibliography statements
   $$contentsref =~
-    s/\\bibliography(\[[^\]]*\])? { 
+    s/\\bibliography(\[[^\]]*\])? {
       (?{ @::refs = (); #print "Found bibliography!\n"; #initialize ref list
         }) (?: \s* ([^}, ]+)(?=[,}]),?
       (?{ # for each word do:
 	  #local (@_flats, @_longs, @_refs);
           #print "Word: \"$+\", "; 
-          $popped = popfile($+); push @::longs, $+;
+          $popped = flattenfilename($+); push @::longs, $+;
           push @::flats,$popped; push @::refs, $popped;
         }) )+ \s* }
       (?{ #print "\n+++ $#::flats; $#::refs; $#::longs\n";
@@ -270,7 +359,7 @@ sub replacebibs {
       #print "\n*** $#::flats; $#::refs; $#::longs\n";
   $" = ' ';			# restore list item separator
 
-  if ($#::flats > 0) {
+  if ($#::flats > -1) {
     #print "Found " . scalar @::longs . " items on line: '@::flats'.\n";
 
     my ($long, $flat);
@@ -283,11 +372,8 @@ sub replacebibs {
       # convert LyX directory dots
       $long =~ s/\\lyxdot /./g;
 
-      my $bibfile = $inputdir . $long;
-
-      if (system("cp $bibfile.bib $outputdir/$flat.bib") != 0) {
-	die "Cannot find $bibfile.bib!\n";
-      }
+      seekfile($long, "$flat.bib", [".bib"], 
+	       ["", "$inputdir"]);
     }
   }
 }