| 1 |
#! /usr/bin/env perl |
| 2 |
|
| 3 |
# Use: |
| 4 |
# $ perldoc texdirflatten |
| 5 |
# to see embedded documentation. Alternatively, you can create manual |
| 6 |
# or HTML pages using pod2man and pod2html. |
| 7 |
|
| 8 |
=pod |
| 9 |
|
| 10 |
=head1 NAME |
| 11 |
|
| 12 |
B<texdirflatten> - Collects all components of a (La)TeX file in a |
| 13 |
single output directory -- i.e., flattens its hierarchy. |
| 14 |
|
| 15 |
=head1 SYNOPSIS |
| 16 |
|
| 17 |
texdirflatten [-1|--onetex] [-f|--file input.tex] [-o outputdir] |
| 18 |
[--imgexts .ext1,.ext2,...] [--debug] [-V|--version] [-?|--help] |
| 19 |
|
| 20 |
=head1 DESCRIPTION |
| 21 |
|
| 22 |
This Perl script parses a LaTeX file recursively, scanning all child |
| 23 |
files, and collects details of any included and other data files, such |
| 24 |
as graphics and BiBTeX bibliography files. These component files, are |
| 25 |
then all put into a single directory (thus "flattening" the document's |
| 26 |
directory tree). This is useful in distributing manuscripts to |
| 27 |
collaborators or in submitting to journals. |
| 28 |
|
| 29 |
=for comment |
| 30 |
Run without parameters to see usage. |
| 31 |
|
| 32 |
=head1 OPTIONS |
| 33 |
|
| 34 |
=over |
| 35 |
|
| 36 |
=item B<--file>, B<-f> I<input.tex> |
| 37 |
|
| 38 |
Specifies input (La)TeX file. |
| 39 |
|
| 40 |
=item B<--onetex>, B<-1> |
| 41 |
|
| 42 |
If specified, produces a single TeX file by expanding all \input and |
| 43 |
\include commands in place. |
| 44 |
|
| 45 |
=item B<--output>, B<-o> I<outputdir> |
| 46 |
|
| 47 |
Directory to collect all files. B<texdirflatten> will copy each source |
| 48 |
file, graphics and bibliography file to this directory. It will be |
| 49 |
created if it is unexistent. If unspecified, it defaults to C<flat/>. |
| 50 |
|
| 51 |
=item B<--imgexts> I<.ext1,.ext2[,...]> |
| 52 |
|
| 53 |
Prepends to the prioritized list of image extensions to search when |
| 54 |
trying to find the image to copy to output folder. The first file with |
| 55 |
the extension found will be copied and search stopped. Default order |
| 56 |
is C<"", ".eps", ".pdf", ".pstex"> (note that it includes files with |
| 57 |
no extension). If using pdflatex, one may want to add PDF before EPS |
| 58 |
with C<--imgexts .pdf>. Don't forget the dot before the extension! |
| 59 |
|
| 60 |
=item B<--debug> |
| 61 |
|
| 62 |
Enables copious amounts of debugging output - useful if something is going wrong. |
| 63 |
|
| 64 |
=item B<--version>, B<-V> |
| 65 |
|
| 66 |
Displays the current version number and the usage and exits. |
| 67 |
|
| 68 |
=item B<--help>, B<-?> |
| 69 |
|
| 70 |
Show this manual page. |
| 71 |
|
| 72 |
=back |
| 73 |
|
| 74 |
=head1 EXAMPLES |
| 75 |
|
| 76 |
The following example scans C<manuscript.tex> in the current directory |
| 77 |
and gathers it and all its components in the C<submit_01/> directory: |
| 78 |
|
| 79 |
$ texdirflatten -f manuscript.tex -o submit_01 |
| 80 |
|
| 81 |
=head1 CAVEATS |
| 82 |
|
| 83 |
Please take backups before running this command. No warranties |
| 84 |
whatsoever provided. |
| 85 |
|
| 86 |
You may need to run C<epstopdf> on EPS files if you are using C<pdflatex>: |
| 87 |
|
| 88 |
$ for i in *.eps; do epstopdf $i; done |
| 89 |
|
| 90 |
=head1 BUGS |
| 91 |
|
| 92 |
Bug reports and patches are welcome. |
| 93 |
|
| 94 |
=head1 AUTHOR |
| 95 |
|
| 96 |
Cengiz Gunay <cengique<AT>users.sf.net> |
| 97 |
|
| 98 |
=head1 COPYRIGHT AND LICENSE |
| 99 |
|
| 100 |
Copyleft 2003-2017, Cengiz Gunay |
| 101 |
|
| 102 |
This library is free software; you may redistribute it and/or modify |
| 103 |
it under the same terms as Perl itself. |
| 104 |
|
| 105 |
=cut |
| 106 |
|
| 107 |
# TODO: |
| 108 |
# - parse BIBINPUTS environment variable and search figures and such |
| 109 |
# there as well. |
| 110 |
# - also copy bibtex style file |
| 111 |
|
| 112 |
package texdirflatten; |
| 113 |
|
| 114 |
use Getopt::Long qw(GetOptions); |
| 115 |
use Pod::Usage; |
| 116 |
use strict; |
| 117 |
use warnings; |
| 118 |
use re 'eval'; |
| 119 |
|
| 120 |
# Global |
| 121 |
my $outputdir = "flat"; # Output directory |
| 122 |
my $file = ""; # File to read |
| 123 |
my @imgexts = ("", ".eps", ".pdf", ".pstex"); # Ordered list of image extensions to search and copy |
| 124 |
my $imgexts_pre; # Prepend to @imgexts |
| 125 |
my $onetex = 0; # Indicates all input and include commands be expanded |
| 126 |
my $version = "1.3"; # 2017-02-02 |
| 127 |
my $DEBUG = 0; # Enable debugging output |
| 128 |
my $help; |
| 129 |
|
| 130 |
# parse helpers |
| 131 |
my %texfiles; |
| 132 |
|
| 133 |
# GetOpt::Long:: |
| 134 |
my $result = GetOptions ("output|o=s" => \$outputdir, |
| 135 |
"onetex|1" => \$onetex, |
| 136 |
"file|f=s" => \$file, |
| 137 |
"imgexts=s" => \$imgexts_pre, |
| 138 |
"debug" => \$DEBUG, |
| 139 |
"version|V" => sub { print "texdirflatten version $version\n"; |
| 140 |
pod2usage( -section => "SYNOPSIS"); }, |
| 141 |
"help|?" => \$help ); |
| 142 |
|
| 143 |
# regexp for none or an even number of backslashes (\): |
| 144 |
my $unescaped = |
| 145 |
'(?<! \\\\)(?(?= \\\\) (?: \\\\\\\\)* )?'; |
| 146 |
|
| 147 |
# counter for files translated (like LyX does) |
| 148 |
$::filecount = 0; |
| 149 |
|
| 150 |
pod2usage( -verbose => 2 ) if ($help); |
| 151 |
|
| 152 |
pod2usage( -section => "SYNOPSIS") if ($file eq ""); |
| 153 |
|
| 154 |
if ($imgexts_pre) { |
| 155 |
unshift(@imgexts, split(/,/, $imgexts_pre)); |
| 156 |
print "imgexts: ". join(",", @imgexts) . "\n" if $DEBUG; |
| 157 |
} |
| 158 |
|
| 159 |
system "mkdir $outputdir" if (not -d $outputdir); |
| 160 |
|
| 161 |
# start recursing |
| 162 |
parseTeX($file, "", popfile($file), 0); |
| 163 |
|
| 164 |
sub parseTeX { |
| 165 |
my $file = shift; |
| 166 |
my $inputdir = shift; |
| 167 |
my $outfile = "$outputdir/" . shift; |
| 168 |
my $level = shift; |
| 169 |
|
| 170 |
# Add a proper TeX suffix if it's missing |
| 171 |
$outfile .= '.tex' if ! ($outfile =~ /\.tex\s*$/i ); |
| 172 |
|
| 173 |
my @flats; |
| 174 |
my @longs; |
| 175 |
my $popped; |
| 176 |
|
| 177 |
open my $FILE, $file or die "Cannot find file to read $file\n"; |
| 178 |
|
| 179 |
# Read to whole file first, and then scan for regexps |
| 180 |
my $contents = ""; |
| 181 |
while(<$FILE>) { |
| 182 |
$contents = $contents . $_; |
| 183 |
} |
| 184 |
close $FILE; |
| 185 |
|
| 186 |
## First, remove comments so that input commands inside are ignored. |
| 187 |
# Also remove trailing newline and all whitespace at the beginning |
| 188 |
# of the following line like TeX does it. Only for %'s following |
| 189 |
# the proper backslash pattern. |
| 190 |
$contents =~ s/( $unescaped )\%.*?\n[ \t]*/$1/gx; |
| 191 |
|
| 192 |
# Look for graphics include statement |
| 193 |
if ($contents =~ /\\input\@path\{\{([^}]*)\}\}/) { |
| 194 |
$inputdir = $1; |
| 195 |
print "Found input directory: $inputdir\n" if $DEBUG; |
| 196 |
} |
| 197 |
|
| 198 |
print "Parsing '$file' in directory '$inputdir'\n"; |
| 199 |
|
| 200 |
# Default value |
| 201 |
#$inputdir = './' if (-z $inputdir); |
| 202 |
|
| 203 |
# three cases: graphics, inputs and bibs |
| 204 |
|
| 205 |
# an \includegraphics statement |
| 206 |
@::flats = (); |
| 207 |
@::longs = (); |
| 208 |
#$::popped = ""; |
| 209 |
|
| 210 |
$contents =~ |
| 211 |
s/\\includegraphics(\[[^\]]*\])?\{([^}]*)\} |
| 212 |
(?{ $::popped = flattenfilename($2); |
| 213 |
push @::longs, $2; push @::flats, $::popped }) |
| 214 |
/\\includegraphics$1\{$::popped\}/gmx; |
| 215 |
|
| 216 |
if ($#::flats > -1) { |
| 217 |
my ($long, $flat); |
| 218 |
|
| 219 |
foreach $long (@::longs) { |
| 220 |
$flat = shift @::flats; |
| 221 |
|
| 222 |
# convert LyX directory dots |
| 223 |
$long =~ s/\\lyxdot /./g; |
| 224 |
$flat =~ s/\\lyxdot /./g; |
| 225 |
|
| 226 |
print "Looking for graphics: '$long'\n" if $DEBUG; |
| 227 |
|
| 228 |
my @dirs = ("", "$inputdir", $inputdir . '../figures/'); |
| 229 |
|
| 230 |
seekfile($long, $flat, \@imgexts, |
| 231 |
["", "$inputdir", $inputdir . '../figures/']); |
| 232 |
|
| 233 |
} |
| 234 |
} |
| 235 |
|
| 236 |
# Must do bibs before input commands because, when expanded, they |
| 237 |
# add new bibs inside. |
| 238 |
replacebibs(\$contents, $inputdir); |
| 239 |
|
| 240 |
# an \input or \include statement |
| 241 |
@flats = (); |
| 242 |
@longs = (); |
| 243 |
$contents =~ |
| 244 |
s/\\(input|include)\{([^}]*)\}/"\\${1}\{" . |
| 245 |
($popped = flattenfilename($2) and push @longs, $2 and push @flats,$popped and $popped) |
| 246 |
. "\}"/egm; |
| 247 |
|
| 248 |
if ($#flats > -1) { |
| 249 |
#print "Found " . scalar @longs . " items on line: '@flats'.\n"; |
| 250 |
|
| 251 |
my ($long, $flat); |
| 252 |
|
| 253 |
foreach $long (@longs) { |
| 254 |
$flat = shift @flats; |
| 255 |
print "Found Input/Include: '$long'\n" if $DEBUG; |
| 256 |
# recurse to parse that tex file unless it's done already |
| 257 |
|
| 258 |
if (not exists $texfiles{$flat}) { |
| 259 |
if (! -r $long ) { |
| 260 |
if (-r $inputdir . $long) { |
| 261 |
$long = $inputdir . $long; |
| 262 |
} elsif (-r "$long.tex") { |
| 263 |
$long .= '.tex'; |
| 264 |
} |
| 265 |
} |
| 266 |
$texfiles{$flat} = $long; |
| 267 |
my $newcontents = parseTeX($long, basedir($long), $flat, $level + 1); #later |
| 268 |
if ( $onetex ) { |
| 269 |
# If specified, expand inclusion directives |
| 270 |
die "Failed to find input/include to expand for '$flat'!\n" |
| 271 |
if (! ( $contents =~ |
| 272 |
s/\\(input|include) \{ \Q$flat\E \} |
| 273 |
(?{ print "===Found \\$1\{$flat\} to expand.\n" if $DEBUG; }) |
| 274 |
/ ($1 eq 'include' ? "\\clearpage\n" : '') . "$newcontents"/gxem ) ); |
| 275 |
} |
| 276 |
} |
| 277 |
} |
| 278 |
} |
| 279 |
|
| 280 |
#print "Writing to \"$outfile\"\n" . |
| 281 |
# "----------------------------------------\n" . |
| 282 |
# "$contents\n" . |
| 283 |
# "----------------------------------------\n"; |
| 284 |
|
| 285 |
if (! $onetex || $level == 0) { # do output at zero recurse level |
| 286 |
# open flat output file |
| 287 |
open my $COPY, ">$outfile" or die "Cannot write file $outfile\n"; |
| 288 |
|
| 289 |
# write to flat copy |
| 290 |
print $COPY $contents; |
| 291 |
|
| 292 |
close $COPY; |
| 293 |
} else { |
| 294 |
return $contents; |
| 295 |
} |
| 296 |
} |
| 297 |
|
| 298 |
# Look for file to copy |
| 299 |
sub seekfile { |
| 300 |
my ($file, $dest, $exts, $dirs) = @_; |
| 301 |
my @exts = @$exts; |
| 302 |
my @dirs = @$dirs; |
| 303 |
|
| 304 |
OUT: foreach my $dir (@dirs) { |
| 305 |
foreach my $ext (@exts) { |
| 306 |
my $tryfile = "$dir$file$ext"; |
| 307 |
if (-r $tryfile) { |
| 308 |
# add the extension to $dest only if it hasn't have any dots |
| 309 |
#(! ($dest =~ /\./)) ? $ext : ""; => better not to have this |
| 310 |
my $dext = $ext; |
| 311 |
|
| 312 |
if (system("cp $tryfile $outputdir/$dest$dext") != 0) { |
| 313 |
die "Cannot copy $tryfile!\n"; |
| 314 |
} |
| 315 |
# If we found it, then get out of the loop |
| 316 |
print "Copied: '$tryfile' (ext: $ext) -> '$dest$dext'\n" if $DEBUG; |
| 317 |
last OUT; |
| 318 |
} |
| 319 |
} |
| 320 |
} |
| 321 |
} |
| 322 |
|
| 323 |
# Return everything after the last / |
| 324 |
sub popfile { |
| 325 |
my $file = shift (@_); |
| 326 |
|
| 327 |
my @a = split /\//, $file; |
| 328 |
|
| 329 |
#print "Popping $file... split: @a ($#a), pop: " . pop(@a) . "\n"; |
| 330 |
pop @a; |
| 331 |
} |
| 332 |
|
| 333 |
# Make a new file by flattening directory names |
| 334 |
sub flattenfilename { |
| 335 |
local $_ = shift; |
| 336 |
|
| 337 |
# convert slashes to pluses, dots to Xs |
| 338 |
tr|/|+|; |
| 339 |
tr|\.|X|; |
| 340 |
|
| 341 |
# add a unique number at the beginning so files don't start with a |
| 342 |
# dot |
| 343 |
$_ = sprintf("%03d_", ++$::filecount) . "$_"; |
| 344 |
|
| 345 |
print "===Flattened file to '$_'\n" if $DEBUG; |
| 346 |
|
| 347 |
return $_; |
| 348 |
} |
| 349 |
|
| 350 |
sub basedir { |
| 351 |
my $file = shift (@_); |
| 352 |
|
| 353 |
my @a = split /\//, $file; |
| 354 |
|
| 355 |
#print "Popping $file... split: @a ($#a), pop: " . pop(@a) . "\n"; |
| 356 |
pop @a; # remove file part |
| 357 |
my $basedir = join('/', @a); # combine the rest with /'s |
| 358 |
$basedir .= '/' if (! $basedir =~ /^$/); # add a trailing / if not empty |
| 359 |
$basedir; |
| 360 |
} |
| 361 |
|
| 362 |
# get the bibliography files |
| 363 |
sub replacebibs { |
| 364 |
my ($contentsref, $inputdir) = @_; |
| 365 |
#my $contents = $$contentsref; |
| 366 |
my $popped; |
| 367 |
|
| 368 |
#print "bib contents: $contents\n"; |
| 369 |
|
| 370 |
local @::flats = (); |
| 371 |
local @::longs = (); |
| 372 |
local @::refs; |
| 373 |
|
| 374 |
$" = ','; # set list item separator |
| 375 |
# search for \bibliography statements |
| 376 |
$$contentsref =~ |
| 377 |
s/\\bibliography(\[[^\]]*\])? { |
| 378 |
(?{ @::refs = (); #print "Found bibliography!\n"; #initialize ref list |
| 379 |
}) (?: \s* ([^}, ]+)(?=[,}]),? |
| 380 |
(?{ # for each word do: |
| 381 |
#local (@_flats, @_longs, @_refs); |
| 382 |
#print "Word: \"$+\", "; |
| 383 |
$popped = flattenfilename($+); push @::longs, $+; |
| 384 |
push @::flats,$popped; push @::refs, $popped; |
| 385 |
}) )+ \s* } |
| 386 |
(?{ #print "\n+++ $#::flats; $#::refs; $#::longs\n"; |
| 387 |
#push @::flats, @_flats; push @longs, @_longs; push @refs, @_refs; |
| 388 |
#print "Refs: @::refs\n"; |
| 389 |
})/"\\bibliography" . ($1 || "") . "{@::refs}"/egmx;# && |
| 390 |
#print "bib contents: $$contentsref\n"; |
| 391 |
#print "\n*** $#::flats; $#::refs; $#::longs\n"; |
| 392 |
$" = ' '; # restore list item separator |
| 393 |
|
| 394 |
if ($#::flats > -1) { |
| 395 |
#print "Found " . scalar @::longs . " items on line: '@::flats'.\n"; |
| 396 |
|
| 397 |
my ($long, $flat); |
| 398 |
|
| 399 |
foreach $long (@::longs) { |
| 400 |
$flat = shift @::flats; |
| 401 |
print "Found bib. file: '$long'\n" if $DEBUG; |
| 402 |
# recurse to parse that tex file unless it's done already |
| 403 |
|
| 404 |
# convert LyX directory dots |
| 405 |
$long =~ s/\\lyxdot /./g; |
| 406 |
|
| 407 |
seekfile($long, "$flat", [".bib"], |
| 408 |
["", "$inputdir"]); |
| 409 |
} |
| 410 |
} |
| 411 |
} |