summaryrefslogtreecommitdiff
path: root/docs/docbook/scripts/collateindex.pl
blob: fd757edb320e86601a991c658b50359afe97a778 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
# -*- Perl -*-
#

use Getopt::Std;

$usage = "Usage: $0 <opts> file
Where <opts> are:
       -p        Link to points in the document.  The default is to link
                 to the closest containing section.
       -g        Group terms with IndexDiv based on the first letter 
                 of the term (or its sortas attribute).
                 (This probably doesn't handle i10n particularly well)
       -s name   Name the IndexDiv that contains symbols.  The default
                 is 'Symbols'.  Meaningless if -g is not used.
       -t name   Title for the index.
       -P file   Read a preamble from file.  The content of file will
                 be inserted before the <index> tag.
       -i id     The ID for the <index> tag.
       -o file   Output to file. Defaults to stdout.
       -S scope  Scope of the index, must be 'all', 'local', or 'global'.
                 If unspecified, 'all' is assumed.
       -I scope  The implied scope, must be 'all', 'local', or 'global'.
                 IndexTerms which do not specify a scope will have the
                 implied scope.  If unspecified, 'all' is assumed.
       -x        Make a SetIndex.
       -f        Force the output file to be written, even if it appears
                 to have been edited by hand.
       -N        New index (generates an empty index file).
       file      The file containing index data generated by Jade
                 with the DocBook HTML Stylesheet.\n";

die $usage if ! getopts('Dfgi:NpP:s:o:S:I:t:x');

$linkpoints   = $opt_p;
$lettergroups = $opt_g;
$symbolsname  = $opt_s || "Symbols";
$title        = $opt_t;
$preamble     = $opt_P;
$outfile      = $opt_o || '-';
$indexid      = $opt_i;
$scope        = uc($opt_S) || 'ALL';
$impliedscope = uc($opt_I) || 'ALL';
$setindex     = $opt_x;
$forceoutput  = $opt_f;
$newindex     = $opt_N;
$debug        = $opt_D;

$indextag     = $setindex ? 'setindex' : 'index';

if ($newindex) {
    safe_open(*OUT, $outfile);
    if ($indexid) {
	print OUT "<$indextag id='$indexid'>\n\n";
    } else {
	print OUT "<$indextag>\n\n";
    }

    print OUT "<!-- This file was produced by collateindex.pl.         -->\n";
    print OUT "<!-- Remove this comment if you edit this file by hand! -->\n";

    print OUT "</$indextag>\n";
    exit 0;
}

$dat = shift @ARGV || die $usage;
die "$0: cannot find $dat.\n" if ! -f $dat;

%legal_scopes = ('ALL' => 1, 'LOCAL' => 1, 'GLOBAL' => 1);
if ($scope && !$legal_scopes{$scope}) {
    die "Invalid scope.\n$usage\n";
}
if ($impliedscope && !$legal_scopes{$impliedscope}) {
    die "Invalid implied scope.\n$usage\n";
}

@term = ();
%id   = ();

$termcount = 0;

print STDERR "Processing $dat...\n";

# Read the index file, creating an array of objects.  Each object 
# represents and indexterm and has fields for the content of the
# indexterm

open (F, $dat);
while (<F>) {
    chop;

    if (/^\/indexterm/i) {
	push (@term, $idx);
	next;
    }

    if (/^indexterm (.*)$/i) {
	$termcount++;
	$idx = {};
	$idx->{'zone'} = {};
	$idx->{'href'} = $1;
	$idx->{'count'} = $termcount;
	$idx->{'scope'} = $impliedscope;
	next;
    }

    if (/^indexpoint (.*)$/i) {
	$idx->{'hrefpoint'} = $1;
	next;
    }

    if (/^title (.*)$/i) {
	$idx->{'title'} = $1;
	next;
    }

    if (/^primary[\[ ](.*)$/i) {
	if (/^primary\[(.*?)\] (.*)$/i) {
	    $idx->{'psortas'} = $1;
	    $idx->{'primary'} = $2;
	} else {
	    $idx->{'psortas'} = $1;
	    $idx->{'primary'} = $1;
	}
	next;
    }

    if (/^secondary[\[ ](.*)$/i) {
	if (/^secondary\[(.*?)\] (.*)$/i) {
	    $idx->{'ssortas'} = $1;
	    $idx->{'secondary'} = $2;
	} else {
	    $idx->{'ssortas'} = $1;
	    $idx->{'secondary'} = $1;
	}
	next;
    }

    if (/^tertiary[\[ ](.*)$/i) {
	if (/^tertiary\[(.*?)\] (.*)$/i) {
	    $idx->{'tsortas'} = $1;
	    $idx->{'tertiary'} = $2;
	} else {
	    $idx->{'tsortas'} = $1;
	    $idx->{'tertiary'} = $1;
	}
	next;
    }

    if (/^see (.*)$/i) {
	$idx->{'see'} = $1;
	next;
    }

    if (/^seealso (.*)$/i) {
	$idx->{'seealso'} = $1;
	next;
    }

    if (/^significance (.*)$/i) {
	$idx->{'significance'} = $1;
	next;
    }

    if (/^class (.*)$/i) {
	$idx->{'class'} = $1;
	next;
    }

    if (/^scope (.*)$/i) {
	$idx->{'scope'} = uc($1);
	next;
    }

    if (/^startref (.*)$/i) {
	$idx->{'startref'} = $1;
	next;
    }

    if (/^id (.*)$/i) {
	$idx->{'id'} = $1;
	$id{$1} = $idx;
	next;
    }

    if (/^zone (.*)$/i) {
	my($href) = $1;
	$_ = scalar(<F>);
	chop;
	die "Bad zone: $_\n" if !/^title (.*)$/i;
	$idx->{'zone'}->{$href} = $1;
	next;
    }

    die "Unrecognized: $_\n";
}
close (F);

print STDERR "$termcount entries loaded...\n";

# Fixup the startrefs...
# In DocBook, STARTREF is a #CONREF attribute; support this by copying
# all of the fields from the indexterm with the id specified by STARTREF
# to the indexterm that has the STARTREF.
foreach $idx (@term) {
    my($ididx, $field);
    if ($idx->{'startref'}) {
	$ididx = $id{$idx->{'startref'}};
	foreach $field ('primary', 'secondary', 'tertiary', 'see', 'seealso',
		        'psortas', 'ssortas', 'tsortas', 'significance',
		        'class', 'scope') {
	    $idx->{$field} = $ididx->{$field};
	}
    }
}

# Sort the index terms
@term = sort termsort @term;

# Move all of the non-alphabetic entries to the front of the index.
@term = sortsymbols(@term);

safe_open(*OUT, $outfile);

# Write the index...
if ($indexid) {
    print OUT "<$indextag id='$indexid'>\n\n";
} else {
    print OUT "<$indextag>\n\n";
}

print OUT "<!-- This file was produced by collateindex.pl.         -->\n";
print OUT "<!-- Remove this comment if you edit this file by hand! -->\n";

print OUT "<!-- ULINK is abused here.
      
      The URL attribute holds the URL that points from the index entry
      back to the appropriate place in the output produced by the HTML
      stylesheet. (It's much easier to calculate this URL in the first
      pass.)

      The Role attribute holds the ID (either real or manufactured) of
      the corresponding INDEXTERM.  This is used by the print backends
      to produce page numbers.

      The entries below are sorted and collated into the correct order.
      Duplicates may be removed in the HTML backend, but in the print
      backends, it is impossible to suppress duplicate pages or coalesce
      sequences of pages into a range.
-->\n\n";

print OUT "<title>$title</title>\n\n" if $title;

$last = {};     # the last indexterm we processed
$first = 1;     # this is the first one
$group = "";    # we're not in a group yet
$lastout = "";  # we've not put anything out yet

foreach $idx (@term) {
    next if $idx->{'startref'}; # no way to represent spans...
    next if ($idx->{'scope'} eq 'LOCAL') && ($scope eq 'GLOBAL');
    next if ($idx->{'scope'} eq 'GLOBAL') && ($scope eq 'LOCAL');
    next if &same($idx, $last); # suppress duplicates

    $termcount--;

    # If primary changes, output a whole new index term, otherwise just
    # output another secondary or tertiary, as appropriate.  We know from
    # sorting that the terms will always be in the right order.
    if (!&tsame($last, $idx, 'primary')) { 
	print "DIFF PRIM\n" if $debug;
	&end_entry() if not $first;

	if ($lettergroups) {
	    # If we're grouping, make the right indexdivs
	    $letter = $idx->{'psortas'};
	    $letter = $idx->{'primary'} if !$letter;
	    $letter = uc(substr($letter, 0, 1));
	    
	    # symbols are a special case
	    if (($letter lt 'A') || ($letter gt 'Z')) {
		if (($group eq '')
		    || (($group ge 'A') && ($group le 'Z'))) {
		    print OUT "</indexdiv>\n" if !$first;
		    print OUT "<indexdiv><title>$symbolsname</title>\n\n";
		    $group = $letter;
		}
	    } elsif (($group eq '') || ($group ne $letter)) {
		print OUT "</indexdiv>\n" if !$first;
		print OUT "<indexdiv><title>$letter</title>\n\n";
		$group = $letter;
	    }
	}

	$first = 0; # there can only be on first ;-)

	print OUT "<indexentry>\n";
	print OUT "  <primaryie>", $idx->{'primary'};
	$lastout = "primaryie";

 	if ($idx->{'secondary'}) {
	    print OUT "\n  </primaryie>\n";
	    print OUT "  <secondaryie>", $idx->{'secondary'};
	    $lastout = "secondaryie";
	};

	if ($idx->{'tertiary'}) {
	    print OUT "\n  </secondaryie>\n";
	    print OUT "  <tertiaryie>", $idx->{'tertiary'};
	    $lastout = "tertiaryie";
	}
    } elsif (!&tsame($last, $idx, 'secondary')) {
	print "DIFF SEC\n" if $debug;

	print OUT "\n  </$lastout>\n" if $lastout;

	print OUT "  <secondaryie>", $idx->{'secondary'};
	$lastout = "secondaryie";
	if ($idx->{'tertiary'}) {
	    print OUT "\n  </secondaryie>\n";
	    print OUT "  <tertiaryie>", $idx->{'tertiary'};
	    $lastout = "tertiaryie";
	}
    } elsif (!&tsame($last, $idx, 'tertiary')) {
	print "DIFF TERT\n" if $debug;

	print OUT "\n  </$lastout>\n" if $lastout;

	if ($idx->{'tertiary'}) {
	    print OUT "  <tertiaryie>", $idx->{'tertiary'};
	    $lastout = "tertiaryie";
	}
    }

    &print_term($idx);
    
    $last = $idx;
}

# Termcount is > 0 iff some entries were skipped.
print STDERR "$termcount entries ignored...\n";

&end_entry();

print OUT "</indexdiv>\n" if $lettergroups;
print OUT "</$indextag>\n";

close (OUT);

print STDERR "Done.\n";

sub same {
    my($a) = shift;
    my($b) = shift;

    my($aP) = $a->{'psortas'} || $a->{'primary'};   
    my($aS) = $a->{'ssortas'} || $a->{'secondary'}; 
    my($aT) = $a->{'tsortas'} || $a->{'tertiary'};  
	                                            
    my($bP) = $b->{'psortas'} || $b->{'primary'};   
    my($bS) = $b->{'ssortas'} || $b->{'secondary'}; 
    my($bT) = $b->{'tsortas'} || $b->{'tertiary'};  

    my($same);

    $aP =~ s/^\s*//; $aP =~ s/\s*$//; $aP = uc($aP);
    $aS =~ s/^\s*//; $aS =~ s/\s*$//; $aS = uc($aS);
    $aT =~ s/^\s*//; $aT =~ s/\s*$//; $aT = uc($aT);
    $bP =~ s/^\s*//; $bP =~ s/\s*$//; $bP = uc($bP);
    $bS =~ s/^\s*//; $bS =~ s/\s*$//; $bS = uc($bS);
    $bT =~ s/^\s*//; $bT =~ s/\s*$//; $bT = uc($bT);

#    print "[$aP]=[$bP]\n";
#    print "[$aS]=[$bS]\n";
#    print "[$aT]=[$bT]\n";

    # Two index terms are the same if:
    # 1. the primary, secondary, and tertiary entries are the same
    #    (or have the same SORTAS)
    # AND
    # 2. They occur in the same titled section
    # AND
    # 3. They point to the same place
    #
    # Notes: Scope is used to suppress some entries, but can't be used
    #          for comparing duplicates.
    #        Interpretation of "the same place" depends on whether or
    #          not $linkpoints is true.

    $same = (($aP eq $bP)
	     && ($aS eq $bS)
	     && ($aT eq $bT)
	     && ($a->{'title'} eq $b->{'title'})
	     && ($a->{'href'} eq $b->{'href'}));

    # If we're linking to points, they're only the same if they link
    # to exactly the same spot.  (surely this is redundant?)
    $same = $same && ($a->{'hrefpoint'} eq $b->{'hrefpoint'})
	if $linkpoints;

    $same;
}

sub tsame {
    # Unlike same(), tsame only compares a single term
    my($a) = shift;
    my($b) = shift;
    my($term) = shift;
    my($sterm) = substr($term, 0, 1) . "sortas";
    my($A, $B);

    $A = $a->{$sterm} || $a->{$term};
    $B = $b->{$sterm} || $b->{$term};

    $A =~ s/^\s*//; $A =~ s/\s*$//; $A = uc($A);
    $B =~ s/^\s*//; $B =~ s/\s*$//; $B = uc($B);

    return $A eq $B;
}

sub end_entry {
    # End any open elements...
    print OUT "\n  </$lastout>\n" if $lastout;
    print OUT "</indexentry>\n\n";
    $lastout = "";
}

sub print_term {
    # Print out the links for an indexterm.  There can be more than
    # one if the term has a ZONE that points to more than one place.
    # (do we do the right thing in that case?)
    my($idx) = shift;
    my($key, $indent, @hrefs);
    my(%href) = ();
    my(%phref) = ();

    $indent = "    ";

    if ($idx->{'see'}) {
	# it'd be nice to make this a link...
	if ($lastout) {
	    print OUT "\n  </$lastout>\n";
	    $lastout = "";
	}
	print OUT $indent, "<seeie>", $idx->{'see'}, "</seeie>\n";
	return;
    }

    if ($idx->{'seealso'}) {
	# it'd be nice to make this a link...
	if ($lastout) {
	    print OUT "\n  </$lastout>\n";
	    $lastout = "";
	}
	print OUT $indent, "<seealsoie>", $idx->{'seealso'}, "</seealsoie>\n";
	return;
    }

    if (keys %{$idx->{'zone'}}) {
	foreach $key (keys %{$idx->{'zone'}}) {
	    $href{$key} = $idx->{'zone'}->{$key};
	    $phref{$key} = $idx->{'zone'}->{$key};
	}
    } else {
	$href{$idx->{'href'}} = $idx->{'title'};
	$phref{$idx->{'href'}} = $idx->{'hrefpoint'};
    }

    # We can't use <LINK> because we don't know the ID of the term in the
    # original source (and, in fact, it might not have one).
    print OUT ",\n";
    @hrefs = keys %href;
    while (@hrefs) {
	my($linkend) = "";
	my($role) = "";
	$key = shift @hrefs;
	if ($linkpoints) {
	    $linkend = $phref{$key};
	} else {
	    $linkend = $key;
	}

	$role = $linkend;
	$role = $1 if $role =~ /\#(.*)$/;

	print OUT $indent;
	print OUT "<ulink url=\"$linkend\" role=\"$role\">";
	print OUT "<emphasis>" if ($idx->{'significance'} eq 'PREFERRED');
	print OUT $href{$key};
	print OUT "</emphasis>" if ($idx->{'significance'} eq 'PREFERRED');
	print OUT "</ulink>";
    }
}

sub termsort {
    my($aP) = $a->{'psortas'} || $a->{'primary'};   
    my($aS) = $a->{'ssortas'} || $a->{'secondary'}; 
    my($aT) = $a->{'tsortas'} || $a->{'tertiary'};  
    my($ap) = $a->{'count'};
	                                            
    my($bP) = $b->{'psortas'} || $b->{'primary'};   
    my($bS) = $b->{'ssortas'} || $b->{'secondary'}; 
    my($bT) = $b->{'tsortas'} || $b->{'tertiary'};  
    my($bp) = $b->{'count'};

    $aP =~ s/^\s*//; $aP =~ s/\s*$//; $aP = uc($aP);
    $aS =~ s/^\s*//; $aS =~ s/\s*$//; $aS = uc($aS);
    $aT =~ s/^\s*//; $aT =~ s/\s*$//; $aT = uc($aT);
    $bP =~ s/^\s*//; $bP =~ s/\s*$//; $bP = uc($bP);
    $bS =~ s/^\s*//; $bS =~ s/\s*$//; $bS = uc($bS);
    $bT =~ s/^\s*//; $bT =~ s/\s*$//; $bT = uc($bT);

    if ($aP eq $bP) {
	if ($aS eq $bS) {
	    if ($aT eq $bT) {
		# make sure seealso's always sort to the bottom
		return 1 if ($a->{'seealso'});
		return -1  if ($b->{'seealso'});
		# if everything else is the same, keep these elements
		# in document order (so the index links are in the right
		# order)
		return $ap <=> $bp;
	    } else {
		return $aT cmp $bT;
	    }
	} else {
	    return $aS cmp $bS;
	}
    } else {
	return $aP cmp $bP;
    }
}

sub sortsymbols {
    my(@term) = @_;
    my(@new) = ();
    my(@sym) = ();
    my($letter);
    my($idx);

    # Move the non-letter things to the front.  Should digits be thier
    # own group?  Maybe...
    foreach $idx (@term) {
	$letter = $idx->{'psortas'};
	$letter = $idx->{'primary'} if !$letter;
	$letter = uc(substr($letter, 0, 1));

	if (($letter lt 'A') || ($letter gt 'Z')) {
	    push (@sym, $idx);
	} else {
	    push (@new, $idx);
	}
    }

    return (@sym, @new);
}

sub safe_open {
    local(*OUT) = shift;
    local(*F, $_);

    if (($outfile ne '-') && (!$forceoutput)) {
	my($handedit) = 1;
	if (open (OUT, $outfile)) {
	    while (<OUT>) {
		if (/<!-- Remove this comment if you edit this file by hand! -->/){
		    $handedit = 0;
		    last;
		}
	    } 
	    close (OUT);
	} else {
	    $handedit = 0;
	}
	
	if ($handedit) {
	    print "\n$outfile appears to have been edited by hand; use -f or\n";
	    print "      change the output file.\n";
	    exit 1;
	}
    }

    open (OUT, ">$outfile") || die "$usage\nCannot write to $outfile.\n";

    if ($preamble) { 
	# Copy the preamble
	if (open(F, $preamble)) {
	    while (<F>) {
		print OUT $_;
	    }
	    close(F);
	} else {
	    warn "$0: cannot open preamble $preamble.\n";
	}
    }
}