xref: /openbsd-src/gnu/usr.bin/perl/cpan/Pod-Simple/lib/Pod/Simple/BlackBox.pm (revision 5a38ef86d0b61900239c7913d24a05e7b88a58f0)
1package Pod::Simple::BlackBox;
2#
3# "What's in the box?"  "Pain."
4#
5###########################################################################
6#
7# This is where all the scary things happen: parsing lines into
8#  paragraphs; and then into directives, verbatims, and then also
9#  turning formatting sequences into treelets.
10#
11# Are you really sure you want to read this code?
12#
13#-----------------------------------------------------------------------------
14#
15# The basic work of this module Pod::Simple::BlackBox is doing the dirty work
16# of parsing Pod into treelets (generally one per non-verbatim paragraph), and
17# to call the proper callbacks on the treelets.
18#
19# Every node in a treelet is a ['name', {attrhash}, ...children...]
20
21use integer; # vroom!
22use strict;
23use Carp ();
24use vars qw($VERSION );
25$VERSION = '3.40';
26#use constant DEBUG => 7;
27
28sub my_qr ($$) {
29
30    # $1 is a pattern to compile and return.  Older perls compile any
31    # syntactically valid property, even if it isn't legal.  To cope with
32    # this, return an empty string unless the compiled pattern also
33    # successfully matches $2, which the caller furnishes.
34
35    my ($input_re, $should_match) = @_;
36    # XXX could have a third parameter $shouldnt_match for extra safety
37
38    my $use_utf8 = ($] le 5.006002) ? 'use utf8;' : "";
39
40    my $re = eval "no warnings; $use_utf8 qr/$input_re/";
41    #print STDERR  __LINE__, ": $input_re: $@\n" if $@;
42    return "" if $@;
43
44    my $matches = eval "no warnings; $use_utf8 '$should_match' =~ /$re/";
45    #print STDERR  __LINE__, ": $input_re: $@\n" if $@;
46    return "" if $@;
47
48    #print STDERR  __LINE__, ": SUCCESS: $re\n" if $matches;
49    return $re if $matches;
50
51    #print STDERR  __LINE__, ": $re: didn't match\n";
52    return "";
53}
54
55BEGIN {
56  require Pod::Simple;
57  *DEBUG = \&Pod::Simple::DEBUG unless defined &DEBUG
58}
59
60# Matches a character iff the character will have a different meaning
61# if we choose CP1252 vs UTF-8 if there is no =encoding line.
62# This is broken for early Perls on non-ASCII platforms.
63my $non_ascii_re = my_qr('[[:^ascii:]]', "\xB6");
64$non_ascii_re = qr/[\x80-\xFF]/ unless $non_ascii_re;
65
66# Use patterns understandable by Perl 5.6, if possible
67my $cs_re = my_qr('\p{IsCs}', "\x{D800}");
68my $cn_re = my_qr('\p{IsCn}', "\x{09E4}");  # <reserved> code point unlikely
69                                            # to get assigned
70my $rare_blocks_re = my_qr('[\p{InIPAExtensions}\p{InSpacingModifierLetters}]',
71                           "\x{250}");
72$rare_blocks_re = my_qr('[\x{0250}-\x{02FF}]', "\x{250}") unless $rare_blocks_re;
73
74my $script_run_re = eval 'no warnings "experimental::script_run";
75                          qr/(*script_run: ^ .* $ )/x';
76my $latin_re = my_qr('[\p{IsLatin}\p{IsInherited}\p{IsCommon}]', "\x{100}");
77unless ($latin_re) {
78    # This was machine generated to be the ranges of the union of the above
79    # three properties, with things that were undefined by Unicode 4.1 filling
80    # gaps.  That is the version in use when Perl advanced enough to
81    # successfully compile and execute the above pattern.
82    $latin_re = my_qr('[\x00-\x{02E9}\x{02EC}-\x{0374}\x{037E}\x{0385}\x{0387}\x{0485}\x{0486}\x{0589}\x{060C}\x{061B}\x{061F}\x{0640}\x{064B}-\x{0655}\x{0670}\x{06DD}\x{0951}-\x{0954}\x{0964}\x{0965}\x{0E3F}\x{10FB}\x{16EB}-\x{16ED}\x{1735}\x{1736}\x{1802}\x{1803}\x{1805}\x{1D00}-\x{1D25}\x{1D2C}-\x{1D5C}\x{1D62}-\x{1D65}\x{1D6B}-\x{1D77}\x{1D79}-\x{1DBE}\x{1DC0}-\x{1EF9}\x{2000}-\x{2125}\x{2127}-\x{27FF}\x{2900}-\x{2B13}\x{2E00}-\x{2E1D}\x{2FF0}-\x{3004}\x{3006}\x{3008}-\x{3020}\x{302A}-\x{302D}\x{3030}-\x{3037}\x{303C}-\x{303F}\x{3099}-\x{309C}\x{30A0}\x{30FB}\x{30FC}\x{3190}-\x{319F}\x{31C0}-\x{31CF}\x{3220}-\x{325F}\x{327F}-\x{32CF}\x{3358}-\x{33FF}\x{4DC0}-\x{4DFF}\x{A700}-\x{A716}\x{FB00}-\x{FB06}\x{FD3E}\x{FD3F}\x{FE00}-\x{FE6B}\x{FEFF}-\x{FF65}\x{FF70}\x{FF9E}\x{FF9F}\x{FFE0}-\x{FFFD}\x{10100}-\x{1013F}\x{1D000}-\x{1D1DD}\x{1D300}-\x{1D7FF}]', "\x{100}");
83}
84
85my $every_char_is_latin_re = my_qr("^(?:$latin_re)*\\z", "A");
86
87# Latin script code points not in the first release of Unicode
88my $later_latin_re = my_qr('[^\P{IsLatin}\p{IsAge=1.1}]', "\x{1F6}");
89
90# If this perl doesn't have the Deprecated property, there's only one code
91# point in it that we need be concerned with.
92my $deprecated_re = my_qr('\p{IsDeprecated}', "\x{149}");
93$deprecated_re = qr/\x{149}/ unless $deprecated_re;
94
95my $utf8_bom;
96if (($] ge 5.007_003)) {
97  $utf8_bom = "\x{FEFF}";
98  utf8::encode($utf8_bom);
99} else {
100  $utf8_bom = "\xEF\xBB\xBF";   # No EBCDIC BOM detection for early Perls.
101}
102
103# This is used so that the 'content_seen' method doesn't return true on a
104# file that just happens to have a line that matches /^=[a-zA-z]/.  Only if
105# there is a valid =foo line will we return that content was seen.
106my $seen_legal_directive = 0;
107
108#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
109
110sub parse_line { shift->parse_lines(@_) } # alias
111
112# - - -  Turn back now!  Run away!  - - -
113
114sub parse_lines {             # Usage: $parser->parse_lines(@lines)
115  # an undef means end-of-stream
116  my $self = shift;
117
118  my $code_handler = $self->{'code_handler'};
119  my $cut_handler  = $self->{'cut_handler'};
120  my $wl_handler   = $self->{'whiteline_handler'};
121  $self->{'line_count'} ||= 0;
122
123  my $scratch;
124
125  DEBUG > 4 and
126   print STDERR "# Parsing starting at line ", $self->{'line_count'}, ".\n";
127
128  DEBUG > 5 and
129   print STDERR "#  About to parse lines: ",
130     join(' ', map defined($_) ? "[$_]" : "EOF", @_), "\n";
131
132  my $paras = ($self->{'paras'} ||= []);
133   # paragraph buffer.  Because we need to defer processing of =over
134   # directives and verbatim paragraphs.  We call _ponder_paragraph_buffer
135   # to process this.
136
137  $self->{'pod_para_count'} ||= 0;
138
139  # An attempt to match the pod portions of a line.  This is not fool proof,
140  # but is good enough to serve as part of the heuristic for guessing the pod
141  # encoding if not specified.
142  my $format_codes = join "", '[', grep { / ^ [A-Za-z] $/x }
143                                                keys %{$self->{accept_codes}};
144  $format_codes .= ']';
145  my $pod_chars_re = qr/ ^ = [A-Za-z]+ | $format_codes < /x;
146
147  my $line;
148  foreach my $source_line (@_) {
149    if( $self->{'source_dead'} ) {
150      DEBUG > 4 and print STDERR "# Source is dead.\n";
151      last;
152    }
153
154    unless( defined $source_line ) {
155      DEBUG > 4 and print STDERR "# Undef-line seen.\n";
156
157      push @$paras, ['~end', {'start_line' => $self->{'line_count'}}];
158      push @$paras, $paras->[-1], $paras->[-1];
159       # So that it definitely fills the buffer.
160      $self->{'source_dead'} = 1;
161      $self->_ponder_paragraph_buffer;
162      next;
163    }
164
165
166    if( $self->{'line_count'}++ ) {
167      ($line = $source_line) =~ tr/\n\r//d;
168       # If we don't have two vars, we'll end up with that there
169       # tr/// modding the (potentially read-only) original source line!
170
171    } else {
172      DEBUG > 2 and print STDERR "First line: [$source_line]\n";
173
174      if( ($line = $source_line) =~ s/^$utf8_bom//s ) {
175        DEBUG and print STDERR "UTF-8 BOM seen.  Faking a '=encoding utf8'.\n";
176        $self->_handle_encoding_line( "=encoding utf8" );
177        delete $self->{'_processed_encoding'};
178        $line =~ tr/\n\r//d;
179
180      } elsif( $line =~ s/^\xFE\xFF//s ) {
181        DEBUG and print STDERR "Big-endian UTF-16 BOM seen.  Aborting parsing.\n";
182        $self->scream(
183          $self->{'line_count'},
184          "UTF16-BE Byte Encoding Mark found; but Pod::Simple v$Pod::Simple::VERSION doesn't implement UTF16 yet."
185        );
186        splice @_;
187        push @_, undef;
188        next;
189
190        # TODO: implement somehow?
191
192      } elsif( $line =~ s/^\xFF\xFE//s ) {
193        DEBUG and print STDERR "Little-endian UTF-16 BOM seen.  Aborting parsing.\n";
194        $self->scream(
195          $self->{'line_count'},
196          "UTF16-LE Byte Encoding Mark found; but Pod::Simple v$Pod::Simple::VERSION doesn't implement UTF16 yet."
197        );
198        splice @_;
199        push @_, undef;
200        next;
201
202        # TODO: implement somehow?
203
204      } else {
205        DEBUG > 2 and print STDERR "First line is BOM-less.\n";
206        ($line = $source_line) =~ tr/\n\r//d;
207      }
208    }
209
210    if(!$self->{'parse_characters'} && !$self->{'encoding'}
211      && ($self->{'in_pod'} || $line =~ /^=/s)
212      && $line =~ /$non_ascii_re/
213    ) {
214
215      my $encoding;
216
217      # No =encoding line, and we are at the first pod line in the input that
218      # contains a non-ascii byte, that is, one whose meaning varies depending
219      # on whether the file is encoded in UTF-8 or CP1252, which are the two
220      # possibilities permitted by the pod spec.  (ASCII is assumed if the
221      # file only contains ASCII bytes.)  In order to process this line, we
222      # need to figure out what encoding we will use for the file.
223      #
224      # Strictly speaking ISO 8859-1 (Latin 1) refers to the code points
225      # 160-255, but it is used here, as it often colloquially is, to refer to
226      # the complete set of code points 0-255, including ASCII (0-127), the C1
227      # controls (128-159), and strict Latin 1 (160-255).
228      #
229      # CP1252 is effectively a superset of Latin 1, because it differs only
230      # from colloquial 8859-1 in the C1 controls, which are very unlikely to
231      # actually be present in 8859-1 files, so can be used for other purposes
232      # without conflict.  CP 1252 uses most of them for graphic characters.
233      #
234      # Note that all ASCII-range bytes represent their corresponding code
235      # points in both CP1252 and UTF-8.  In ASCII platform UTF-8, all other
236      # code points require multiple (non-ASCII) bytes to represent.  (A
237      # separate paragraph for EBCDIC is below.)  The multi-byte
238      # representation is quite structured.  If we find an isolated byte that
239      # would require multiple bytes to represent in UTF-8, we know that the
240      # encoding is not UTF-8.  If we find a sequence of bytes that violates
241      # the UTF-8 structure, we also can presume the encoding isn't UTF-8, and
242      # hence must be 1252.
243      #
244      # But there are ambiguous cases where we could guess wrong.  If so, the
245      # user will end up having to supply an =encoding line.  We use all
246      # readily available information to improve our chances of guessing
247      # right.  The odds of something not being UTF-8, but still passing a
248      # UTF-8 validity test go down very rapidly with increasing length of the
249      # sequence.  Therefore we look at all non-ascii sequences on the line.
250      # If any of the sequences can't be UTF-8, we quit there and choose
251      # CP1252.  If all could be UTF-8, we see if any of the code points
252      # represented are unlikely to be in pod.  If so, we guess CP1252.  If
253      # not, we check if the line is all in the same script; if not guess
254      # CP1252; otherwise UTF-8.  For perls that don't have convenient script
255      # run testing, see if there is both Latin and non-Latin.  If so, CP1252,
256      # otherwise UTF-8.
257      #
258      # On EBCDIC platforms, the situation is somewhat different.  In
259      # UTF-EBCDIC, not only do ASCII-range bytes represent their code points,
260      # but so do the bytes that are for the C1 controls.  Recall that these
261      # correspond to the unused portion of 8859-1 that 1252 mostly takes
262      # over.  That means that there are fewer code points that are
263      # represented by multi-bytes.  But, note that the these controls are
264      # very unlikely to be in pod text.  So if we encounter one of them, it
265      # means that it is quite likely CP1252 and not UTF-8.  The net result is
266      # the same code below is used for both platforms.
267      #
268      # XXX probably if the line has E<foo> that evaluates to illegal CP1252,
269      # then it is UTF-8.  But we haven't processed E<> yet.
270
271      goto set_1252 if $] lt 5.006_000;    # No UTF-8 on very early perls
272
273      my $copy;
274
275      no warnings 'utf8';
276
277      if ($] ge 5.007_003) {
278        $copy = $line;
279
280        # On perls that have this function, we can use it to easily see if the
281        # sequence is valid UTF-8 or not; if valid it turns on the UTF-8 flag
282        # needed below for script run detection
283        goto set_1252 if ! utf8::decode($copy);
284      }
285      elsif (ord("A") != 65) {  # Early EBCDIC, assume UTF-8.  What's a windows
286                                # code page doing here anyway?
287        goto set_utf8;
288      }
289      else { # ASCII, no decode(): do it ourselves using the fundamental
290             # characteristics of UTF-8
291        use if $] le 5.006002, 'utf8';
292
293        my $char_ord;
294        my $needed;         # How many continuation bytes to gobble up
295
296        # Initialize the translated line with a dummy character that will be
297        # deleted after everything else is done.  This dummy makes sure that
298        # $copy will be in UTF-8.  Doing it now avoids the bugs in early perls
299        # with upgrading in the middle
300        $copy = chr(0x100);
301
302        # Parse through the line
303        for (my $i = 0; $i < length $line; $i++) {
304          my $byte = substr($line, $i, 1);
305
306          # ASCII bytes are trivially dealt with
307          if ($byte !~ $non_ascii_re) {
308            $copy .= $byte;
309            next;
310          }
311
312          my $b_ord = ord $byte;
313
314          # Now figure out what this code point would be if the input is
315          # actually in UTF-8.  If, in the process, we discover that it isn't
316          # well-formed UTF-8, we guess CP1252.
317          #
318          # Start the process.  If it is UTF-8, we are at the first, start
319          # byte, of a multi-byte sequence.  We look at this byte to figure
320          # out how many continuation bytes are needed, and to initialize the
321          # code point accumulator with the data from this byte.
322          #
323          # Normally the minimum continuation byte is 0x80, but in certain
324          # instances the minimum is a higher number.  So the code below
325          # overrides this for those instances.
326          my $min_cont = 0x80;
327
328          if ($b_ord < 0xC2) { #  A start byte < C2 is malformed
329            goto set_1252;
330          }
331          elsif ($b_ord <= 0xDF) {
332            $needed = 1;
333            $char_ord = $b_ord & 0x1F;
334          }
335          elsif ($b_ord <= 0xEF) {
336            $min_cont = 0xA0 if $b_ord == 0xE0;
337            $needed = 2;
338            $char_ord = $b_ord & (0x1F >> 1);
339          }
340          elsif ($b_ord <= 0xF4) {
341            $min_cont = 0x90 if $b_ord == 0xF0;
342            $needed = 3;
343            $char_ord = $b_ord & (0x1F >> 2);
344          }
345          else { # F4 is the highest start byte for legal Unicode; higher is
346                 # unlikely to be in pod.
347            goto set_1252;
348          }
349
350          # ? not enough continuation bytes available
351          goto set_1252 if $i + $needed >= length $line;
352
353          # Accumulate the ordinal of the character from the remaining
354          # (continuation) bytes.
355          while ($needed-- > 0) {
356            my $cont = substr($line, ++$i, 1);
357            $b_ord = ord $cont;
358            goto set_1252 if $b_ord < $min_cont || $b_ord > 0xBF;
359
360            # In all cases, any next continuation bytes all have the same
361            # minimum legal value
362            $min_cont = 0x80;
363
364            # Accumulate this byte's contribution to the code point
365            $char_ord <<= 6;
366            $char_ord |= ($b_ord & 0x3F);
367          }
368
369          # Here, the sequence that formed this code point was valid UTF-8,
370          # so add the completed character to the output
371          $copy .= chr $char_ord;
372        } # End of loop through line
373
374        # Delete the dummy first character
375        $copy = substr($copy, 1);
376      }
377
378      # Here, $copy is legal UTF-8.
379
380      # If it can't be legal CP1252, no need to look further.  (These bytes
381      # aren't valid in CP1252.)  This test could have been placed higher in
382      # the code, but it seemed wrong to set the encoding to UTF-8 without
383      # making sure that the very first instance is well-formed.  But what if
384      # it isn't legal CP1252 either?  We have to choose one or the other, and
385      # It seems safer to favor the single-byte encoding over the multi-byte.
386      goto set_utf8 if ord("A") == 65 && $line =~ /[\x81\x8D\x8F\x90\x9D]/;
387
388      # The C1 controls are not likely to appear in pod
389      goto set_1252 if ord("A") == 65 && $copy =~ /[\x80-\x9F]/;
390
391      # Nor are surrogates nor unassigned, nor deprecated.
392      DEBUG > 8 and print STDERR __LINE__, ": $copy: surrogate\n" if $copy =~ $cs_re;
393      goto set_1252 if $cs_re && $copy =~ $cs_re;
394      DEBUG > 8 and print STDERR __LINE__, ": $copy: unassigned\n" if $cn_re && $copy =~ $cn_re;
395      goto set_1252 if $cn_re && $copy =~ $cn_re;
396      DEBUG > 8 and print STDERR __LINE__, ": $copy: deprecated\n" if $copy =~ $deprecated_re;
397      goto set_1252 if $copy =~ $deprecated_re;
398
399      # Nor are rare code points.  But this is hard to determine.  khw
400      # believes that IPA characters and the modifier letters are unlikely to
401      # be in pod (and certainly very unlikely to be the in the first line in
402      # the pod containing non-ASCII)
403      DEBUG > 8 and print STDERR __LINE__, ": $copy: rare\n" if $copy =~ $rare_blocks_re;
404      goto set_1252 if $rare_blocks_re && $copy =~ $rare_blocks_re;
405
406      # The first Unicode version included essentially every Latin character
407      # in modern usage.  So, a Latin character not in the first release will
408      # unlikely be in pod.
409      DEBUG > 8 and print STDERR __LINE__, ": $copy: later_latin\n" if $later_latin_re && $copy =~ $later_latin_re;
410      goto set_1252 if $later_latin_re && $copy =~ $later_latin_re;
411
412      # On perls that handle script runs, if the UTF-8 interpretation yields
413      # a single script, we guess UTF-8, otherwise just having a mixture of
414      # scripts is suspicious, so guess CP1252.  We first strip off, as best
415      # we can, the ASCII characters that look like they are pod directives,
416      # as these would always show as mixed with non-Latin text.
417      $copy =~ s/$pod_chars_re//g;
418
419      if ($script_run_re) {
420        goto set_utf8 if $copy =~ $script_run_re;
421        DEBUG > 8 and print STDERR __LINE__, ":  not script run\n";
422        goto set_1252;
423      }
424
425      # Even without script runs, but on recent enough perls and Unicodes, we
426      # can check if there is a mixture of both Latin and non-Latin.  Again,
427      # having a mixture of scripts is suspicious, so assume CP1252
428
429      # If it's all non-Latin, there is no CP1252, as that is Latin
430      # characters and punct, etc.
431      DEBUG > 8 and print STDERR __LINE__, ": $copy: not latin\n" if $copy !~ $latin_re;
432      goto set_utf8 if $copy !~ $latin_re;
433
434      DEBUG > 8 and print STDERR __LINE__, ": $copy: all latin\n" if $copy =~ $every_char_is_latin_re;
435      goto set_utf8 if $copy =~ $every_char_is_latin_re;
436
437      DEBUG > 8 and print STDERR __LINE__, ": $copy: mixed\n";
438
439     set_1252:
440      DEBUG > 9 and print STDERR __LINE__, ": $copy: is 1252\n";
441      $encoding = 'CP1252';
442      goto done_set;
443
444     set_utf8:
445      DEBUG > 9 and print STDERR __LINE__, ": $copy: is UTF-8\n";
446      $encoding = 'UTF-8';
447
448     done_set:
449      $self->_handle_encoding_line( "=encoding $encoding" );
450      delete $self->{'_processed_encoding'};
451      $self->{'_transcoder'} && $self->{'_transcoder'}->($line);
452
453      my ($word) = $line =~ /(\S*$non_ascii_re\S*)/;
454
455      $self->whine(
456        $self->{'line_count'},
457        "Non-ASCII character seen before =encoding in '$word'. Assuming $encoding"
458      );
459    }
460
461    DEBUG > 5 and print STDERR "# Parsing line: [$line]\n";
462
463    if(!$self->{'in_pod'}) {
464      if($line =~ m/^=([a-zA-Z][a-zA-Z0-9]*)(?:\s|$)/s) {
465        if($1 eq 'cut') {
466          $self->scream(
467            $self->{'line_count'},
468            "=cut found outside a pod block.  Skipping to next block."
469          );
470
471          ## Before there were errata sections in the world, it was
472          ## least-pessimal to abort processing the file.  But now we can
473          ## just barrel on thru (but still not start a pod block).
474          #splice @_;
475          #push @_, undef;
476
477          next;
478        } else {
479          $self->{'in_pod'} = $self->{'start_of_pod_block'}
480                            = $self->{'last_was_blank'}     = 1;
481          # And fall thru to the pod-mode block further down
482        }
483      } else {
484        DEBUG > 5 and print STDERR "# It's a code-line.\n";
485        $code_handler->(map $_, $line, $self->{'line_count'}, $self)
486         if $code_handler;
487        # Note: this may cause code to be processed out of order relative
488        #  to pods, but in order relative to cuts.
489
490        # Note also that we haven't yet applied the transcoding to $line
491        #  by time we call $code_handler!
492
493        if( $line =~ m/^#\s*line\s+(\d+)\s*(?:\s"([^"]+)")?\s*$/ ) {
494          # That RE is from perlsyn, section "Plain Old Comments (Not!)",
495          #$fname = $2 if defined $2;
496          #DEBUG > 1 and defined $2 and print STDERR "# Setting fname to \"$fname\"\n";
497          DEBUG > 1 and print STDERR "# Setting nextline to $1\n";
498          $self->{'line_count'} = $1 - 1;
499        }
500
501        next;
502      }
503    }
504
505    # . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
506    # Else we're in pod mode:
507
508    # Apply any necessary transcoding:
509    $self->{'_transcoder'} && $self->{'_transcoder'}->($line);
510
511    # HERE WE CATCH =encoding EARLY!
512    if( $line =~ m/^=encoding\s+\S+\s*$/s ) {
513      next if $self->parse_characters;   # Ignore this line
514      $line = $self->_handle_encoding_line( $line );
515    }
516
517    if($line =~ m/^=cut/s) {
518      # here ends the pod block, and therefore the previous pod para
519      DEBUG > 1 and print STDERR "Noting =cut at line ${$self}{'line_count'}\n";
520      $self->{'in_pod'} = 0;
521      # ++$self->{'pod_para_count'};
522      $self->_ponder_paragraph_buffer();
523       # by now it's safe to consider the previous paragraph as done.
524      DEBUG > 6 and print STDERR "Processing any cut handler, line ${$self}{'line_count'}\n";
525      $cut_handler->(map $_, $line, $self->{'line_count'}, $self)
526       if $cut_handler;
527
528      # TODO: add to docs: Note: this may cause cuts to be processed out
529      #  of order relative to pods, but in order relative to code.
530
531    } elsif($line =~ m/^(\s*)$/s) {  # it's a blank line
532      if (defined $1 and $1 =~ /[^\S\r\n]/) { # it's a white line
533        $wl_handler->(map $_, $line, $self->{'line_count'}, $self)
534          if $wl_handler;
535      }
536
537      if(!$self->{'start_of_pod_block'} and @$paras and $paras->[-1][0] eq '~Verbatim') {
538        DEBUG > 1 and print STDERR "Saving blank line at line ${$self}{'line_count'}\n";
539        push @{$paras->[-1]}, $line;
540      }  # otherwise it's not interesting
541
542      if(!$self->{'start_of_pod_block'} and !$self->{'last_was_blank'}) {
543        DEBUG > 1 and print STDERR "Noting para ends with blank line at ${$self}{'line_count'}\n";
544      }
545
546      $self->{'last_was_blank'} = 1;
547
548    } elsif($self->{'last_was_blank'}) {  # A non-blank line starting a new para...
549
550      if($line =~ m/^(=[a-zA-Z][a-zA-Z0-9]*)(\s+|$)(.*)/s) {
551        # THIS IS THE ONE PLACE WHERE WE CONSTRUCT NEW DIRECTIVE OBJECTS
552        my $new = [$1, {'start_line' => $self->{'line_count'}}, $3];
553        $new->[1]{'~orig_spacer'} = $2 if $2 && $2 ne " ";
554         # Note that in "=head1 foo", the WS is lost.
555         # Example: ['=head1', {'start_line' => 123}, ' foo']
556
557        ++$self->{'pod_para_count'};
558
559        $self->_ponder_paragraph_buffer();
560         # by now it's safe to consider the previous paragraph as done.
561
562        push @$paras, $new; # the new incipient paragraph
563        DEBUG > 1 and print STDERR "Starting new ${$paras}[-1][0] para at line ${$self}{'line_count'}\n";
564
565      } elsif($line =~ m/^\s/s) {
566
567        if(!$self->{'start_of_pod_block'} and @$paras and $paras->[-1][0] eq '~Verbatim') {
568          DEBUG > 1 and print STDERR "Resuming verbatim para at line ${$self}{'line_count'}\n";
569          push @{$paras->[-1]}, $line;
570        } else {
571          ++$self->{'pod_para_count'};
572          $self->_ponder_paragraph_buffer();
573           # by now it's safe to consider the previous paragraph as done.
574          DEBUG > 1 and print STDERR "Starting verbatim para at line ${$self}{'line_count'}\n";
575          push @$paras, ['~Verbatim', {'start_line' => $self->{'line_count'}}, $line];
576        }
577      } else {
578        ++$self->{'pod_para_count'};
579        $self->_ponder_paragraph_buffer();
580         # by now it's safe to consider the previous paragraph as done.
581        push @$paras, ['~Para',  {'start_line' => $self->{'line_count'}}, $line];
582        DEBUG > 1 and print STDERR "Starting plain para at line ${$self}{'line_count'}\n";
583      }
584      $self->{'last_was_blank'} = $self->{'start_of_pod_block'} = 0;
585
586    } else {
587      # It's a non-blank line /continuing/ the current para
588      if(@$paras) {
589        DEBUG > 2 and print STDERR "Line ${$self}{'line_count'} continues current paragraph\n";
590        push @{$paras->[-1]}, $line;
591      } else {
592        # Unexpected case!
593        die "Continuing a paragraph but \@\$paras is empty?";
594      }
595      $self->{'last_was_blank'} = $self->{'start_of_pod_block'} = 0;
596    }
597
598  } # ends the big while loop
599
600  DEBUG > 1 and print STDERR (pretty(@$paras), "\n");
601  return $self;
602}
603
604#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
605
606sub _handle_encoding_line {
607  my($self, $line) = @_;
608
609  return if $self->parse_characters;
610
611  # The point of this routine is to set $self->{'_transcoder'} as indicated.
612
613  return $line unless $line =~ m/^=encoding\s+(\S+)\s*$/s;
614  DEBUG > 1 and print STDERR "Found an encoding line \"=encoding $1\"\n";
615
616  my $e    = $1;
617  my $orig = $e;
618  push @{ $self->{'encoding_command_reqs'} }, "=encoding $orig";
619
620  my $enc_error;
621
622  # Cf.   perldoc Encode   and   perldoc Encode::Supported
623
624  require Pod::Simple::Transcode;
625
626  if( $self->{'encoding'} ) {
627    my $norm_current = $self->{'encoding'};
628    my $norm_e = $e;
629    foreach my $that ($norm_current, $norm_e) {
630      $that =  lc($that);
631      $that =~ s/[-_]//g;
632    }
633    if($norm_current eq $norm_e) {
634      DEBUG > 1 and print STDERR "The '=encoding $orig' line is ",
635       "redundant.  ($norm_current eq $norm_e).  Ignoring.\n";
636      $enc_error = '';
637       # But that doesn't necessarily mean that the earlier one went okay
638    } else {
639      $enc_error = "Encoding is already set to " . $self->{'encoding'};
640      DEBUG > 1 and print STDERR $enc_error;
641    }
642  } elsif (
643    # OK, let's turn on the encoding
644    do {
645      DEBUG > 1 and print STDERR " Setting encoding to $e\n";
646      $self->{'encoding'} = $e;
647      1;
648    }
649    and $e eq 'HACKRAW'
650  ) {
651    DEBUG and print STDERR " Putting in HACKRAW (no-op) encoding mode.\n";
652
653  } elsif( Pod::Simple::Transcode::->encoding_is_available($e) ) {
654
655    die($enc_error = "WHAT? _transcoder is already set?!")
656     if $self->{'_transcoder'};   # should never happen
657    require Pod::Simple::Transcode;
658    $self->{'_transcoder'} = Pod::Simple::Transcode::->make_transcoder($e);
659    eval {
660      my @x = ('', "abc", "123");
661      $self->{'_transcoder'}->(@x);
662    };
663    $@ && die( $enc_error =
664      "Really unexpected error setting up encoding $e: $@\nAborting"
665    );
666    $self->{'detected_encoding'} = $e;
667
668  } else {
669    my @supported = Pod::Simple::Transcode::->all_encodings;
670
671    # Note unsupported, and complain
672    DEBUG and print STDERR " Encoding [$e] is unsupported.",
673      "\nSupporteds: @supported\n";
674    my $suggestion = '';
675
676    # Look for a near match:
677    my $norm = lc($e);
678    $norm =~ tr[-_][]d;
679    my $n;
680    foreach my $enc (@supported) {
681      $n = lc($enc);
682      $n =~ tr[-_][]d;
683      next unless $n eq $norm;
684      $suggestion = "  (Maybe \"$e\" should be \"$enc\"?)";
685      last;
686    }
687    my $encmodver = Pod::Simple::Transcode::->encmodver;
688    $enc_error = join '' =>
689      "This document probably does not appear as it should, because its ",
690      "\"=encoding $e\" line calls for an unsupported encoding.",
691      $suggestion, "  [$encmodver\'s supported encodings are: @supported]"
692    ;
693
694    $self->scream( $self->{'line_count'}, $enc_error );
695  }
696  push @{ $self->{'encoding_command_statuses'} }, $enc_error;
697  if (defined($self->{'_processed_encoding'})) {
698    # Double declaration.
699    $self->scream( $self->{'line_count'}, 'Cannot have multiple =encoding directives');
700  }
701  $self->{'_processed_encoding'} = $orig;
702
703  return $line;
704}
705
706# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
707
708sub _handle_encoding_second_level {
709  # By time this is called, the encoding (if well formed) will already
710  #  have been acted on.
711  my($self, $para) = @_;
712  my @x = @$para;
713  my $content = join ' ', splice @x, 2;
714  $content =~ s/^\s+//s;
715  $content =~ s/\s+$//s;
716
717  DEBUG > 2 and print STDERR "Ogling encoding directive: =encoding $content\n";
718
719  if (defined($self->{'_processed_encoding'})) {
720    #if($content ne $self->{'_processed_encoding'}) {
721    #  Could it happen?
722    #}
723    delete $self->{'_processed_encoding'};
724    # It's already been handled.  Check for errors.
725    if(! $self->{'encoding_command_statuses'} ) {
726      DEBUG > 2 and print STDERR " CRAZY ERROR: It wasn't really handled?!\n";
727    } elsif( $self->{'encoding_command_statuses'}[-1] ) {
728      $self->whine( $para->[1]{'start_line'},
729        sprintf "Couldn't do %s: %s",
730          $self->{'encoding_command_reqs'  }[-1],
731          $self->{'encoding_command_statuses'}[-1],
732      );
733    } else {
734      DEBUG > 2 and print STDERR " (Yup, it was successfully handled already.)\n";
735    }
736
737  } else {
738    # Otherwise it's a syntax error
739    $self->whine( $para->[1]{'start_line'},
740      "Invalid =encoding syntax: $content"
741    );
742  }
743
744  return;
745}
746
747#~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`~`
748
749{
750my $m = -321;   # magic line number
751
752sub _gen_errata {
753  my $self = $_[0];
754  # Return 0 or more fake-o paragraphs explaining the accumulated
755  #  errors on this document.
756
757  return() unless $self->{'errata'} and keys %{$self->{'errata'}};
758
759  my @out;
760
761  foreach my $line (sort {$a <=> $b} keys %{$self->{'errata'}}) {
762    push @out,
763      ['=item', {'start_line' => $m}, "Around line $line:"],
764      map( ['~Para', {'start_line' => $m, '~cooked' => 1},
765        #['~Top', {'start_line' => $m},
766        $_
767        #]
768        ],
769        @{$self->{'errata'}{$line}}
770      )
771    ;
772  }
773
774  # TODO: report of unknown entities? unrenderable characters?
775
776  unshift @out,
777    ['=head1', {'start_line' => $m, 'errata' => 1}, 'POD ERRORS'],
778    ['~Para', {'start_line' => $m, '~cooked' => 1, 'errata' => 1},
779     "Hey! ",
780     ['B', {},
781      'The above document had some coding errors, which are explained below:'
782     ]
783    ],
784    ['=over',  {'start_line' => $m, 'errata' => 1}, ''],
785  ;
786
787  push @out,
788    ['=back',  {'start_line' => $m, 'errata' => 1}, ''],
789  ;
790
791  DEBUG and print STDERR "\n<<\n", pretty(\@out), "\n>>\n\n";
792
793  return @out;
794}
795
796}
797
798#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
799
800##############################################################################
801##
802##  stop reading now stop reading now stop reading now stop reading now stop
803##
804##                         HERE IT BECOMES REALLY SCARY
805##
806##  stop reading now stop reading now stop reading now stop reading now stop
807##
808##############################################################################
809
810sub _ponder_paragraph_buffer {
811
812  # Para-token types as found in the buffer.
813  #   ~Verbatim, ~Para, ~end, =head1..4, =for, =begin, =end,
814  #   =over, =back, =item
815  #   and the null =pod (to be complained about if over one line)
816  #
817  # "~data" paragraphs are something we generate at this level, depending on
818  # a currently open =over region
819
820  # Events fired:  Begin and end for:
821  #                   directivename (like head1 .. head4), item, extend,
822  #                   for (from =begin...=end, =for),
823  #                   over-bullet, over-number, over-text, over-block,
824  #                   item-bullet, item-number, item-text,
825  #                   Document,
826  #                   Data, Para, Verbatim
827  #                   B, C, longdirname (TODO -- wha?), etc. for all directives
828  #
829
830  my $self = $_[0];
831  my $paras;
832  return unless @{$paras = $self->{'paras'}};
833  my $curr_open = ($self->{'curr_open'} ||= []);
834
835  my $scratch;
836
837  DEBUG > 10 and print STDERR "# Paragraph buffer: <<", pretty($paras), ">>\n";
838
839  # We have something in our buffer.  So apparently the document has started.
840  unless($self->{'doc_has_started'}) {
841    $self->{'doc_has_started'} = 1;
842
843    my $starting_contentless;
844    $starting_contentless =
845     (
846       !@$curr_open
847       and @$paras and ! grep $_->[0] ne '~end', @$paras
848        # i.e., if the paras is all ~ends
849     )
850    ;
851    DEBUG and print STDERR "# Starting ",
852      $starting_contentless ? 'contentless' : 'contentful',
853      " document\n"
854    ;
855
856    $self->_handle_element_start(
857      ($scratch = 'Document'),
858      {
859        'start_line' => $paras->[0][1]{'start_line'},
860        $starting_contentless ? ( 'contentless' => 1 ) : (),
861      },
862    );
863  }
864
865  my($para, $para_type);
866  while(@$paras) {
867
868    # If a directive, assume it's legal; subtract below if found not to be
869    $seen_legal_directive++ if $paras->[0][0] =~ /^=/;
870
871    last if      @$paras == 1
872            and (    $paras->[0][0] eq '=over'
873                 or  $paras->[0][0] eq '=item'
874                 or ($paras->[0][0] eq '~Verbatim' and $self->{'in_pod'}));
875    # Those're the three kinds of paragraphs that require lookahead.
876    #   Actually, an "=item Foo" inside an <over type=text> region
877    #   and any =item inside an <over type=block> region (rare)
878    #   don't require any lookahead, but all others (bullets
879    #   and numbers) do.
880    # The verbatim is different from the other two, because those might be
881    # like:
882    #
883    #   =item
884    #   ...
885    #   =cut
886    #   ...
887    #   =item
888    #
889    # The =cut here finishes the paragraph but doesn't terminate the =over
890    # they should be in. (khw apologizes that he didn't comment at the time
891    # why the 'in_pod' works, and no longer remembers why, and doesn't think
892    # it is currently worth the effort to re-figure it out.)
893
894# TODO: whinge about many kinds of directives in non-resolving =for regions?
895# TODO: many?  like what?  =head1 etc?
896
897    $para = shift @$paras;
898    $para_type = $para->[0];
899
900    DEBUG > 1 and print STDERR "Pondering a $para_type paragraph, given the stack: (",
901      $self->_dump_curr_open(), ")\n";
902
903    if($para_type eq '=for') {
904      next if $self->_ponder_for($para,$curr_open,$paras);
905
906    } elsif($para_type eq '=begin') {
907      next if $self->_ponder_begin($para,$curr_open,$paras);
908
909    } elsif($para_type eq '=end') {
910      next if $self->_ponder_end($para,$curr_open,$paras);
911
912    } elsif($para_type eq '~end') { # The virtual end-document signal
913      next if $self->_ponder_doc_end($para,$curr_open,$paras);
914    }
915
916
917    # ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
918    #~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
919    if(grep $_->[1]{'~ignore'}, @$curr_open) {
920      DEBUG > 1 and
921       print STDERR "Skipping $para_type paragraph because in ignore mode.\n";
922      next;
923    }
924    #~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
925    # ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
926
927    if($para_type eq '=pod') {
928      $self->_ponder_pod($para,$curr_open,$paras);
929
930    } elsif($para_type eq '=over') {
931      next if $self->_ponder_over($para,$curr_open,$paras);
932
933    } elsif($para_type eq '=back') {
934      next if $self->_ponder_back($para,$curr_open,$paras);
935
936    } else {
937
938      # All non-magical codes!!!
939
940      # Here we start using $para_type for our own twisted purposes, to
941      #  mean how it should get treated, not as what the element name
942      #  should be.
943
944      DEBUG > 1 and print STDERR "Pondering non-magical $para_type\n";
945
946      my $i;
947
948      # Enforce some =headN discipline
949      if($para_type =~ m/^=head\d$/s
950         and ! $self->{'accept_heads_anywhere'}
951         and @$curr_open
952         and $curr_open->[-1][0] eq '=over'
953      ) {
954        DEBUG > 2 and print STDERR "'=$para_type' inside an '=over'!\n";
955        $self->whine(
956          $para->[1]{'start_line'},
957          "You forgot a '=back' before '$para_type'"
958        );
959        unshift @$paras, ['=back', {}, ''], $para;   # close the =over
960        next;
961      }
962
963
964      if($para_type eq '=item') {
965
966        my $over;
967        unless(@$curr_open and
968               $over = (grep { $_->[0] eq '=over' } @$curr_open)[-1]) {
969          $self->whine(
970            $para->[1]{'start_line'},
971            "'=item' outside of any '=over'"
972          );
973          unshift @$paras,
974            ['=over', {'start_line' => $para->[1]{'start_line'}}, ''],
975            $para
976          ;
977          next;
978        }
979
980
981        my $over_type = $over->[1]{'~type'};
982
983        if(!$over_type) {
984          # Shouldn't happen1
985          die "Typeless over in stack, starting at line "
986           . $over->[1]{'start_line'};
987
988        } elsif($over_type eq 'block') {
989          unless($curr_open->[-1][1]{'~bitched_about'}) {
990            $curr_open->[-1][1]{'~bitched_about'} = 1;
991            $self->whine(
992              $curr_open->[-1][1]{'start_line'},
993              "You can't have =items (as at line "
994              . $para->[1]{'start_line'}
995              . ") unless the first thing after the =over is an =item"
996            );
997          }
998          # Just turn it into a paragraph and reconsider it
999          $para->[0] = '~Para';
1000          unshift @$paras, $para;
1001          next;
1002
1003        } elsif($over_type eq 'text') {
1004          my $item_type = $self->_get_item_type($para);
1005            # That kills the content of the item if it's a number or bullet.
1006          DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1007
1008          if($item_type eq 'text') {
1009            # Nothing special needs doing for 'text'
1010          } elsif($item_type eq 'number' or $item_type eq 'bullet') {
1011            $self->whine(
1012              $para->[1]{'start_line'},
1013              "Expected text after =item, not a $item_type"
1014            );
1015            # Undo our clobbering:
1016            push @$para, $para->[1]{'~orig_content'};
1017            delete $para->[1]{'number'};
1018             # Only a PROPER item-number element is allowed
1019             #  to have a number attribute.
1020          } else {
1021            die "Unhandled item type $item_type"; # should never happen
1022          }
1023
1024          # =item-text thingies don't need any assimilation, it seems.
1025
1026        } elsif($over_type eq 'number') {
1027          my $item_type = $self->_get_item_type($para);
1028            # That kills the content of the item if it's a number or bullet.
1029          DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1030
1031          my $expected_value = ++ $curr_open->[-1][1]{'~counter'};
1032
1033          if($item_type eq 'bullet') {
1034            # Hm, it's not numeric.  Correct for this.
1035            $para->[1]{'number'} = $expected_value;
1036            $self->whine(
1037              $para->[1]{'start_line'},
1038              "Expected '=item $expected_value'"
1039            );
1040            push @$para, $para->[1]{'~orig_content'};
1041              # restore the bullet, blocking the assimilation of next para
1042
1043          } elsif($item_type eq 'text') {
1044            # Hm, it's not numeric.  Correct for this.
1045            $para->[1]{'number'} = $expected_value;
1046            $self->whine(
1047              $para->[1]{'start_line'},
1048              "Expected '=item $expected_value'"
1049            );
1050            # Text content will still be there and will block next ~Para
1051
1052          } elsif($item_type ne 'number') {
1053            die "Unknown item type $item_type"; # should never happen
1054
1055          } elsif($expected_value == $para->[1]{'number'}) {
1056            DEBUG > 1 and print STDERR " Numeric item has the expected value of $expected_value\n";
1057
1058          } else {
1059            DEBUG > 1 and print STDERR " Numeric item has ", $para->[1]{'number'},
1060             " instead of the expected value of $expected_value\n";
1061            $self->whine(
1062              $para->[1]{'start_line'},
1063              "You have '=item " . $para->[1]{'number'} .
1064              "' instead of the expected '=item $expected_value'"
1065            );
1066            $para->[1]{'number'} = $expected_value;  # correcting!!
1067          }
1068
1069          if(@$para == 2) {
1070            # For the cases where we /didn't/ push to @$para
1071            if($paras->[0][0] eq '~Para') {
1072              DEBUG and print STDERR "Assimilating following ~Para content into $over_type item\n";
1073              push @$para, splice @{shift @$paras},2;
1074            } else {
1075              DEBUG and print STDERR "Can't assimilate following ", $paras->[0][0], "\n";
1076              push @$para, '';  # Just so it's not contentless
1077            }
1078          }
1079
1080
1081        } elsif($over_type eq 'bullet') {
1082          my $item_type = $self->_get_item_type($para);
1083            # That kills the content of the item if it's a number or bullet.
1084          DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1085
1086          if($item_type eq 'bullet') {
1087            # as expected!
1088
1089            if( $para->[1]{'~_freaky_para_hack'} ) {
1090              DEBUG and print STDERR "Accomodating '=item * Foo' tolerance hack.\n";
1091              push @$para, $para->[1]{'~_freaky_para_hack'};
1092            }
1093
1094          } elsif($item_type eq 'number') {
1095            $self->whine(
1096              $para->[1]{'start_line'},
1097              "Expected '=item *'"
1098            );
1099            push @$para, $para->[1]{'~orig_content'};
1100             # and block assimilation of the next paragraph
1101            delete $para->[1]{'number'};
1102             # Only a PROPER item-number element is allowed
1103             #  to have a number attribute.
1104          } elsif($item_type eq 'text') {
1105            $self->whine(
1106              $para->[1]{'start_line'},
1107              "Expected '=item *'"
1108            );
1109             # But doesn't need processing.  But it'll block assimilation
1110             #  of the next para.
1111          } else {
1112            die "Unhandled item type $item_type"; # should never happen
1113          }
1114
1115          if(@$para == 2) {
1116            # For the cases where we /didn't/ push to @$para
1117            if($paras->[0][0] eq '~Para') {
1118              DEBUG and print STDERR "Assimilating following ~Para content into $over_type item\n";
1119              push @$para, splice @{shift @$paras},2;
1120            } else {
1121              DEBUG and print STDERR "Can't assimilate following ", $paras->[0][0], "\n";
1122              push @$para, '';  # Just so it's not contentless
1123            }
1124          }
1125
1126        } else {
1127          die "Unhandled =over type \"$over_type\"?";
1128          # Shouldn't happen!
1129        }
1130
1131        $para_type = 'Plain';
1132        $para->[0] .= '-' . $over_type;
1133        # Whew.  Now fall thru and process it.
1134
1135
1136      } elsif($para_type eq '=extend') {
1137        # Well, might as well implement it here.
1138        $self->_ponder_extend($para);
1139        next;  # and skip
1140      } elsif($para_type eq '=encoding') {
1141        # Not actually acted on here, but we catch errors here.
1142        $self->_handle_encoding_second_level($para);
1143        next unless $self->keep_encoding_directive;
1144        $para_type = 'Plain';
1145      } elsif($para_type eq '~Verbatim') {
1146        $para->[0] = 'Verbatim';
1147        $para_type = '?Verbatim';
1148      } elsif($para_type eq '~Para') {
1149        $para->[0] = 'Para';
1150        $para_type = '?Plain';
1151      } elsif($para_type eq 'Data') {
1152        $para->[0] = 'Data';
1153        $para_type = '?Data';
1154      } elsif( $para_type =~ s/^=//s
1155        and defined( $para_type = $self->{'accept_directives'}{$para_type} )
1156      ) {
1157        DEBUG > 1 and print STDERR " Pondering known directive ${$para}[0] as $para_type\n";
1158      } else {
1159        # An unknown directive!
1160        $seen_legal_directive--;
1161        DEBUG > 1 and printf STDERR "Unhandled directive %s (Handled: %s)\n",
1162         $para->[0], join(' ', sort keys %{$self->{'accept_directives'}} )
1163        ;
1164        $self->whine(
1165          $para->[1]{'start_line'},
1166          "Unknown directive: $para->[0]"
1167        );
1168
1169        # And maybe treat it as text instead of just letting it go?
1170        next;
1171      }
1172
1173      if($para_type =~ s/^\?//s) {
1174        if(! @$curr_open) {  # usual case
1175          DEBUG and print STDERR "Treating $para_type paragraph as such because stack is empty.\n";
1176        } else {
1177          my @fors = grep $_->[0] eq '=for', @$curr_open;
1178          DEBUG > 1 and print STDERR "Containing fors: ",
1179            join(',', map $_->[1]{'target'}, @fors), "\n";
1180
1181          if(! @fors) {
1182            DEBUG and print STDERR "Treating $para_type paragraph as such because stack has no =for's\n";
1183
1184          #} elsif(grep $_->[1]{'~resolve'}, @fors) {
1185          #} elsif(not grep !$_->[1]{'~resolve'}, @fors) {
1186          } elsif( $fors[-1][1]{'~resolve'} ) {
1187            # Look to the immediately containing for
1188
1189            if($para_type eq 'Data') {
1190              DEBUG and print STDERR "Treating Data paragraph as Plain/Verbatim because the containing =for ($fors[-1][1]{'target'}) is a resolver\n";
1191              $para->[0] = 'Para';
1192              $para_type = 'Plain';
1193            } else {
1194              DEBUG and print STDERR "Treating $para_type paragraph as such because the containing =for ($fors[-1][1]{'target'}) is a resolver\n";
1195            }
1196          } else {
1197            DEBUG and print STDERR "Treating $para_type paragraph as Data because the containing =for ($fors[-1][1]{'target'}) is a non-resolver\n";
1198            $para->[0] = $para_type = 'Data';
1199          }
1200        }
1201      }
1202
1203      #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1204      if($para_type eq 'Plain') {
1205        $self->_ponder_Plain($para);
1206      } elsif($para_type eq 'Verbatim') {
1207        $self->_ponder_Verbatim($para);
1208      } elsif($para_type eq 'Data') {
1209        $self->_ponder_Data($para);
1210      } else {
1211        die "\$para type is $para_type -- how did that happen?";
1212        # Shouldn't happen.
1213      }
1214
1215      #~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1216      $para->[0] =~ s/^[~=]//s;
1217
1218      DEBUG and print STDERR "\n", pretty($para), "\n";
1219
1220      # traverse the treelet (which might well be just one string scalar)
1221      $self->{'content_seen'} ||= 1 if   $seen_legal_directive
1222                                    && ! $self->{'~tried_gen_errata'};
1223      $self->_traverse_treelet_bit(@$para);
1224    }
1225  }
1226
1227  return;
1228}
1229
1230###########################################################################
1231# The sub-ponderers...
1232
1233
1234
1235sub _ponder_for {
1236  my ($self,$para,$curr_open,$paras) = @_;
1237
1238  # Fake it out as a begin/end
1239  my $target;
1240
1241  if(grep $_->[1]{'~ignore'}, @$curr_open) {
1242    DEBUG > 1 and print STDERR "Ignoring ignorable =for\n";
1243    return 1;
1244  }
1245
1246  for(my $i = 2; $i < @$para; ++$i) {
1247    if($para->[$i] =~ s/^\s*(\S+)\s*//s) {
1248      $target = $1;
1249      last;
1250    }
1251  }
1252  unless(defined $target) {
1253    $self->whine(
1254      $para->[1]{'start_line'},
1255      "=for without a target?"
1256    );
1257    return 1;
1258  }
1259  DEBUG > 1 and
1260   print STDERR "Faking out a =for $target as a =begin $target / =end $target\n";
1261
1262  $para->[0] = 'Data';
1263
1264  unshift @$paras,
1265    ['=begin',
1266      {'start_line' => $para->[1]{'start_line'}, '~really' => '=for'},
1267      $target,
1268    ],
1269    $para,
1270    ['=end',
1271      {'start_line' => $para->[1]{'start_line'}, '~really' => '=for'},
1272      $target,
1273    ],
1274  ;
1275
1276  return 1;
1277}
1278
1279sub _ponder_begin {
1280  my ($self,$para,$curr_open,$paras) = @_;
1281  my $content = join ' ', splice @$para, 2;
1282  $content =~ s/^\s+//s;
1283  $content =~ s/\s+$//s;
1284  unless(length($content)) {
1285    $self->whine(
1286      $para->[1]{'start_line'},
1287      "=begin without a target?"
1288    );
1289    DEBUG and print STDERR "Ignoring targetless =begin\n";
1290    return 1;
1291  }
1292
1293  my ($target, $title) = $content =~ m/^(\S+)\s*(.*)$/;
1294  $para->[1]{'title'} = $title if ($title);
1295  $para->[1]{'target'} = $target;  # without any ':'
1296  $content = $target; # strip off the title
1297
1298  $content =~ s/^:!/!:/s;
1299  my $neg;  # whether this is a negation-match
1300  $neg = 1        if $content =~ s/^!//s;
1301  my $to_resolve;  # whether to process formatting codes
1302  $to_resolve = 1 if $content =~ s/^://s;
1303
1304  my $dont_ignore; # whether this target matches us
1305
1306  foreach my $target_name (
1307    split(',', $content, -1),
1308    $neg ? () : '*'
1309  ) {
1310    DEBUG > 2 and
1311     print STDERR " Considering whether =begin $content matches $target_name\n";
1312    next unless $self->{'accept_targets'}{$target_name};
1313
1314    DEBUG > 2 and
1315     print STDERR "  It DOES match the acceptable target $target_name!\n";
1316    $to_resolve = 1
1317      if $self->{'accept_targets'}{$target_name} eq 'force_resolve';
1318    $dont_ignore = 1;
1319    $para->[1]{'target_matching'} = $target_name;
1320    last; # stop looking at other target names
1321  }
1322
1323  if($neg) {
1324    if( $dont_ignore ) {
1325      $dont_ignore = '';
1326      delete $para->[1]{'target_matching'};
1327      DEBUG > 2 and print STDERR " But the leading ! means that this is a NON-match!\n";
1328    } else {
1329      $dont_ignore = 1;
1330      $para->[1]{'target_matching'} = '!';
1331      DEBUG > 2 and print STDERR " But the leading ! means that this IS a match!\n";
1332    }
1333  }
1334
1335  $para->[0] = '=for';  # Just what we happen to call these, internally
1336  $para->[1]{'~really'} ||= '=begin';
1337  $para->[1]{'~ignore'}   = (! $dont_ignore) || 0;
1338  $para->[1]{'~resolve'}  = $to_resolve || 0;
1339
1340  DEBUG > 1 and print STDERR " Making note to ", $dont_ignore ? 'not ' : '',
1341    "ignore contents of this region\n";
1342  DEBUG > 1 and $dont_ignore and print STDERR " Making note to treat contents as ",
1343    ($to_resolve ? 'verbatim/plain' : 'data'), " paragraphs\n";
1344  DEBUG > 1 and print STDERR " (Stack now: ", $self->_dump_curr_open(), ")\n";
1345
1346  push @$curr_open, $para;
1347  if(!$dont_ignore or scalar grep $_->[1]{'~ignore'}, @$curr_open) {
1348    DEBUG > 1 and print STDERR "Ignoring ignorable =begin\n";
1349  } else {
1350    $self->{'content_seen'} ||= 1 unless $self->{'~tried_gen_errata'};
1351    $self->_handle_element_start((my $scratch='for'), $para->[1]);
1352  }
1353
1354  return 1;
1355}
1356
1357sub _ponder_end {
1358  my ($self,$para,$curr_open,$paras) = @_;
1359  my $content = join ' ', splice @$para, 2;
1360  $content =~ s/^\s+//s;
1361  $content =~ s/\s+$//s;
1362  DEBUG and print STDERR "Ogling '=end $content' directive\n";
1363
1364  unless(length($content)) {
1365    $self->whine(
1366      $para->[1]{'start_line'},
1367      "'=end' without a target?" . (
1368        ( @$curr_open and $curr_open->[-1][0] eq '=for' )
1369        ? ( " (Should be \"=end " . $curr_open->[-1][1]{'target'} . '")' )
1370        : ''
1371      )
1372    );
1373    DEBUG and print STDERR "Ignoring targetless =end\n";
1374    return 1;
1375  }
1376
1377  unless($content =~ m/^\S+$/) {  # i.e., unless it's one word
1378    $self->whine(
1379      $para->[1]{'start_line'},
1380      "'=end $content' is invalid.  (Stack: "
1381      . $self->_dump_curr_open() . ')'
1382    );
1383    DEBUG and print STDERR "Ignoring mistargetted =end $content\n";
1384    return 1;
1385  }
1386
1387  unless(@$curr_open and $curr_open->[-1][0] eq '=for') {
1388    $self->whine(
1389      $para->[1]{'start_line'},
1390      "=end $content without matching =begin.  (Stack: "
1391      . $self->_dump_curr_open() . ')'
1392    );
1393    DEBUG and print STDERR "Ignoring mistargetted =end $content\n";
1394    return 1;
1395  }
1396
1397  unless($content eq $curr_open->[-1][1]{'target'}) {
1398    $self->whine(
1399      $para->[1]{'start_line'},
1400      "=end $content doesn't match =begin "
1401      . $curr_open->[-1][1]{'target'}
1402      . ".  (Stack: "
1403      . $self->_dump_curr_open() . ')'
1404    );
1405    DEBUG and print STDERR "Ignoring mistargetted =end $content at line $para->[1]{'start_line'}\n";
1406    return 1;
1407  }
1408
1409  # Else it's okay to close...
1410  if(grep $_->[1]{'~ignore'}, @$curr_open) {
1411    DEBUG > 1 and print STDERR "Not firing any event for this =end $content because in an ignored region\n";
1412    # And that may be because of this to-be-closed =for region, or some
1413    #  other one, but it doesn't matter.
1414  } else {
1415    $curr_open->[-1][1]{'start_line'} = $para->[1]{'start_line'};
1416      # what's that for?
1417
1418    $self->{'content_seen'} ||= 1 unless $self->{'~tried_gen_errata'};
1419    $self->_handle_element_end( my $scratch = 'for', $para->[1]);
1420  }
1421  DEBUG > 1 and print STDERR "Popping $curr_open->[-1][0] $curr_open->[-1][1]{'target'} because of =end $content\n";
1422  pop @$curr_open;
1423
1424  return 1;
1425}
1426
1427sub _ponder_doc_end {
1428  my ($self,$para,$curr_open,$paras) = @_;
1429  if(@$curr_open) { # Deal with things left open
1430    DEBUG and print STDERR "Stack is nonempty at end-document: (",
1431      $self->_dump_curr_open(), ")\n";
1432
1433    DEBUG > 9 and print STDERR "Stack: ", pretty($curr_open), "\n";
1434    unshift @$paras, $self->_closers_for_all_curr_open;
1435    # Make sure there is exactly one ~end in the parastack, at the end:
1436    @$paras = grep $_->[0] ne '~end', @$paras;
1437    push @$paras, $para, $para;
1438     # We need two -- once for the next cycle where we
1439     #  generate errata, and then another to be at the end
1440     #  when that loop back around to process the errata.
1441    return 1;
1442
1443  } else {
1444    DEBUG and print STDERR "Okay, stack is empty now.\n";
1445  }
1446
1447  # Try generating errata section, if applicable
1448  unless($self->{'~tried_gen_errata'}) {
1449    $self->{'~tried_gen_errata'} = 1;
1450    my @extras = $self->_gen_errata();
1451    if(@extras) {
1452      unshift @$paras, @extras;
1453      DEBUG and print STDERR "Generated errata... relooping...\n";
1454      return 1;  # I.e., loop around again to process these fake-o paragraphs
1455    }
1456  }
1457
1458  splice @$paras; # Well, that's that for this paragraph buffer.
1459  DEBUG and print STDERR "Throwing end-document event.\n";
1460
1461  $self->_handle_element_end( my $scratch = 'Document' );
1462  return 1; # Hasta la byebye
1463}
1464
1465sub _ponder_pod {
1466  my ($self,$para,$curr_open,$paras) = @_;
1467  $self->whine(
1468    $para->[1]{'start_line'},
1469    "=pod directives shouldn't be over one line long!  Ignoring all "
1470     . (@$para - 2) . " lines of content"
1471  ) if @$para > 3;
1472
1473  # Content ignored unless 'pod_handler' is set
1474  if (my $pod_handler = $self->{'pod_handler'}) {
1475      my ($line_num, $line) = map $_, $para->[1]{'start_line'}, $para->[2];
1476      $line = $line eq '' ? "=pod" : "=pod $line"; # imitate cut_handler output
1477      $pod_handler->($line, $line_num, $self);
1478  }
1479
1480  # The surrounding methods set content_seen, so let us remain consistent.
1481  # I do not know why it was not here before -- should it not be here?
1482  # $self->{'content_seen'} ||= 1 unless $self->{'~tried_gen_errata'};
1483
1484  return;
1485}
1486
1487sub _ponder_over {
1488  my ($self,$para,$curr_open,$paras) = @_;
1489  return 1 unless @$paras;
1490  my $list_type;
1491
1492  if($paras->[0][0] eq '=item') { # most common case
1493    $list_type = $self->_get_initial_item_type($paras->[0]);
1494
1495  } elsif($paras->[0][0] eq '=back') {
1496    # Ignore empty lists by default
1497    if ($self->{'parse_empty_lists'}) {
1498      $list_type = 'empty';
1499    } else {
1500      shift @$paras;
1501      return 1;
1502    }
1503  } elsif($paras->[0][0] eq '~end') {
1504    $self->whine(
1505      $para->[1]{'start_line'},
1506      "=over is the last thing in the document?!"
1507    );
1508    return 1; # But feh, ignore it.
1509  } else {
1510    $list_type = 'block';
1511  }
1512  $para->[1]{'~type'} = $list_type;
1513  push @$curr_open, $para;
1514   # yes, we reuse the paragraph as a stack item
1515
1516  my $content = join ' ', splice @$para, 2;
1517  $para->[1]{'~orig_content'} = $content;
1518  my $overness;
1519  if($content =~ m/^\s*$/s) {
1520    $para->[1]{'indent'} = 4;
1521  } elsif($content =~ m/^\s*((?:\d*\.)?\d+)\s*$/s) {
1522    no integer;
1523    $para->[1]{'indent'} = $1;
1524    if($1 == 0) {
1525      $self->whine(
1526        $para->[1]{'start_line'},
1527        "Can't have a 0 in =over $content"
1528      );
1529      $para->[1]{'indent'} = 4;
1530    }
1531  } else {
1532    $self->whine(
1533      $para->[1]{'start_line'},
1534      "=over should be: '=over' or '=over positive_number'"
1535    );
1536    $para->[1]{'indent'} = 4;
1537  }
1538  DEBUG > 1 and print STDERR "=over found of type $list_type\n";
1539
1540  $self->{'content_seen'} ||= 1 unless $self->{'~tried_gen_errata'};
1541  $self->_handle_element_start((my $scratch = 'over-' . $list_type), $para->[1]);
1542
1543  return;
1544}
1545
1546sub _ponder_back {
1547  my ($self,$para,$curr_open,$paras) = @_;
1548  # TODO: fire off </item-number> or </item-bullet> or </item-text> ??
1549
1550  my $content = join ' ', splice @$para, 2;
1551  if($content =~ m/\S/) {
1552    $self->whine(
1553      $para->[1]{'start_line'},
1554      "=back doesn't take any parameters, but you said =back $content"
1555    );
1556  }
1557
1558  if(@$curr_open and $curr_open->[-1][0] eq '=over') {
1559    DEBUG > 1 and print STDERR "=back happily closes matching =over\n";
1560    # Expected case: we're closing the most recently opened thing
1561    #my $over = pop @$curr_open;
1562    $self->{'content_seen'} ||= 1 unless $self->{'~tried_gen_errata'};
1563    $self->_handle_element_end( my $scratch =
1564      'over-' . ( (pop @$curr_open)->[1]{'~type'} ), $para->[1]
1565    );
1566  } else {
1567    DEBUG > 1 and print STDERR "=back found without a matching =over.  Stack: (",
1568        join(', ', map $_->[0], @$curr_open), ").\n";
1569    $self->whine(
1570      $para->[1]{'start_line'},
1571      '=back without =over'
1572    );
1573    return 1; # and ignore it
1574  }
1575}
1576
1577sub _ponder_item {
1578  my ($self,$para,$curr_open,$paras) = @_;
1579  my $over;
1580  unless(@$curr_open and
1581         $over = (grep { $_->[0] eq '=over' } @$curr_open)[-1]) {
1582    $self->whine(
1583      $para->[1]{'start_line'},
1584      "'=item' outside of any '=over'"
1585    );
1586    unshift @$paras,
1587      ['=over', {'start_line' => $para->[1]{'start_line'}}, ''],
1588      $para
1589    ;
1590    return 1;
1591  }
1592
1593
1594  my $over_type = $over->[1]{'~type'};
1595
1596  if(!$over_type) {
1597    # Shouldn't happen1
1598    die "Typeless over in stack, starting at line "
1599     . $over->[1]{'start_line'};
1600
1601  } elsif($over_type eq 'block') {
1602    unless($curr_open->[-1][1]{'~bitched_about'}) {
1603      $curr_open->[-1][1]{'~bitched_about'} = 1;
1604      $self->whine(
1605        $curr_open->[-1][1]{'start_line'},
1606        "You can't have =items (as at line "
1607        . $para->[1]{'start_line'}
1608        . ") unless the first thing after the =over is an =item"
1609      );
1610    }
1611    # Just turn it into a paragraph and reconsider it
1612    $para->[0] = '~Para';
1613    unshift @$paras, $para;
1614    return 1;
1615
1616  } elsif($over_type eq 'text') {
1617    my $item_type = $self->_get_item_type($para);
1618      # That kills the content of the item if it's a number or bullet.
1619    DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1620
1621    if($item_type eq 'text') {
1622      # Nothing special needs doing for 'text'
1623    } elsif($item_type eq 'number' or $item_type eq 'bullet') {
1624      $self->whine(
1625          $para->[1]{'start_line'},
1626          "Expected text after =item, not a $item_type"
1627      );
1628      # Undo our clobbering:
1629      push @$para, $para->[1]{'~orig_content'};
1630      delete $para->[1]{'number'};
1631       # Only a PROPER item-number element is allowed
1632       #  to have a number attribute.
1633    } else {
1634      die "Unhandled item type $item_type"; # should never happen
1635    }
1636
1637    # =item-text thingies don't need any assimilation, it seems.
1638
1639  } elsif($over_type eq 'number') {
1640    my $item_type = $self->_get_item_type($para);
1641      # That kills the content of the item if it's a number or bullet.
1642    DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1643
1644    my $expected_value = ++ $curr_open->[-1][1]{'~counter'};
1645
1646    if($item_type eq 'bullet') {
1647      # Hm, it's not numeric.  Correct for this.
1648      $para->[1]{'number'} = $expected_value;
1649      $self->whine(
1650        $para->[1]{'start_line'},
1651        "Expected '=item $expected_value'"
1652      );
1653      push @$para, $para->[1]{'~orig_content'};
1654        # restore the bullet, blocking the assimilation of next para
1655
1656    } elsif($item_type eq 'text') {
1657      # Hm, it's not numeric.  Correct for this.
1658      $para->[1]{'number'} = $expected_value;
1659      $self->whine(
1660        $para->[1]{'start_line'},
1661        "Expected '=item $expected_value'"
1662      );
1663      # Text content will still be there and will block next ~Para
1664
1665    } elsif($item_type ne 'number') {
1666      die "Unknown item type $item_type"; # should never happen
1667
1668    } elsif($expected_value == $para->[1]{'number'}) {
1669      DEBUG > 1 and print STDERR " Numeric item has the expected value of $expected_value\n";
1670
1671    } else {
1672      DEBUG > 1 and print STDERR " Numeric item has ", $para->[1]{'number'},
1673       " instead of the expected value of $expected_value\n";
1674      $self->whine(
1675        $para->[1]{'start_line'},
1676        "You have '=item " . $para->[1]{'number'} .
1677        "' instead of the expected '=item $expected_value'"
1678      );
1679      $para->[1]{'number'} = $expected_value;  # correcting!!
1680    }
1681
1682    if(@$para == 2) {
1683      # For the cases where we /didn't/ push to @$para
1684      if($paras->[0][0] eq '~Para') {
1685        DEBUG and print STDERR "Assimilating following ~Para content into $over_type item\n";
1686        push @$para, splice @{shift @$paras},2;
1687      } else {
1688        DEBUG and print STDERR "Can't assimilate following ", $paras->[0][0], "\n";
1689        push @$para, '';  # Just so it's not contentless
1690      }
1691    }
1692
1693
1694  } elsif($over_type eq 'bullet') {
1695    my $item_type = $self->_get_item_type($para);
1696      # That kills the content of the item if it's a number or bullet.
1697    DEBUG and print STDERR " Item is of type ", $para->[0], " under $over_type\n";
1698
1699    if($item_type eq 'bullet') {
1700      # as expected!
1701
1702      if( $para->[1]{'~_freaky_para_hack'} ) {
1703        DEBUG and print STDERR "Accomodating '=item * Foo' tolerance hack.\n";
1704        push @$para, $para->[1]{'~_freaky_para_hack'};
1705      }
1706
1707    } elsif($item_type eq 'number') {
1708      $self->whine(
1709        $para->[1]{'start_line'},
1710        "Expected '=item *'"
1711      );
1712      push @$para, $para->[1]{'~orig_content'};
1713       # and block assimilation of the next paragraph
1714      delete $para->[1]{'number'};
1715       # Only a PROPER item-number element is allowed
1716       #  to have a number attribute.
1717    } elsif($item_type eq 'text') {
1718      $self->whine(
1719        $para->[1]{'start_line'},
1720        "Expected '=item *'"
1721      );
1722       # But doesn't need processing.  But it'll block assimilation
1723       #  of the next para.
1724    } else {
1725      die "Unhandled item type $item_type"; # should never happen
1726    }
1727
1728    if(@$para == 2) {
1729      # For the cases where we /didn't/ push to @$para
1730      if($paras->[0][0] eq '~Para') {
1731        DEBUG and print STDERR "Assimilating following ~Para content into $over_type item\n";
1732        push @$para, splice @{shift @$paras},2;
1733      } else {
1734        DEBUG and print STDERR "Can't assimilate following ", $paras->[0][0], "\n";
1735        push @$para, '';  # Just so it's not contentless
1736      }
1737    }
1738
1739  } else {
1740    die "Unhandled =over type \"$over_type\"?";
1741    # Shouldn't happen!
1742  }
1743  $para->[0] .= '-' . $over_type;
1744
1745  return;
1746}
1747
1748sub _ponder_Plain {
1749  my ($self,$para) = @_;
1750  DEBUG and print STDERR " giving plain treatment...\n";
1751  unless( @$para == 2 or ( @$para == 3 and $para->[2] eq '' )
1752    or $para->[1]{'~cooked'}
1753  ) {
1754    push @$para,
1755    @{$self->_make_treelet(
1756      join("\n", splice(@$para, 2)),
1757      $para->[1]{'start_line'}
1758    )};
1759  }
1760  # Empty paragraphs don't need a treelet for any reason I can see.
1761  # And precooked paragraphs already have a treelet.
1762  return;
1763}
1764
1765sub _ponder_Verbatim {
1766  my ($self,$para) = @_;
1767  DEBUG and print STDERR " giving verbatim treatment...\n";
1768
1769  $para->[1]{'xml:space'} = 'preserve';
1770
1771  unless ($self->{'_output_is_for_JustPod'}) {
1772    # Fix illegal settings for expand_verbatim_tabs()
1773    # This is because this module doesn't do input error checking, but khw
1774    # doesn't want to add yet another instance of that.
1775    $self->expand_verbatim_tabs(8)
1776                            if ! defined $self->expand_verbatim_tabs()
1777                            ||   $self->expand_verbatim_tabs() =~ /\D/;
1778
1779    my $indent = $self->strip_verbatim_indent;
1780    if ($indent && ref $indent eq 'CODE') {
1781        my @shifted = (shift @{$para}, shift @{$para});
1782        $indent = $indent->($para);
1783        unshift @{$para}, @shifted;
1784    }
1785
1786    for(my $i = 2; $i < @$para; $i++) {
1787      foreach my $line ($para->[$i]) { # just for aliasing
1788        # Strip indentation.
1789        $line =~ s/^\Q$indent// if $indent;
1790        next unless $self->expand_verbatim_tabs;
1791
1792            # This is commented out because of github issue #85, and the
1793            # current maintainers don't know why it was there in the first
1794            # place.
1795            #&& !($self->{accept_codes} && $self->{accept_codes}{VerbatimFormatted});
1796        while( $line =~
1797          # Sort of adapted from Text::Tabs.
1798          s/^([^\t]*)(\t+)/$1.(" " x ((length($2)
1799                                       * $self->expand_verbatim_tabs)
1800                                       -(length($1)&7)))/e
1801        ) {}
1802
1803        # TODO: whinge about (or otherwise treat) unindented or overlong lines
1804
1805      }
1806    }
1807  }
1808
1809  # Now the VerbatimFormatted hoodoo...
1810  if( $self->{'accept_codes'} and
1811      $self->{'accept_codes'}{'VerbatimFormatted'}
1812  ) {
1813    while(@$para > 3 and $para->[-1] !~ m/\S/) { pop @$para }
1814     # Kill any number of terminal newlines
1815    $self->_verbatim_format($para);
1816  } elsif ($self->{'codes_in_verbatim'}) {
1817    push @$para,
1818    @{$self->_make_treelet(
1819      join("\n", splice(@$para, 2)),
1820      $para->[1]{'start_line'}, $para->[1]{'xml:space'}
1821    )};
1822    $para->[-1] =~ s/\n+$//s; # Kill any number of terminal newlines
1823  } else {
1824    push @$para, join "\n", splice(@$para, 2) if @$para > 3;
1825    $para->[-1] =~ s/\n+$//s; # Kill any number of terminal newlines
1826  }
1827  return;
1828}
1829
1830sub _ponder_Data {
1831  my ($self,$para) = @_;
1832  DEBUG and print STDERR " giving data treatment...\n";
1833  $para->[1]{'xml:space'} = 'preserve';
1834  push @$para, join "\n", splice(@$para, 2) if @$para > 3;
1835  return;
1836}
1837
1838
1839
1840
1841###########################################################################
1842
1843sub _traverse_treelet_bit {  # for use only by the routine above
1844  my($self, $name) = splice @_,0,2;
1845
1846  my $scratch;
1847  $self->_handle_element_start(($scratch=$name), shift @_);
1848
1849  while (@_) {
1850    my $x = shift;
1851    if (ref($x)) {
1852      &_traverse_treelet_bit($self, @$x);
1853    } else {
1854      $x .= shift while @_ && !ref($_[0]);
1855      $self->_handle_text($x);
1856    }
1857  }
1858
1859  $self->_handle_element_end($scratch=$name);
1860  return;
1861}
1862
1863#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
1864
1865sub _closers_for_all_curr_open {
1866  my $self = $_[0];
1867  my @closers;
1868  foreach my $still_open (@{  $self->{'curr_open'} || return  }) {
1869    my @copy = @$still_open;
1870    $copy[1] = {%{ $copy[1] }};
1871    #$copy[1]{'start_line'} = -1;
1872    if($copy[0] eq '=for') {
1873      $copy[0] = '=end';
1874    } elsif($copy[0] eq '=over') {
1875      $self->whine(
1876        $still_open->[1]{start_line} ,
1877        "=over without closing =back"
1878      );
1879
1880      $copy[0] = '=back';
1881    } else {
1882      die "I don't know how to auto-close an open $copy[0] region";
1883    }
1884
1885    unless( @copy > 2 ) {
1886      push @copy, $copy[1]{'target'};
1887      $copy[-1] = '' unless defined $copy[-1];
1888       # since =over's don't have targets
1889    }
1890
1891    $copy[1]{'fake-closer'} = 1;
1892
1893    DEBUG and print STDERR "Queuing up fake-o event: ", pretty(\@copy), "\n";
1894    unshift @closers, \@copy;
1895  }
1896  return @closers;
1897}
1898
1899#--------------------------------------------------------------------------
1900
1901sub _verbatim_format {
1902  my($it, $p) = @_;
1903
1904  my $formatting;
1905
1906  for(my $i = 2; $i < @$p; $i++) { # work backwards over the lines
1907    DEBUG and print STDERR "_verbatim_format appends a newline to $i: $p->[$i]\n";
1908    $p->[$i] .= "\n";
1909     # Unlike with simple Verbatim blocks, we don't end up just doing
1910     # a join("\n", ...) on the contents, so we have to append a
1911     # newline to every line, and then nix the last one later.
1912  }
1913
1914  if( DEBUG > 4 ) {
1915    print STDERR "<<\n";
1916    for(my $i = $#$p; $i >= 2; $i--) { # work backwards over the lines
1917      print STDERR "_verbatim_format $i: $p->[$i]";
1918    }
1919    print STDERR ">>\n";
1920  }
1921
1922  for(my $i = $#$p; $i > 2; $i--) {
1923    # work backwards over the lines, except the first (#2)
1924
1925    #next unless $p->[$i]   =~ m{^#:([ \^\/\%]*)\n?$}s
1926    #        and $p->[$i-1] !~ m{^#:[ \^\/\%]*\n?$}s;
1927     # look at a formatty line preceding a nonformatty one
1928    DEBUG > 5 and print STDERR "Scrutinizing line $i: $$p[$i]\n";
1929    if($p->[$i]   =~ m{^#:([ \^\/\%]*)\n?$}s) {
1930      DEBUG > 5 and print STDERR "  It's a formatty line.  ",
1931       "Peeking at previous line ", $i-1, ": $$p[$i-1]: \n";
1932
1933      if( $p->[$i-1] =~ m{^#:[ \^\/\%]*\n?$}s ) {
1934        DEBUG > 5 and print STDERR "  Previous line is formatty!  Skipping this one.\n";
1935        next;
1936      } else {
1937        DEBUG > 5 and print STDERR "  Previous line is non-formatty!  Yay!\n";
1938      }
1939    } else {
1940      DEBUG > 5 and print STDERR "  It's not a formatty line.  Ignoring\n";
1941      next;
1942    }
1943
1944    # A formatty line has to have #: in the first two columns, and uses
1945    # "^" to mean bold, "/" to mean underline, and "%" to mean bold italic.
1946    # Example:
1947    #   What do you want?  i like pie. [or whatever]
1948    # #:^^^^^^^^^^^^^^^^^              /////////////
1949
1950
1951    DEBUG > 4 and print STDERR "_verbatim_format considers:\n<$p->[$i-1]>\n<$p->[$i]>\n";
1952
1953    $formatting = '  ' . $1;
1954    $formatting =~ s/\s+$//s; # nix trailing whitespace
1955    unless(length $formatting and $p->[$i-1] =~ m/\S/) { # no-op
1956      splice @$p,$i,1; # remove this line
1957      $i--; # don't consider next line
1958      next;
1959    }
1960
1961    if( length($formatting) >= length($p->[$i-1]) ) {
1962      $formatting = substr($formatting, 0, length($p->[$i-1]) - 1) . ' ';
1963    } else {
1964      $formatting .= ' ' x (length($p->[$i-1]) - length($formatting));
1965    }
1966    # Make $formatting and the previous line be exactly the same length,
1967    # with $formatting having a " " as the last character.
1968
1969    DEBUG > 4 and print STDERR "Formatting <$formatting>    on <", $p->[$i-1], ">\n";
1970
1971
1972    my @new_line;
1973    while( $formatting =~ m{\G(( +)|(\^+)|(\/+)|(\%+))}g ) {
1974      #print STDERR "Format matches $1\n";
1975
1976      if($2) {
1977        #print STDERR "SKIPPING <$2>\n";
1978        push @new_line,
1979          substr($p->[$i-1], pos($formatting)-length($1), length($1));
1980      } else {
1981        #print STDERR "SNARING $+\n";
1982        push @new_line, [
1983          (
1984            $3 ? 'VerbatimB'  :
1985            $4 ? 'VerbatimI'  :
1986            $5 ? 'VerbatimBI' : die("Should never get called")
1987          ), {},
1988          substr($p->[$i-1], pos($formatting)-length($1), length($1))
1989        ];
1990        #print STDERR "Formatting <$new_line[-1][-1]> as $new_line[-1][0]\n";
1991      }
1992    }
1993    my @nixed =
1994      splice @$p, $i-1, 2, @new_line; # replace myself and the next line
1995    DEBUG > 10 and print STDERR "Nixed count: ", scalar(@nixed), "\n";
1996
1997    DEBUG > 6 and print STDERR "New version of the above line is these tokens (",
1998      scalar(@new_line), "):",
1999      map( ref($_)?"<@$_> ":"<$_>", @new_line ), "\n";
2000    $i--; # So the next line we scrutinize is the line before the one
2001          #  that we just went and formatted
2002  }
2003
2004  $p->[0] = 'VerbatimFormatted';
2005
2006  # Collapse adjacent text nodes, just for kicks.
2007  for( my $i = 2; $i > $#$p; $i++ ) { # work forwards over the tokens except for the last
2008    if( !ref($p->[$i]) and !ref($p->[$i + 1]) ) {
2009      DEBUG > 5 and print STDERR "_verbatim_format merges {$p->[$i]} and {$p->[$i+1]}\n";
2010      $p->[$i] .= splice @$p, $i+1, 1; # merge
2011      --$i;  # and back up
2012    }
2013  }
2014
2015  # Now look for the last text token, and remove the terminal newline
2016  for( my $i = $#$p; $i >= 2; $i-- ) {
2017    # work backwards over the tokens, even the first
2018    if( !ref($p->[$i]) ) {
2019      if($p->[$i] =~ s/\n$//s) {
2020        DEBUG > 5 and print STDERR "_verbatim_format killed the terminal newline on #$i: {$p->[$i]}, after {$p->[$i-1]}\n";
2021      } else {
2022        DEBUG > 5 and print STDERR
2023         "No terminal newline on #$i: {$p->[$i]}, after {$p->[$i-1]} !?\n";
2024      }
2025      last; # we only want the next one
2026    }
2027  }
2028
2029  return;
2030}
2031
2032
2033#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
2034
2035
2036sub _treelet_from_formatting_codes {
2037  # Given a paragraph, returns a treelet.  Full of scary tokenizing code.
2038  #  Like [ '~Top', {'start_line' => $start_line},
2039  #            "I like ",
2040  #            [ 'B', {}, "pie" ],
2041  #            "!"
2042  #       ]
2043  # This illustrates the general format of a treelet.  It is an array:
2044  #     [0]       is a scalar indicating its type.  In the example above, the
2045  #               types are '~Top' and 'B'
2046  #     [1]       is a hash of various flags about it, possibly empty
2047  #     [2] - [N] are an ordered list of the subcomponents of the treelet.
2048  #               Scalars are literal text, refs are sub-treelets, to
2049  #               arbitrary levels.  Stringifying a treelet will recursively
2050  #               stringify the sub-treelets, concatentating everything
2051  #               together to form the exact text of the treelet.
2052
2053  my($self, $para, $start_line, $preserve_space) = @_;
2054
2055  my $treelet = ['~Top', {'start_line' => $start_line},];
2056
2057  unless ($preserve_space || $self->{'preserve_whitespace'}) {
2058    $para =~ s/\s+/ /g; # collapse and trim all whitespace first.
2059    $para =~ s/ $//;
2060    $para =~ s/^ //;
2061  }
2062
2063  # Only apparent problem the above code is that N<<  >> turns into
2064  # N<< >>.  But then, word wrapping does that too!  So don't do that!
2065
2066
2067  # As a Start-code is encountered, the number of opening bracket '<'
2068  # characters minus 1 is pushed onto @stack (so 0 means a single bracket,
2069  # etc).  When closing brackets are found in the text, at least this number
2070  # (plus the 1) will be required to mean the Start-code is terminated.  When
2071  # those are found, @stack is popped.
2072  my @stack;
2073
2074  my @lineage = ($treelet);
2075  my $raw = ''; # raw content of L<> fcode before splitting/processing
2076    # XXX 'raw' is not 100% accurate: all surrounding whitespace is condensed
2077    # into just 1 ' '. Is this the regex's doing or 'raw's?  Answer is it's
2078    # the 'collapse and trim all whitespace first' lines just above.
2079  my $inL = 0;
2080
2081  DEBUG > 4 and print STDERR "Paragraph:\n$para\n\n";
2082
2083  # Here begins our frightening tokenizer RE.  The following regex matches
2084  # text in four main parts:
2085  #
2086  #  * Start-codes.  The first alternative matches C< or C<<, the latter
2087  #    followed by some whitespace.  $1 will hold the entire start code
2088  #    (including any space following a multiple-angle-bracket delimiter),
2089  #    and $2 will hold only the additional brackets past the first in a
2090  #    multiple-bracket delimiter.  length($2) + 1 will be the number of
2091  #    closing brackets we have to find.
2092  #
2093  #  * Closing brackets.  Match some amount of whitespace followed by
2094  #    multiple close brackets.  The logic to see if this closes anything
2095  #    is down below.  Note that in order to parse C<<  >> correctly, we
2096  #    have to use look-behind (?<=\s\s), since the match of the starting
2097  #    code will have consumed the whitespace.
2098  #
2099  #  * A single closing bracket, to close a simple code like C<>.
2100  #
2101  #  * Something that isn't a start or end code.  We have to be careful
2102  #    about accepting whitespace, since perlpodspec says that any whitespace
2103  #    before a multiple-bracket closing delimiter should be ignored.
2104  #
2105  while($para =~
2106    m/\G
2107      (?:
2108        # Match starting codes, including the whitespace following a
2109        # multiple-delimiter start code.  $1 gets the whole start code and
2110        # $2 gets all but one of the <s in the multiple-bracket case.
2111        ([A-Z]<(?:(<+)\s+)?)
2112        |
2113        # Match multiple-bracket end codes.  $3 gets the whitespace that
2114        # should be discarded before an end bracket but kept in other cases
2115        # and $4 gets the end brackets themselves.  ($3 can be empty if the
2116        # construct is empty, like C<<  >>, and all the white-space has been
2117        # gobbled up already, considered to be space after the opening
2118        # bracket.  In this case we use look-behind to verify that there are
2119        # at least 2 spaces in a row before the ">".)
2120        (\s+|(?<=\s\s))(>{2,})
2121        |
2122        (\s?>)          # $5: simple end-codes
2123        |
2124        (               # $6: stuff containing no start-codes or end-codes
2125          (?:
2126            [^A-Z\s>]
2127            |
2128            (?:
2129              [A-Z](?!<)
2130            )
2131            |
2132            # whitespace is ok, but we don't want to eat the whitespace before
2133            # a multiple-bracket end code.
2134            # NOTE: we may still have problems with e.g. S<<    >>
2135            (?:
2136              \s(?!\s*>{2,})
2137            )
2138          )+
2139        )
2140      )
2141    /xgo
2142  ) {
2143    DEBUG > 4 and print STDERR "\nParagraphic tokenstack = (@stack)\n";
2144    if(defined $1) {
2145      my $bracket_count;    # How many '<<<' in a row this has.  Needed for
2146                            # Pod::Simple::JustPod
2147      if(defined $2) {
2148        DEBUG > 3 and print STDERR "Found complex start-text code \"$1\"\n";
2149        $bracket_count = length($2) + 1;
2150        push @stack, $bracket_count; # length of the necessary complex
2151                                     # end-code string
2152      } else {
2153        DEBUG > 3 and print STDERR "Found simple start-text code \"$1\"\n";
2154        push @stack, 0;  # signal that we're looking for simple
2155        $bracket_count = 1;
2156      }
2157      my $code = substr($1,0,1);
2158      if ('L' eq $code) {
2159        if ($inL) {
2160            $raw .= $1;
2161            $self->scream( $start_line,
2162                           'Nested L<> are illegal.  Pretending inner one is '
2163                         . 'X<...> so can continue looking for other errors.');
2164            $code = "X";
2165        }
2166        else {
2167            $raw = ""; # reset raw content accumulator
2168            $inL = @stack;
2169        }
2170      } else {
2171        $raw .= $1 if $inL;
2172      }
2173      push @lineage, [ $code, {}, ];  # new node object
2174
2175      # Tell Pod::Simple::JustPod how many brackets there were, but to save
2176      # space, not in the most usual case of there was just 1.  It can be
2177      # inferred by the absence of this element.  Similarly, if there is more
2178      # than one bracket, extract the white space between the final bracket
2179      # and the real beginning of the interior.  Save that if it isn't just a
2180      # single space
2181      if ($self->{'_output_is_for_JustPod'} && $bracket_count > 1) {
2182        $lineage[-1][1]{'~bracket_count'} = $bracket_count;
2183        my $lspacer = substr($1, 1 + $bracket_count);
2184        $lineage[-1][1]{'~lspacer'} = $lspacer if $lspacer ne " ";
2185      }
2186      push @{ $lineage[-2] }, $lineage[-1];
2187    } elsif(defined $4) {
2188      DEBUG > 3 and print STDERR "Found apparent complex end-text code \"$3$4\"\n";
2189      # This is where it gets messy...
2190      if(! @stack) {
2191        # We saw " >>>>" but needed nothing.  This is ALL just stuff then.
2192        DEBUG > 4 and print STDERR " But it's really just stuff.\n";
2193        push @{ $lineage[-1] }, $3, $4;
2194        next;
2195      } elsif(!$stack[-1]) {
2196        # We saw " >>>>" but needed only ">".  Back pos up.
2197        DEBUG > 4 and print STDERR " And that's more than we needed to close simple.\n";
2198        push @{ $lineage[-1] }, $3; # That was a for-real space, too.
2199        pos($para) = pos($para) - length($4) + 1;
2200      } elsif($stack[-1] == length($4)) {
2201        # We found " >>>>", and it was exactly what we needed.  Commonest case.
2202        DEBUG > 4 and print STDERR " And that's exactly what we needed to close complex.\n";
2203      } elsif($stack[-1] < length($4)) {
2204        # We saw " >>>>" but needed only " >>".  Back pos up.
2205        DEBUG > 4 and print STDERR " And that's more than we needed to close complex.\n";
2206        pos($para) = pos($para) - length($4) + $stack[-1];
2207      } else {
2208        # We saw " >>>>" but needed " >>>>>>".  So this is all just stuff!
2209        DEBUG > 4 and print STDERR " But it's really just stuff, because we needed more.\n";
2210        push @{ $lineage[-1] }, $3, $4;
2211        next;
2212      }
2213      #print STDERR "\nHOOBOY ", scalar(@{$lineage[-1]}), "!!!\n";
2214
2215      if ($3 ne " " && $self->{'_output_is_for_JustPod'}) {
2216        if ($3 ne "") {
2217          $lineage[-1][1]{'~rspacer'} = $3;
2218        }
2219        elsif ($lineage[-1][1]{'~lspacer'} eq "  ") {
2220
2221          # Here we had something like C<<  >> which was a false positive
2222          delete $lineage[-1][1]{'~lspacer'};
2223        }
2224        else {
2225          $lineage[-1][1]{'~rspacer'}
2226                                = substr($lineage[-1][1]{'~lspacer'}, -1, 1);
2227          chop $lineage[-1][1]{'~lspacer'};
2228        }
2229      }
2230
2231      push @{ $lineage[-1] }, '' if 2 == @{ $lineage[-1] };
2232      # Keep the element from being childless
2233
2234      if ($inL == @stack) {
2235        $lineage[-1][1]{'raw'} = $raw;
2236        $inL = 0;
2237      }
2238
2239      pop @stack;
2240      pop @lineage;
2241
2242      $raw .= $3.$4 if $inL;
2243
2244    } elsif(defined $5) {
2245      DEBUG > 3 and print STDERR "Found apparent simple end-text code \"$5\"\n";
2246
2247      if(@stack and ! $stack[-1]) {
2248        # We're indeed expecting a simple end-code
2249        DEBUG > 4 and print STDERR " It's indeed an end-code.\n";
2250
2251        if(length($5) == 2) { # There was a space there: " >"
2252          push @{ $lineage[-1] }, ' ';
2253        } elsif( 2 == @{ $lineage[-1] } ) { # Closing a childless element
2254          push @{ $lineage[-1] }, ''; # keep it from being really childless
2255        }
2256
2257        if ($inL == @stack) {
2258          $lineage[-1][1]{'raw'} = $raw;
2259          $inL = 0;
2260        }
2261
2262        pop @stack;
2263        pop @lineage;
2264      } else {
2265        DEBUG > 4 and print STDERR " It's just stuff.\n";
2266        push @{ $lineage[-1] }, $5;
2267      }
2268
2269      $raw .= $5 if $inL;
2270
2271    } elsif(defined $6) {
2272      DEBUG > 3 and print STDERR "Found stuff \"$6\"\n";
2273      push @{ $lineage[-1] }, $6;
2274      $raw .= $6 if $inL;
2275        # XXX does not capture multiplace whitespaces -- 'raw' ends up with
2276        #     at most 1 leading/trailing whitespace, why not all of it?
2277        #     Answer, because we deliberately trimmed it above
2278
2279    } else {
2280      # should never ever ever ever happen
2281      DEBUG and print STDERR "AYYAYAAAAA at line ", __LINE__, "\n";
2282      die "SPORK 512512!";
2283    }
2284  }
2285
2286  if(@stack) { # Uhoh, some sequences weren't closed.
2287    my $x= "...";
2288    while(@stack) {
2289      push @{ $lineage[-1] }, '' if 2 == @{ $lineage[-1] };
2290      # Hmmmmm!
2291
2292      my $code         = (pop @lineage)->[0];
2293      my $ender_length =  pop @stack;
2294      if($ender_length) {
2295        --$ender_length;
2296        $x = $code . ("<" x $ender_length) . " $x " . (">" x $ender_length);
2297      } else {
2298        $x = $code . "<$x>";
2299      }
2300    }
2301    DEBUG > 1 and print STDERR "Unterminated $x sequence\n";
2302    $self->whine($start_line,
2303      "Unterminated $x sequence",
2304    );
2305  }
2306
2307  return $treelet;
2308}
2309
2310#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
2311
2312sub text_content_of_treelet {  # method: $parser->text_content_of_treelet($lol)
2313  return stringify_lol($_[1]);
2314}
2315
2316sub stringify_lol {  # function: stringify_lol($lol)
2317  my $string_form = '';
2318  _stringify_lol( $_[0] => \$string_form );
2319  return $string_form;
2320}
2321
2322sub _stringify_lol {  # the real recursor
2323  my($lol, $to) = @_;
2324  for(my $i = 2; $i < @$lol; ++$i) {
2325    if( ref($lol->[$i] || '') and UNIVERSAL::isa($lol->[$i], 'ARRAY') ) {
2326      _stringify_lol( $lol->[$i], $to);  # recurse!
2327    } else {
2328      $$to .= $lol->[$i];
2329    }
2330  }
2331  return;
2332}
2333
2334#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
2335
2336sub _dump_curr_open { # return a string representation of the stack
2337  my $curr_open = $_[0]{'curr_open'};
2338
2339  return '[empty]' unless @$curr_open;
2340  return join '; ',
2341    map {;
2342           ($_->[0] eq '=for')
2343             ? ( ($_->[1]{'~really'} || '=over')
2344               . ' ' . $_->[1]{'target'})
2345             : $_->[0]
2346        }
2347    @$curr_open
2348  ;
2349}
2350
2351###########################################################################
2352my %pretty_form = (
2353  "\a" => '\a', # ding!
2354  "\b" => '\b', # BS
2355  "\e" => '\e', # ESC
2356  "\f" => '\f', # FF
2357  "\t" => '\t', # tab
2358  "\cm" => '\cm',
2359  "\cj" => '\cj',
2360  "\n" => '\n', # probably overrides one of either \cm or \cj
2361  '"' => '\"',
2362  '\\' => '\\\\',
2363  '$' => '\\$',
2364  '@' => '\\@',
2365  '%' => '\\%',
2366  '#' => '\\#',
2367);
2368
2369sub pretty { # adopted from Class::Classless
2370  # Not the most brilliant routine, but passable.
2371  # Don't give it a cyclic data structure!
2372  my @stuff = @_; # copy
2373  my $x;
2374  my $out =
2375    # join ",\n" .
2376    join ", ",
2377    map {;
2378    if(!defined($_)) {
2379      "undef";
2380    } elsif(ref($_) eq 'ARRAY' or ref($_) eq 'Pod::Simple::LinkSection') {
2381      $x = "[ " . pretty(@$_) . " ]" ;
2382      $x;
2383    } elsif(ref($_) eq 'SCALAR') {
2384      $x = "\\" . pretty($$_) ;
2385      $x;
2386    } elsif(ref($_) eq 'HASH') {
2387      my $hr = $_;
2388      $x = "{" . join(", ",
2389        map(pretty($_) . '=>' . pretty($hr->{$_}),
2390            sort keys %$hr ) ) . "}" ;
2391      $x;
2392    } elsif(!length($_)) { q{''} # empty string
2393    } elsif(
2394      $_ eq '0' # very common case
2395      or(
2396         m/^-?(?:[123456789]\d*|0)(?:\.\d+)?$/s
2397         and $_ ne '-0' # the strange case that RE lets thru
2398      )
2399    ) { $_;
2400    } else {
2401        # Yes, explicitly name every character desired. There are shorcuts one
2402        # could make, but I (Karl Williamson) was afraid that some Perl
2403        # releases would have bugs in some of them. For example [A-Z] works
2404        # even on EBCDIC platforms to match exactly the 26 uppercase English
2405        # letters, but I don't know if it has always worked without bugs. It
2406        # seemed safest just to list the characters.
2407        # s<([^\x20\x21\x23\x27-\x3F\x41-\x5B\x5D-\x7E])>
2408        s<([^ !"#'()*+,\-./0123456789:;\<=\>?ABCDEFGHIJKLMNOPQRSTUVWXYZ\[\]^_`abcdefghijklmnopqrstuvwxyz{|}~])>
2409         <$pretty_form{$1} || '\\x{'.sprintf("%x", ord($1)).'}'>eg;
2410         #<$pretty_form{$1} || '\\x'.(unpack("H2",$1))>eg;
2411      qq{"$_"};
2412    }
2413  } @stuff;
2414  # $out =~ s/\n */ /g if length($out) < 75;
2415  return $out;
2416}
2417
2418#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
2419
2420# A rather unsubtle method of blowing away all the state information
2421# from a parser object so it can be reused. Provided as a utility for
2422# backward compatibility in Pod::Man, etc. but not recommended for
2423# general use.
2424
2425sub reinit {
2426  my $self = shift;
2427  foreach (qw(source_dead source_filename doc_has_started
2428start_of_pod_block content_seen last_was_blank paras curr_open
2429line_count pod_para_count in_pod ~tried_gen_errata all_errata errata errors_seen
2430Title)) {
2431
2432    delete $self->{$_};
2433  }
2434}
2435
2436#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
24371;
2438
2439