Commit | Line | Data |
---|---|---|
94bdecf9 | 1 | #!/usr/bin/perl -w |
00d68f0d KW |
2 | |
3 | use Text::Tabs; | |
a9a7c81e | 4 | # |
6294c161 DM |
5 | # Unconditionally regenerate: |
6 | # | |
7 | # pod/perlintern.pod | |
8 | # pod/perlapi.pod | |
9 | # | |
10 | # from information stored in | |
11 | # | |
12 | # embed.fnc | |
9b983db7 | 13 | # plus all the core .c, .h, and .pod files listed in MANIFEST |
6294c161 DM |
14 | # |
15 | # Has an optional arg, which is the directory to chdir to before reading | |
9b983db7 | 16 | # MANIFEST and the files |
6294c161 | 17 | # |
52a9d53b | 18 | # This script is invoked as part of 'make all' |
151c3fe5 | 19 | # |
9b983db7 KW |
20 | # The generated pod consists of sections of related elements, functions, |
21 | # macros, and variables. The keys of %valid_sections give the current legal | |
22 | # ones. Just add a new key to add a section. | |
23 | # | |
24 | # Throughout the files read by this script are lines like | |
25 | # | |
26 | # =for apidoc_section Section Name | |
27 | # | |
28 | # "Section Name" (after having been stripped of leading space) must be one of | |
29 | # the legal section names, or an error is thrown. All API elements defined | |
30 | # between this line and the next 'apidoc_section' line will go into the | |
31 | # section "Section Name", sorted by dictionary order within it. perlintern | |
32 | # and perlapi are parallel documents, each potentially with a section "Section | |
33 | # Name". Each element is marked as to which document it goes into. If there | |
34 | # are none for a particular section in perlapi, that section is | |
35 | # omitted. | |
36 | # | |
37 | # Also, in .[ch] files, there may be | |
38 | # | |
39 | # =head1 Section Name | |
40 | # | |
41 | # lines in comments. These are also used by this program to switch to section | |
42 | # "Section Name". The difference is that if there are any lines after the | |
43 | # =head1, inside the same comment, and before any =for apidoc-ish lines, they | |
44 | # are used as a heading for section "Section Name" (in both perlintern and | |
45 | # perlapi). This includes any =head[2-5]. If more than one '=head1 Section | |
46 | # Name' line has content, they appear in the generated pod in an undefined | |
47 | # order. | |
48 | # | |
49 | # The next =head1, =for apidoc_section, or file end terminates what goes into | |
50 | # the current section | |
51 | # | |
52 | # The %valid_sections hash below also can have header content, which will | |
53 | # appear before any =head1 content. The hash can also have footer content | |
54 | # content, which will appear at the end of the section, after all the | |
55 | # elements. | |
56 | # | |
57 | # The lines that define the actual functions, etc are documented in embed.fnc, | |
58 | # because they have flags which must be kept in sync with that file. | |
1fcde0e9 | 59 | |
56a0c332 | 60 | use strict; |
04adaf08 | 61 | use warnings; |
a64c954a | 62 | |
e4f5ddf4 KW |
63 | # 80 column terminal - 1 for pager adding a column; -7 for nroff |
64 | # indent; | |
65 | my $max_width = 80 - 1 - 7; | |
66 | ||
7882b24a NC |
67 | if (@ARGV) { |
68 | my $workdir = shift; | |
69 | chdir $workdir | |
70 | or die "Couldn't chdir to '$workdir': $!"; | |
71 | } | |
3d7c117d MB |
72 | require './regen/regen_lib.pl'; |
73 | require './regen/embed_lib.pl'; | |
7882b24a | 74 | |
a026e430 | 75 | my %described_elsewhere; |
b87d9527 | 76 | |
94bdecf9 | 77 | # |
346f75ff | 78 | # See database of global and static function prototypes in embed.fnc |
94bdecf9 JH |
79 | # This is used to generate prototype headers under various configurations, |
80 | # export symbols lists for different platforms, and macros to provide an | |
81 | # implicit interpreter context argument. | |
82 | # | |
83 | ||
6a235718 | 84 | my %docs; |
df6bd76f | 85 | my %seen; |
5ce57792 | 86 | my %funcflags; |
5ce57792 | 87 | my %missing; |
00d68f0d KW |
88 | my %missing_macros; |
89 | ||
90 | my $link_text = "Described in"; | |
91 | ||
783469eb KW |
92 | my $description_indent = 4; |
93 | my $usage_indent = 3; # + initial blank yields 4 total | |
94 | ||
3951f999 KW |
95 | my $av_scn = 'AV Handling'; |
96 | my $callback_scn = 'Callback Functions'; | |
97 | my $casting_scn = 'Casting'; | |
98 | my $casing_scn = 'Character case changing'; | |
99 | my $classification_scn = 'Character classification'; | |
100 | my $names_scn = 'Character names'; | |
101 | my $scope_scn = 'Compile-time scope hooks'; | |
102 | my $compiler_scn = 'Compiler and Preprocessor information'; | |
103 | my $directives_scn = 'Compiler directives'; | |
104 | my $concurrency_scn = 'Concurrency'; | |
105 | my $COP_scn = 'COP Hint Hashes'; | |
106 | my $CV_scn = 'CV Handling'; | |
107 | my $custom_scn = 'Custom Operators'; | |
108 | my $dump_scn = 'Display and Dump functions'; | |
109 | my $embedding_scn = 'Embedding and Interpreter Cloning'; | |
110 | my $errno_scn = 'Errno'; | |
111 | my $exceptions_scn = 'Exception Handling (simple) Macros'; | |
112 | my $filesystem_scn = 'Filesystem configuration values'; | |
113 | my $floating_scn = 'Floating point configuration values'; | |
114 | my $formats_scn = 'Formats'; | |
115 | my $genconfig_scn = 'General Configuration'; | |
116 | my $globals_scn = 'Global Variables'; | |
117 | my $GV_scn = 'GV Handling'; | |
118 | my $hook_scn = 'Hook manipulation'; | |
119 | my $HV_scn = 'HV Handling'; | |
120 | my $io_scn = 'Input/Output'; | |
121 | my $integer_scn = 'Integer configuration values'; | |
122 | my $lexer_scn = 'Lexer interface'; | |
123 | my $locales_scn = 'Locales'; | |
124 | my $magic_scn = 'Magic'; | |
125 | my $memory_scn = 'Memory Management'; | |
126 | my $mro_scn = 'MRO'; | |
127 | my $multicall_scn = 'Multicall Functions'; | |
128 | my $numeric_scn = 'Numeric Functions'; | |
129 | my $optree_construction_scn = 'Optree construction'; | |
130 | my $optree_manipulation_scn = 'Optree Manipulation Functions'; | |
131 | my $pack_scn = 'Pack and Unpack'; | |
132 | my $pad_scn = 'Pad Data Structures'; | |
133 | my $password_scn = 'Password and Group access'; | |
134 | my $paths_scn = 'Paths to system commands'; | |
135 | my $intrpvar_scn = 'Per-Interpreter Variables'; | |
136 | my $prototypes_scn = 'Prototype information'; | |
137 | my $regexp_scn = 'REGEXP Functions'; | |
138 | my $signals_scn = 'Signals'; | |
139 | my $site_scn = 'Site configuration'; | |
140 | my $sockets_scn = 'Sockets configuration values'; | |
141 | my $filters_scn = 'Source Filters'; | |
142 | my $stack_scn = 'Stack Manipulation Macros'; | |
143 | my $string_scn = 'String Handling'; | |
144 | my $SV_flags_scn = 'SV Flags'; | |
145 | my $SV_scn = 'SV Handling'; | |
146 | my $time_scn = 'Time'; | |
147 | my $typedefs_scn = 'Typedef names'; | |
148 | my $unicode_scn = 'Unicode Support'; | |
149 | my $utility_scn = 'Utility Functions'; | |
150 | my $versioning_scn = 'Versioning'; | |
151 | my $warning_scn = 'Warning and Dieing'; | |
152 | my $xs_scn = 'XS'; | |
153 | ||
00d68f0d | 154 | my %valid_sections = ( |
3951f999 KW |
155 | $av_scn => {}, |
156 | $callback_scn => {}, | |
157 | $casting_scn => {}, | |
158 | $casing_scn => {}, | |
159 | $classification_scn => {}, | |
160 | $scope_scn => {}, | |
161 | $compiler_scn => {}, | |
162 | $directives_scn => {}, | |
163 | $concurrency_scn => {}, | |
164 | $COP_scn => {}, | |
165 | $CV_scn => {}, | |
166 | $custom_scn => {}, | |
167 | $dump_scn => {}, | |
168 | $embedding_scn => {}, | |
169 | $errno_scn => {}, | |
170 | $exceptions_scn => {}, | |
171 | $filesystem_scn => { | |
6927cd44 KW |
172 | header => <<~'EOT', |
173 | Also see L</List of capability HAS_foo symbols>. | |
174 | EOT | |
175 | }, | |
3951f999 | 176 | $floating_scn => { |
00d68f0d | 177 | header => <<~'EOT', |
6927cd44 KW |
178 | Also L</List of capability HAS_foo symbols> lists capabilities |
179 | that arent in this section. For example C<HAS_ASINH>, for the | |
180 | hyperbolic sine function. | |
181 | EOT | |
182 | }, | |
3951f999 | 183 | $formats_scn => { |
6927cd44 KW |
184 | header => <<~'EOT', |
185 | These are used for formatting the corresponding type For example, | |
186 | instead of saying | |
187 | ||
188 | Perl_newSVpvf(pTHX_ "Create an SV with a %d in it\n", iv); | |
189 | ||
190 | use | |
191 | ||
192 | Perl_newSVpvf(pTHX_ "Create an SV with a " IVdf " in it\n", iv); | |
193 | ||
194 | This keeps you from having to know if, say an IV, needs to be | |
195 | printed as C<%d>, C<%ld>, or something else. | |
00d68f0d KW |
196 | EOT |
197 | }, | |
3951f999 | 198 | $genconfig_scn => { |
00d68f0d KW |
199 | header => <<~'EOT', |
200 | This section contains configuration information not otherwise | |
201 | found in the more specialized sections of this document. At the | |
202 | end is a list of C<#defines> whose name should be enough to tell | |
203 | you what they do, and a list of #defines which tell you if you | |
204 | need to C<#include> files to get the corresponding functionality. | |
205 | EOT | |
206 | ||
207 | footer => <<~'EOT', | |
208 | ||
209 | =head2 List of capability C<HAS_I<foo>> symbols | |
210 | ||
6927cd44 KW |
211 | This is a list of those symbols that dont appear elsewhere in ths |
212 | document that indicate if the current platform has a certain | |
213 | capability. Their names all begin with C<HAS_>. Only those | |
214 | symbols whose capability is directly derived from the name are | |
215 | listed here. All others have their meaning expanded out elsewhere | |
216 | in this document. This (relatively) compact list is because we | |
217 | think that the expansion would add little or no value and take up | |
218 | a lot of space (because there are so many). If you think certain | |
219 | ones should be expanded, send email to | |
220 | L<perl5-porters@perl.org|mailto:perl5-porters@perl.org>. | |
00d68f0d KW |
221 | |
222 | Each symbol here will be C<#define>d if and only if the platform | |
223 | has the capability. If you need more detail, see the | |
6927cd44 KW |
224 | corresponding entry in F<config.h>. For convenience, the list is |
225 | split so that the ones that indicate there is a reentrant version | |
226 | of a capability are listed separately | |
00d68f0d KW |
227 | |
228 | __HAS_LIST__ | |
229 | ||
6927cd44 KW |
230 | And, the reentrant capabilities: |
231 | ||
232 | __HAS_R_LIST__ | |
233 | ||
00d68f0d KW |
234 | Example usage: |
235 | ||
236 | =over | |
237 | ||
238 | #ifdef HAS_STRNLEN | |
239 | use strnlen() | |
240 | #else | |
241 | use an alternative implementation | |
242 | #endif | |
243 | ||
244 | =back | |
245 | ||
246 | =head2 List of C<#include> needed symbols | |
247 | ||
248 | This list contains symbols that indicate if certain C<#include> | |
249 | files are present on the platform. If your code accesses the | |
250 | functionality that one of these is for, you will need to | |
251 | C<#include> it if the symbol on this list is C<#define>d. For | |
252 | more detail, see the corresponding entry in F<config.h>. | |
253 | ||
254 | __INCLUDE_LIST__ | |
255 | ||
256 | Example usage: | |
257 | ||
258 | =over | |
259 | ||
260 | #ifdef I_WCHAR | |
261 | #include <wchar.h> | |
262 | #endif | |
263 | ||
264 | =back | |
265 | EOT | |
266 | }, | |
3951f999 KW |
267 | $globals_scn => {}, |
268 | $GV_scn => {}, | |
269 | $hook_scn => {}, | |
270 | $HV_scn => {}, | |
271 | $io_scn => {}, | |
272 | $integer_scn => {}, | |
273 | $lexer_scn => {}, | |
274 | $locales_scn => {}, | |
275 | $magic_scn => {}, | |
276 | $memory_scn => {}, | |
277 | $mro_scn => {}, | |
278 | $multicall_scn => {}, | |
279 | $numeric_scn => {}, | |
280 | $optree_construction_scn => {}, | |
281 | $optree_manipulation_scn => {}, | |
282 | $pack_scn => {}, | |
283 | $pad_scn => {}, | |
284 | $password_scn => {}, | |
285 | $paths_scn => {}, | |
286 | $intrpvar_scn => {}, | |
287 | $prototypes_scn => {}, | |
288 | $regexp_scn => {}, | |
289 | $signals_scn => {}, | |
290 | $site_scn => { | |
6927cd44 KW |
291 | header => <<~'EOT', |
292 | These variables give details as to where various libraries, | |
293 | installation destinations, I<etc.>, go, as well as what various | |
294 | installation options were selected | |
295 | EOT | |
296 | }, | |
3951f999 KW |
297 | $sockets_scn => {}, |
298 | $filters_scn => {}, | |
299 | $stack_scn => {}, | |
300 | $string_scn => { | |
6927cd44 KW |
301 | header => <<~'EOT', |
302 | See also C<L</Unicode Support>>. | |
303 | EOT | |
304 | }, | |
3951f999 KW |
305 | $SV_flags_scn => {}, |
306 | $SV_scn => {}, | |
307 | $time_scn => {}, | |
308 | $typedefs_scn => {}, | |
309 | $unicode_scn => {}, | |
310 | $utility_scn => {}, | |
311 | $versioning_scn => {}, | |
312 | $warning_scn => {}, | |
313 | $xs_scn => {}, | |
00d68f0d | 314 | ); |
94bdecf9 | 315 | |
9e019b59 KW |
316 | # Somewhat loose match for an apidoc line so we can catch minor typos. |
317 | # Parentheses are used to capture portions so that below we verify | |
318 | # that things are the actual correct syntax. | |
319 | my $apidoc_re = qr/ ^ (\s*) # $1 | |
320 | (=?) # $2 | |
321 | (\s*) # $3 | |
322 | for (\s*) # $4 | |
323 | apidoc (_item)? # $5 | |
324 | (\s*) # $6 | |
325 | (.*?) # $7 | |
326 | \s* \n /x; | |
183f6e59 KW |
327 | # Only certain flags, dealing with display, are acceptable for apidoc_item |
328 | my $display_flags = "fFnDopsT"; | |
9e019b59 KW |
329 | |
330 | sub check_api_doc_line ($$) { | |
331 | my ($file, $in) = @_; | |
332 | ||
333 | return unless $in =~ $apidoc_re; | |
334 | ||
335 | my $is_item = defined $5; | |
336 | my $is_in_proper_form = length $1 == 0 | |
337 | && length $2 > 0 | |
338 | && length $3 == 0 | |
339 | && length $4 > 0 | |
340 | && length $6 > 0 | |
341 | && length $7 > 0; | |
342 | my $proto_in_file = $7; | |
343 | my $proto = $proto_in_file; | |
344 | $proto = "||$proto" if $proto !~ /\|/; | |
345 | my ($flags, $ret_type, $name, @args) = split /\s*\|\s*/, $proto; | |
346 | ||
347 | $name && $is_in_proper_form or die <<EOS; | |
348 | Bad apidoc at $file line $.: | |
349 | $in | |
350 | Expected: | |
351 | =for apidoc flags|returntype|name|arg|arg|... | |
352 | =for apidoc flags|returntype|name | |
353 | =for apidoc name | |
354 | (or 'apidoc_item') | |
355 | EOS | |
356 | ||
56aac0d0 | 357 | die "Only [$display_flags] allowed in apidoc_item:\n$in" |
783469eb KW |
358 | if $is_item && $flags =~ /[^$display_flags]/; |
359 | ||
9e019b59 KW |
360 | return ($name, $flags, $ret_type, $is_item, $proto_in_file, @args); |
361 | } | |
94bdecf9 | 362 | |
4be3b9ca KW |
363 | sub embed_override($) { |
364 | my ($element_name) = shift; | |
365 | ||
366 | # If the entry is also in embed.fnc, it should be defined | |
367 | # completely there, but not here | |
368 | my $embed_docref = delete $funcflags{$element_name}; | |
369 | ||
370 | return unless $embed_docref and %$embed_docref; | |
371 | ||
372 | my $flags = $embed_docref->{'flags'}; | |
373 | warn "embed.fnc entry '$element_name' missing 'd' flag" | |
374 | unless $flags =~ /d/; | |
375 | ||
376 | return ($flags, $embed_docref->{'ret_type'}, $embed_docref->{args}->@*); | |
377 | } | |
378 | ||
94bdecf9 JH |
379 | sub autodoc ($$) { # parse a file and extract documentation info |
380 | my($fh,$file) = @_; | |
63da8140 | 381 | my($in, $line_num, $header, $section); |
f554dfc5 | 382 | |
3cfa66c4 KW |
383 | my $file_is_C = $file =~ / \. [ch] $ /x; |
384 | ||
f554dfc5 | 385 | # Count lines easier |
b1cb8a49 | 386 | my $get_next_line = sub { $line_num++; return <$fh> }; |
f554dfc5 | 387 | |
63da8140 KW |
388 | # Read the file |
389 | while ($in = $get_next_line->()) { | |
390 | last unless defined $in; | |
3cfa66c4 | 391 | |
63da8140 KW |
392 | next unless ( $in =~ / ^ =for [ ]+ apidoc /x |
393 | # =head1 lines only have effect in C files | |
394 | || ($file_is_C && $in =~ /^=head1/)); | |
395 | ||
396 | # Here, the line introduces a portion of the input that we care about. | |
397 | # Either it is for an API element, or heading text which we expect | |
398 | # will be used for elements later in the file | |
399 | ||
400 | my ($text, $element_name, $flags, $ret_type, $is_item, $proto_in_file); | |
183f6e59 | 401 | my (@args, @items); |
63da8140 KW |
402 | |
403 | # If the line starts a new section ... | |
404 | if ($in=~ /^ = (?: for [ ]+ apidoc_section | head1 ) [ ]+ (.*) /x) { | |
183f6e59 | 405 | |
b1cb8a49 | 406 | $section = $1; |
6927cd44 KW |
407 | die "Unknown section name '$section' in $file near line $.\n" |
408 | unless defined $valid_sections{$section}; | |
409 | ||
3cfa66c4 | 410 | } |
63da8140 KW |
411 | elsif ($in=~ /^ =for [ ]+ apidoc \B /x) { # Otherwise better be a |
412 | # plain apidoc line | |
183f6e59 KW |
413 | die "Unkown apidoc-type line '$in'" unless $in=~ /^=for apidoc_item/; |
414 | die "apidoc_item doesn't immediately follow an apidoc entry: '$in'"; | |
63da8140 KW |
415 | } |
416 | else { # Plain apidoc | |
417 | ||
418 | ($element_name, $flags, $ret_type, $is_item, $proto_in_file, @args) | |
419 | = check_api_doc_line($file, $in); | |
4be3b9ca KW |
420 | # Override this line with any info in embed.fnc |
421 | my ($embed_flags, $embed_ret_type, @embed_args) | |
422 | = embed_override($element_name); | |
423 | if ($embed_ret_type) { | |
63da8140 | 424 | warn "embed.fnc entry overrides redundant information in" |
4be3b9ca KW |
425 | . " '$proto_in_file' in $file" |
426 | if $flags || $ret_type || @args; | |
427 | $flags = $embed_flags; | |
428 | $ret_type = $embed_ret_type; | |
429 | @args = @embed_args; | |
430 | } | |
9e728159 KW |
431 | elsif ($flags !~ /[my]/) { # Not in embed.fnc, is missing if not |
432 | # a macro or typedef | |
63da8140 KW |
433 | $missing{$element_name} = $file; |
434 | } | |
3cfa66c4 | 435 | |
56aac0d0 | 436 | die "flag '$1' is not legal (for function $element_name (from $file))" |
9e728159 | 437 | if $flags =~ / ( [^AabCDdEeFfhiMmNnTOoPpRrSsUuWXxy] ) /x; |
151c3fe5 | 438 | |
9e728159 KW |
439 | die "'u' flag must also have 'm' or 'y' flags' for $element_name" |
440 | if $flags =~ /u/ && $flags !~ /[my]/; | |
63da8140 KW |
441 | warn ("'$element_name' not \\w+ in '$proto_in_file' in $file") |
442 | if $flags !~ /N/ && $element_name !~ / ^ [_[:alpha:]] \w* $ /x; | |
f554dfc5 | 443 | |
63da8140 KW |
444 | if (exists $seen{$element_name} && $flags !~ /h/) { |
445 | die ("'$element_name' in $file was already documented in $seen{$element_name}"); | |
446 | } | |
447 | else { | |
448 | $seen{$element_name} = $file; | |
151c3fe5 | 449 | } |
94bdecf9 | 450 | } |
df6bd76f | 451 | |
63da8140 KW |
452 | # Here we have processed the initial line in the heading text or API |
453 | # element, and have saved the important information from it into the | |
454 | # corresponding variables. Now accumulate the text that applies to it | |
455 | # up to a terminating line, which is one of: | |
456 | # 1) =cut | |
457 | # 2) =head (in a C file only =head1) | |
458 | # 3) an end comment line in a C file: m:^\s*\*/: | |
183f6e59 | 459 | # 4) =for apidoc... (except apidoc_item lines) |
63da8140 KW |
460 | $text = ""; |
461 | my $head_ender_num = ($file_is_C) ? 1 : ""; | |
462 | while (defined($in = $get_next_line->())) { | |
463 | ||
464 | last if $in =~ /^=cut/x; | |
465 | last if $in =~ /^=head$head_ender_num/; | |
466 | ||
467 | if ($file_is_C && $in =~ m: ^ \s* \* / $ :x) { | |
468 | ||
469 | # End of comment line in C files is a fall-back terminator, | |
470 | # but warn only if there actually is some accumulated text | |
471 | warn "=cut missing? $file:$line_num:$in" if $text =~ /\S/; | |
472 | last; | |
df6bd76f | 473 | } |
df6bd76f | 474 | |
63da8140 KW |
475 | if ($in !~ / ^ =for [ ]+ apidoc /x) { |
476 | $text .= $in; | |
477 | next; | |
478 | } | |
479 | ||
183f6e59 | 480 | # Here, the line is an apidoc line. All but apidoc_item terminate |
63da8140 | 481 | # the text being accumulated. |
183f6e59 KW |
482 | last if $in =~ / ^ =for [ ]+ apidoc_section /x; |
483 | ||
484 | my ($item_name, $item_flags, $item_ret_type, $is_item, | |
4be3b9ca | 485 | $item_proto, @item_args) = check_api_doc_line($file, $in); |
183f6e59 KW |
486 | last unless $is_item; |
487 | ||
488 | # Here, is an apidoc_item_line; They can only come within apidoc | |
489 | # paragraphs. | |
4be3b9ca KW |
490 | die "Unexpected api_doc_item line '$item_proto'" |
491 | unless $element_name; | |
183f6e59 KW |
492 | |
493 | # We accept blank lines between these, but nothing else; | |
56aac0d0 KW |
494 | die "apidoc_item lines must immediately follow apidoc lines for " |
495 | . " '$element_name' in $file" | |
183f6e59 | 496 | if $text =~ /\S/; |
4be3b9ca KW |
497 | # Override this line with any info in embed.fnc |
498 | my ($embed_flags, $embed_ret_type, @embed_args) | |
499 | = embed_override($item_name); | |
500 | if ($embed_ret_type) { | |
501 | warn "embed.fnc entry overrides redundant information in" | |
502 | . " '$item_proto' in $file" | |
503 | if $item_flags || $item_ret_type || @item_args; | |
504 | ||
505 | $item_flags = $embed_flags; | |
506 | $item_ret_type = $embed_ret_type; | |
507 | @item_args = @embed_args; | |
508 | } | |
183f6e59 KW |
509 | |
510 | # Use the base entry flags if none for this item; otherwise add in | |
511 | # any non-display base entry flags. | |
512 | if ($item_flags) { | |
513 | $item_flags .= $flags =~ s/[$display_flags]//rg; | |
514 | } | |
515 | else { | |
516 | $item_flags = $flags; | |
517 | } | |
518 | $item_ret_type = $ret_type unless $item_ret_type; | |
519 | @item_args = @args unless @item_args; | |
520 | push @items, { name => $item_name, | |
521 | ret_type => $item_ret_type, | |
522 | flags => $item_flags, | |
523 | args => [ @item_args ], | |
524 | }; | |
525 | ||
526 | # This line shows that this element is documented. | |
527 | delete $funcflags{$item_name}; | |
a9a7c81e | 528 | } |
3cfa66c4 | 529 | |
63da8140 KW |
530 | # Here, are done accumulating the text for this item. Trim it |
531 | $text =~ s/ ^ \s* //x; | |
532 | $text =~ s/ \s* $ //x; | |
533 | $text .= "\n" if $text ne ""; | |
6523e108 | 534 | |
63da8140 KW |
535 | # And treat all-spaces as nothing at all |
536 | undef $text unless $text =~ /\S/; | |
df6bd76f | 537 | |
63da8140 | 538 | if ($element_name) { |
df6bd76f | 539 | |
63da8140 KW |
540 | # Here, we have accumulated into $text, the pod for $element_name |
541 | my $where = $flags =~ /A/ ? 'api' : 'guts'; | |
5ce57792 | 542 | |
63da8140 KW |
543 | $section = "Functions in file $file" unless defined $section; |
544 | die "No =for apidoc_section nor =head1 in $file for '$element_name'\n" | |
545 | unless defined $section; | |
546 | if (exists $docs{$where}{$section}{$element_name}) { | |
547 | warn "$0: duplicate API entry for '$element_name' in" | |
548 | . " $where/$section\n"; | |
549 | next; | |
20046047 | 550 | } |
5ce57792 | 551 | |
63da8140 KW |
552 | # Override the text with just a link if the flags call for that |
553 | my $is_link_only = ($flags =~ /h/); | |
554 | if ($is_link_only) { | |
555 | if ($file_is_C) { | |
56aac0d0 | 556 | die "Can't currently handle link with items to it:\n$in" if @items; |
63da8140 KW |
557 | redo; # Don't put anything if C source |
558 | } | |
559 | ||
560 | # Here, is an 'h' flag in pod. We add a reference to the pod (and | |
561 | # nothing else) to perlapi/intern. (It would be better to add a | |
562 | # reference to the correct =item,=header, but something that makes | |
563 | # it harder is that it that might be a duplicate, like '=item *'; | |
564 | # so that is a future enhancement XXX. Another complication is | |
565 | # there might be more than one deserving candidates.) | |
566 | my $podname = $file =~ s!.*/!!r; # Rmv directory name(s) | |
567 | $podname =~ s/\.pod//; | |
568 | $text = "Described in L<$podname>.\n"; | |
569 | ||
570 | # Don't output a usage example for linked to documentation if | |
571 | # it is trivial (has no arguments) and we aren't to add a | |
572 | # semicolon | |
573 | $flags .= 'U' if $flags =~ /n/ && $flags !~ /[Us]/; | |
574 | ||
575 | # Keep track of all the pod files that we refer to. | |
576 | push $described_elsewhere{$podname}->@*, $podname; | |
577 | } | |
5ce57792 | 578 | |
5bf02caf KW |
579 | $docs{$where}{$section}{$element_name}{flags} = $flags; |
580 | $docs{$where}{$section}{$element_name}{pod} = $text; | |
581 | $docs{$where}{$section}{$element_name}{file} = $file; | |
582 | $docs{$where}{$section}{$element_name}{ret_type} = $ret_type; | |
583 | push $docs{$where}{$section}{$element_name}{args}->@*, @args; | |
183f6e59 | 584 | push $docs{$where}{$section}{$element_name}{items}->@*, @items; |
a9a7c81e | 585 | } |
63da8140 KW |
586 | elsif ($text) { |
587 | $valid_sections{$section}{header} = "" unless | |
588 | defined $valid_sections{$section}{header}; | |
589 | $valid_sections{$section}{header} .= "\n$text"; | |
a9a7c81e | 590 | } |
63da8140 KW |
591 | |
592 | # We already have the first line of what's to come in $in | |
593 | redo; | |
594 | ||
595 | } # End of loop through input | |
94bdecf9 JH |
596 | } |
597 | ||
00d68f0d KW |
598 | my %configs; |
599 | my @has_defs; | |
600 | my @has_r_defs; # Reentrant symbols | |
601 | my @include_defs; | |
602 | sub parse_config_h { | |
603 | use re '/aa'; # Everthing is ASCII in this file | |
604 | ||
605 | # Process config.h | |
d3871820 KW |
606 | my $config_h = 'config.h'; |
607 | $config_h = 'win32/config.h' unless -e $config_h; | |
608 | die "Can't find $config_h" unless -e $config_h; | |
609 | open my $fh, '<', $config_h or die "Can't open $config_h: $!"; | |
00d68f0d KW |
610 | while (<$fh>) { |
611 | ||
612 | # Look for lines like /* FOO_BAR: | |
613 | # By convention all config.h descriptions begin like that | |
614 | if (m[ ^ /\* [ ] ( [[:alpha:]] \w+ ) : \s* $ ]ax) { | |
615 | my $name = $1; | |
616 | ||
617 | # Here we are starting the description for $name in config.h. We | |
618 | # accumulate the entire description for it into @description. | |
619 | # Flowing text from one input line to another is appended into the | |
620 | # same array element to make a single flowing line element, but | |
621 | # verbatim lines are kept as separate elements in @description. | |
622 | # This will facilitate later doing pattern matching without regard | |
623 | # to line boundaries on non-verbatim text. | |
624 | ||
625 | die "Multiple config.h entries for '$name'" | |
626 | if defined $configs{$name}{description}; | |
627 | ||
628 | # Get first line of description | |
629 | $_ = <$fh>; | |
630 | ||
631 | # Each line in the description begins with blanks followed by '/*' | |
632 | # and some spaces. | |
633 | die "Unexpected config.h initial line for $name: '$_'" | |
634 | unless s/ ^ ( \s* \* \s* ) //x; | |
635 | my $initial_text = $1; | |
636 | ||
637 | # Initialize the description with this first line (after having | |
638 | # stripped the prefix text) | |
639 | my @description = $_; | |
640 | ||
641 | # The first line is used as a template for how much indentation | |
642 | # each normal succeeding line has. Lines indented further | |
643 | # will be considered as intended to be verbatim. But, empty lines | |
644 | # likely won't have trailing blanks, so just strip the whole thing | |
645 | # for them. | |
646 | my $strip_initial_qr = qr! \s* \* \s* $ | |
647 | | \Q$initial_text\E | |
648 | !x; | |
649 | $configs{$name}{verbatim} = 0; | |
650 | ||
651 | # Read in the remainder of the description | |
652 | while (<$fh>) { | |
653 | last if s| ^ \s* \* / ||x; # A '*/' ends it | |
654 | ||
655 | die "Unexpected config.h description line for $name: '$_'" | |
656 | unless s/$strip_initial_qr//; | |
657 | ||
658 | # Fix up the few flawed lines in config.h wherein a new | |
659 | # sentence begins with a tab (and maybe a space after that). | |
660 | # Although none of them currently do, let it recognize | |
661 | # something like | |
662 | # | |
663 | # "... text"). The next sentence ... | |
664 | # | |
665 | s/ ( \w "? \)? \. ) \t \s* ( [[:alpha:]] ) /$1 $2/xg; | |
666 | ||
667 | # If this line has extra indentation or looks to have columns, | |
668 | # it should be treated as verbatim. Columns are indicated by | |
669 | # use of interior: tabs, 3 spaces in a row, or even 2 spaces | |
670 | # not preceded by punctuation. | |
671 | if ($_ !~ m/ ^ \s | |
672 | | \S (?: \t | |
673 | | \s{3} | |
674 | | (*nlb:[[:punct:]]) \s{2} | |
675 | ) | |
676 | /x) | |
677 | { | |
678 | # But here, is not a verbatim line. Add an empty line if | |
679 | # this is the first non-verbatim after a run of verbatims | |
680 | if ($description[-1] =~ /^\s/) { | |
681 | push @description, "\n", $_; | |
682 | } | |
683 | else { # Otherwise, append this flowing line to the | |
684 | # current flowing line | |
685 | $description[-1] .= $_; | |
686 | } | |
687 | } | |
688 | else { | |
689 | $configs{$name}{verbatim} = 1; | |
690 | ||
691 | # The first verbatim line in a run of them is separated by an | |
692 | # empty line from the flowing lines above it | |
693 | push @description, "\n" if $description[-1] =~ /^\S/; | |
694 | ||
695 | $_ = Text::Tabs::expand($_); | |
696 | ||
697 | # Only a single space so less likely to wrap | |
698 | s/ ^ \s* / /x; | |
699 | ||
700 | push @description, $_; | |
701 | } | |
702 | } | |
703 | ||
704 | push $configs{$name}{description}->@*, @description | |
705 | ||
706 | } # Not a description; see if it is a macro definition. | |
707 | elsif (m! ^ | |
708 | (?: / \* )? # Optional commented-out | |
709 | # indication | |
710 | \# \s* define \s+ ( \w+ ) # $1 is the name | |
711 | ( \s* ) # $2 indicates if args or not | |
712 | ( .*? ) # $3 is any definition | |
713 | (?: / \s* \* \* / )? # Optional trailing /**/ or / **/ | |
714 | $ | |
715 | !x) | |
716 | { | |
717 | my $name = $1; | |
718 | ||
719 | # There can be multiple definitions for a name. We want to know | |
720 | # if any of them has arguments, and if any has a body. | |
721 | $configs{$name}{has_args} //= $2 eq ""; | |
722 | $configs{$name}{has_args} ||= $2 eq ""; | |
723 | $configs{$name}{has_defn} //= $3 ne ""; | |
724 | $configs{$name}{has_defn} ||= $3 ne ""; | |
725 | } | |
726 | } | |
727 | ||
728 | # We now have stored the description and information about every #define | |
729 | # in the file. The description is in a form convenient to operate on to | |
730 | # convert to pod. Do that now. | |
731 | foreach my $name (keys %configs) { | |
732 | next unless defined $configs{$name}{description}; | |
733 | ||
734 | # All adjacent non-verbatim lines of the description are appended | |
735 | # together in a single element in the array. This allows the patterns | |
736 | # to work across input line boundaries. | |
737 | ||
738 | my $pod = ""; | |
739 | while (defined ($_ = shift $configs{$name}{description}->@*)) { | |
740 | chomp; | |
741 | ||
742 | if (/ ^ \S /x) { # Don't edit verbatim lines | |
743 | ||
744 | # Enclose known file/path names not already so enclosed | |
745 | # with <...>. (Some entries in config.h are already | |
746 | # '<path/to/file>') | |
747 | my $file_name_qr = qr! [ \w / ]+ \. | |
748 | (?: c | h | xs | p [lm] | pmc | PL | |
749 | | sh | SH | exe ) \b | |
750 | !xx; | |
751 | my $path_name_qr = qr! (?: / \w+ )+ !x; | |
752 | for my $re ($file_name_qr, $path_name_qr) { | |
753 | s! (*nlb:[ < \w / ]) ( $re ) !<$1>!gxx; | |
754 | } | |
755 | ||
756 | # Enclose <... file/path names with F<...> (but no double | |
757 | # angle brackets) | |
758 | for my $re ($file_name_qr, $path_name_qr) { | |
759 | s! < ( $re ) > !F<$1>!gxx; | |
760 | } | |
761 | ||
762 | # Explain metaconfig units | |
763 | s/ ( \w+ \. U \b ) /$1 (part of metaconfig)/gx; | |
764 | ||
765 | # Convert "See foo" to "See C<L</foo>>" if foo is described in | |
766 | # this file. Also create a link to the known file INSTALL. | |
767 | # And, to be more general, handle "See also foo and bar", and | |
768 | # "See also foo, bar, and baz" | |
769 | while (m/ \b [Ss]ee \s+ | |
770 | (?: also \s+ )? ( \w+ ) | |
771 | (?: , \s+ ( \w+ ) )? | |
772 | (?: ,? \s+ and \s+ ( \w+ ) )? /xg) { | |
773 | my @links = $1; | |
774 | push @links, $2 if defined $2; | |
775 | push @links, $3 if defined $3; | |
776 | foreach my $link (@links) { | |
777 | if ($link eq 'INSTALL') { | |
778 | s/ \b INSTALL \b /C<L<INSTALL>>/xg; | |
779 | } | |
780 | elsif (grep { $link =~ / \b $_ \b /x } keys %configs) { | |
781 | s| \b $link \b |C<L</$link>>|xg; | |
782 | $configs{$link}{linked} = 1; | |
783 | $configs{$name}{linked} = 1; | |
784 | } | |
785 | } | |
786 | } | |
787 | ||
788 | # Enclose what we think are symbols with C<...>. | |
789 | no warnings 'experimental::vlb'; | |
790 | s/ (*nlb:<) | |
791 | ( | |
792 | # Any word followed immediately with parens or | |
793 | # brackets | |
794 | \b \w+ (?: \( [^)]* \) # parameter list | |
795 | | \[ [^]]* \] # or array reference | |
796 | ) | |
797 | | (*plb: ^ | \s ) -D \w+ # Also -Dsymbols. | |
798 | | \b (?: struct | union ) \s \w+ | |
799 | ||
800 | # Words that contain underscores (which are | |
801 | # definitely not text) or three uppercase letters in | |
802 | # a row. Length two ones, like IV, aren't enclosed, | |
803 | # because they often don't look as nice. | |
804 | | \b \w* (?: _ | [[:upper:]]{3,} ) \w* \b | |
805 | ) | |
806 | (*nla:>) | |
807 | /C<$1>/xg; | |
808 | ||
809 | # These include foo when the name is HAS_foo. This is a | |
810 | # heuristic which works in most cases. | |
811 | if ($name =~ / ^ HAS_ (.*) /x) { | |
812 | my $symbol = lc $1; | |
813 | ||
814 | # Don't include path components, nor things already in | |
815 | # <>, or with trailing '(', '[' | |
816 | s! \b (*nlb:[/<]) $symbol (*nla:[[/>(]) \b !C<$symbol>!xg; | |
817 | } | |
818 | } | |
819 | ||
820 | $pod .= "$_\n"; | |
821 | } | |
822 | delete $configs{$name}{description}; | |
823 | ||
824 | $configs{$name}{pod} = $pod; | |
825 | } | |
826 | ||
827 | # Now have converted the description to pod. We also now have enough | |
828 | # information that we can do cross checking to find definitions without | |
829 | # corresponding pod, and see if they are mentioned in some description; | |
830 | # otherwise they aren't documented. | |
831 | NAME: | |
832 | foreach my $name (keys %configs) { | |
833 | ||
834 | # A definition without pod | |
835 | if (! defined $configs{$name}{pod}) { | |
836 | ||
837 | # Leading/trailing underscore means internal to config.h, e.g., | |
838 | # _GNU_SOURCE | |
839 | next if $name =~ / ^ _ /x; | |
840 | next if $name =~ / _ $ /x; | |
841 | ||
842 | # MiXeD case names are internal to config.h; the first 4 | |
843 | # characters are sufficient to determine this | |
844 | next if $name =~ / ^ [[:upper:]] [[:lower:]] | |
845 | [[:upper:]] [[:lower:]] | |
846 | /x; | |
847 | ||
848 | # Here, not internal to config.h. Look to see if this symbol is | |
849 | # mentioned in the pod of some other. If so, assume it is | |
850 | # documented. | |
851 | foreach my $check_name (keys %configs) { | |
852 | my $this_element = $configs{$check_name}; | |
853 | my $this_pod = $this_element->{pod}; | |
854 | if (defined $this_pod) { | |
855 | next NAME if $this_pod =~ / \b $name \b /x; | |
856 | } | |
857 | } | |
858 | ||
859 | warn "$name has no documentation\n"; | |
860 | $missing_macros{$name} = 'config.h'; | |
861 | ||
862 | next; | |
863 | } | |
864 | ||
865 | my $has_defn = $configs{$name}{has_defn}; | |
866 | my $has_args = $configs{$name}{has_args}; | |
867 | ||
868 | # Check if any section already has an entry for this element. | |
869 | # If so, it better be a placeholder, in which case we replace it | |
870 | # with this entry. | |
871 | foreach my $section (keys $docs{'api'}->%*) { | |
872 | if (exists $docs{'api'}{$section}{$name}) { | |
873 | my $was = $docs{'api'}{$section}{$name}->{pod}; | |
874 | $was = "" unless $was; | |
875 | chomp $was; | |
876 | if ($was ne "" && $was !~ m/$link_text/) { | |
877 | die "Multiple descriptions for $name\n" | |
878 | . "$section contained '$was'"; | |
879 | } | |
880 | $docs{'api'}{$section}{$name}->{pod} = $configs{$name}{pod}; | |
881 | $configs{$name}{section} = $section; | |
882 | last; | |
883 | } | |
884 | } | |
885 | ||
886 | my $handled = 0; # Haven't handled this yet | |
887 | ||
888 | if (defined $configs{$name}{'section'}) { | |
889 | # This has been taken care of elsewhere. | |
890 | $handled = 1; | |
891 | } | |
892 | else { | |
9e728159 | 893 | my $flags = ""; |
00d68f0d KW |
894 | if ($has_defn && ! $has_args) { |
895 | $configs{$name}{args} = 1; | |
896 | } | |
897 | ||
898 | # Symbols of the form I_FOO are for #include files. They have | |
899 | # special usage information | |
900 | if ($name =~ / ^ I_ ( .* ) /x) { | |
901 | my $file = lc $1 . '.h'; | |
902 | $configs{$name}{usage} = <<~"EOT"; | |
903 | #ifdef $name | |
904 | #include <$file> | |
905 | #endif | |
906 | EOT | |
907 | } | |
908 | ||
909 | # Compute what section this variable should go into. This | |
910 | # heuristic was determined by manually inspecting the current | |
911 | # things in config.h, and should be adjusted as necessary as | |
912 | # deficiencies are found. | |
913 | # | |
914 | # This is the default section for macros with a definiton but | |
915 | # no arguments, meaning it is replaced unconditionally | |
916 | # | |
917 | my $sb = qr/ _ | \b /x; # segment boundary | |
918 | my $dash_or_spaces = qr/ - | \s+ /x; | |
919 | my $pod = $configs{$name}{pod}; | |
920 | if ($name =~ / ^ USE_ /x) { | |
3951f999 | 921 | $configs{$name}{'section'} = $site_scn; |
00d68f0d KW |
922 | } |
923 | elsif ($name =~ / SLEEP | (*nlb:SYS_) TIME | TZ | $sb TM $sb /x) | |
924 | { | |
3951f999 | 925 | $configs{$name}{'section'} = $time_scn; |
00d68f0d KW |
926 | } |
927 | elsif ($name =~ / DOUBLE | FLOAT | LONGDBL | LDBL | ^ NV | |
928 | | $sb CASTFLAGS $sb | |
929 | | QUADMATH | |
930 | | $sb (?: IS )? NAN | |
931 | | $sb (?: IS )? FINITE | |
932 | /x) | |
933 | { | |
934 | $configs{$name}{'section'} = | |
3951f999 | 935 | $floating_scn; |
00d68f0d KW |
936 | } |
937 | elsif ($name =~ / (?: POS | OFF | DIR ) 64 /x) { | |
3951f999 | 938 | $configs{$name}{'section'} = $filesystem_scn; |
00d68f0d KW |
939 | } |
940 | elsif ( $name =~ / $sb (?: BUILTIN | CPP ) $sb | ^ CPP /x | |
941 | || $configs{$name}{pod} =~ m/ \b align /x) | |
942 | { | |
3951f999 | 943 | $configs{$name}{'section'} = $compiler_scn; |
00d68f0d KW |
944 | } |
945 | elsif ($name =~ / ^ [IU] [ \d V ] | |
946 | | ^ INT | SHORT | LONG | QUAD | 64 | 32 /xx) | |
947 | { | |
3951f999 | 948 | $configs{$name}{'section'} = $integer_scn; |
00d68f0d KW |
949 | } |
950 | elsif ($name =~ / $sb t $sb /x) { | |
3951f999 | 951 | $configs{$name}{'section'} = $typedefs_scn; |
9e728159 | 952 | $flags .= 'y'; |
00d68f0d KW |
953 | } |
954 | elsif ( $name =~ / ^ PERL_ ( PRI | SCN ) | $sb FORMAT $sb /x | |
955 | && $configs{$name}{pod} =~ m/ \b format \b /ix) | |
956 | { | |
3951f999 | 957 | $configs{$name}{'section'} = $formats_scn; |
00d68f0d KW |
958 | } |
959 | elsif ($name =~ / BACKTRACE /x) { | |
3951f999 | 960 | $configs{$name}{'section'} = $dump_scn; |
00d68f0d KW |
961 | } |
962 | elsif ($name =~ / ALLOC $sb /x) { | |
3951f999 | 963 | $configs{$name}{'section'} = $memory_scn; |
00d68f0d KW |
964 | } |
965 | elsif ( $name =~ / STDIO | FCNTL | EOF | FFLUSH | |
966 | | $sb FILE $sb | |
967 | | $sb DIR $sb | |
968 | | $sb LSEEK | |
969 | | $sb INO $sb | |
970 | | $sb OPEN | |
971 | | $sb CLOSE | |
972 | | ^ DIR | |
973 | | ^ INO $sb | |
974 | | DIR $ | |
975 | | FILENAMES | |
976 | /x | |
977 | || $configs{$name}{pod} =~ m! I/O | stdio | |
978 | | file \s+ descriptor | |
979 | | file \s* system | |
980 | | statfs | |
981 | !x) | |
982 | { | |
3951f999 | 983 | $configs{$name}{'section'} = $filesystem_scn; |
00d68f0d KW |
984 | } |
985 | elsif ($name =~ / ^ SIG | SIGINFO | signal /ix) { | |
3951f999 | 986 | $configs{$name}{'section'} = $signals_scn; |
00d68f0d KW |
987 | } |
988 | elsif ($name =~ / $sb ( PROTO (?: TYPE)? S? ) $sb /x) { | |
3951f999 | 989 | $configs{$name}{'section'} = $prototypes_scn; |
00d68f0d KW |
990 | } |
991 | elsif ( $name =~ / ^ LOC_ /x | |
992 | || $configs{$name}{pod} =~ /full path/i) | |
993 | { | |
3951f999 | 994 | $configs{$name}{'section'} = $paths_scn; |
00d68f0d KW |
995 | } |
996 | elsif ($name =~ / $sb LC_ | LOCALE | langinfo /xi) { | |
3951f999 | 997 | $configs{$name}{'section'} = $locales_scn; |
00d68f0d KW |
998 | } |
999 | elsif ($configs{$name}{pod} =~ / GCC | C99 | C\+\+ /xi) { | |
3951f999 | 1000 | $configs{$name}{'section'} = $compiler_scn; |
00d68f0d KW |
1001 | } |
1002 | elsif ($name =~ / PASSW (OR)? D | ^ PW | ( PW | GR ) ENT /x) | |
1003 | { | |
3951f999 | 1004 | $configs{$name}{'section'} = $password_scn; |
00d68f0d KW |
1005 | } |
1006 | elsif ($name =~ / SOCKET | $sb SOCK /x) { | |
3951f999 | 1007 | $configs{$name}{'section'} = $sockets_scn; |
00d68f0d KW |
1008 | } |
1009 | elsif ( $name =~ / THREAD | MULTIPLICITY /x | |
1010 | || $configs{$name}{pod} =~ m/ \b pthread /ix) | |
1011 | { | |
3951f999 | 1012 | $configs{$name}{'section'} = $concurrency_scn; |
00d68f0d KW |
1013 | } |
1014 | elsif ($name =~ / PERL | ^ PRIV | SITE | ARCH | BIN | |
1015 | | VENDOR | ^ USE | |
1016 | /x) | |
1017 | { | |
3951f999 | 1018 | $configs{$name}{'section'} = $site_scn; |
00d68f0d KW |
1019 | } |
1020 | elsif ( $pod =~ / \b floating $dash_or_spaces point \b /ix | |
1021 | || $pod =~ / \b (double | single) $dash_or_spaces precision \b /ix | |
1022 | || $pod =~ / \b doubles \b /ix | |
1023 | || $pod =~ / \b (?: a | the | long ) \s+ (?: double | NV ) \b /ix) | |
1024 | { | |
1025 | $configs{$name}{'section'} = | |
3951f999 | 1026 | $floating_scn; |
00d68f0d KW |
1027 | } |
1028 | else { | |
1029 | # Above are the specific sections. The rest go into a | |
1030 | # grab-bag of general configuration values. However, we put | |
1031 | # two classes of them into lists of their names, without their | |
1032 | # descriptions, when we think that the description doesn't add | |
1033 | # any real value. One list contains the #include variables: | |
1034 | # the description is basically boiler plate for each of these. | |
1035 | # The other list contains the very many things that are of the | |
1036 | # form HAS_foo, and \bfoo\b is contained in its description, | |
1037 | # and there is no verbatim text in the pod or links to/from it | |
1038 | # (which would add value). That means that it is likely the | |
1039 | # intent of the variable can be gleaned from just its name, | |
1040 | # and unlikely the description adds signficant value, so just | |
1041 | # listing them suffices. Giving their descriptions would | |
1042 | # expand this pod significantly with little added value. | |
1043 | if ( ! $has_defn | |
1044 | && ! $configs{$name}{verbatim} | |
1045 | && ! $configs{$name}{linked}) | |
1046 | { | |
1047 | if ($name =~ / ^ I_ ( .* ) /x) { | |
1048 | push @include_defs, $name; | |
1049 | next; | |
1050 | } | |
1051 | elsif ($name =~ / ^ HAS_ ( .* ) /x) { | |
1052 | my $canonical_name = $1; | |
1053 | $canonical_name =~ s/_//g; | |
1054 | ||
1055 | my $canonical_pod = $configs{$name}{pod}; | |
1056 | $canonical_pod =~ s/_//g; | |
1057 | ||
1058 | if ($canonical_pod =~ / \b $canonical_name \b /xi) { | |
1059 | if ($name =~ / $sb R $sb /x) { | |
1060 | push @has_r_defs, $name; | |
1061 | } | |
1062 | else { | |
1063 | push @has_defs, $name; | |
1064 | } | |
1065 | next; | |
1066 | } | |
1067 | } | |
1068 | } | |
1069 | ||
3951f999 | 1070 | $configs{$name}{'section'} = $genconfig_scn; |
00d68f0d KW |
1071 | } |
1072 | ||
1073 | my $section = $configs{$name}{'section'}; | |
6927cd44 KW |
1074 | die "Internal error: '$section' not in \%valid_sections" |
1075 | unless grep { $_ eq $section } keys %valid_sections; | |
9e728159 | 1076 | $flags .= 'AdmnT'; |
00d68f0d | 1077 | $flags .= 'U' unless defined $configs{$name}{usage}; |
9e728159 | 1078 | |
00d68f0d KW |
1079 | $docs{'api'}{$section}{$name}{flags} = $flags; |
1080 | $docs{'api'}{$section}{$name}{pod} = $configs{$name}{pod}; | |
1081 | $docs{'api'}{$section}{$name}{ret_type} = ""; | |
1082 | $docs{'api'}{$section}{$name}{file} = 'config.h'; | |
1083 | $docs{'api'}{$section}{$name}{usage} | |
1084 | = $configs{$name}{usage} if defined $configs{$name}{usage}; | |
1085 | push $docs{'api'}{$section}{$name}{args}->@*, (); | |
1086 | push $docs{'api'}{$section}{$name}{items}->@*, (); | |
1087 | } | |
1088 | } | |
1089 | } | |
1090 | ||
94bdecf9 | 1091 | sub docout ($$$) { # output the docs for one function |
b1cb8a49 | 1092 | my($fh, $element_name, $docref) = @_; |
5bf02caf | 1093 | |
783469eb KW |
1094 | # Trim trailing space |
1095 | $element_name =~ s/\s*$//; | |
1096 | ||
5bf02caf | 1097 | my $flags = $docref->{flags}; |
63da8140 | 1098 | my $pod = $docref->{pod} // ""; |
5bf02caf | 1099 | my $file = $docref->{file}; |
783469eb | 1100 | |
183f6e59 | 1101 | my @items = $docref->{items}->@*; |
5bf02caf | 1102 | |
783469eb KW |
1103 | # Make the main element the first of the items. This allows uniform |
1104 | # treatment below | |
1105 | unshift @items, { name => $element_name, | |
1106 | flags => $flags, | |
1107 | ret_type => $docref->{ret_type}, | |
1108 | args => [ $docref->{args}->@* ], | |
1109 | }; | |
94bdecf9 | 1110 | |
7a997878 | 1111 | warn("Empty pod for $element_name (from $file)") unless $pod =~ /\S/; |
9639517b | 1112 | |
783469eb KW |
1113 | print $fh "\n=over $description_indent\n"; |
1114 | print $fh "\n=item C<$_->{name}>\n" for @items; | |
63da8140 | 1115 | |
783469eb KW |
1116 | # If we're printing only a link to an element, this isn't the major entry, |
1117 | # so no X<> here. | |
1118 | if ($flags !~ /h/) { | |
1119 | print $fh "X<$_->{name}>" for @items; | |
1120 | print $fh "\n"; | |
d4e99c76 | 1121 | } |
63da8140 | 1122 | |
783469eb KW |
1123 | for my $item (@items) { |
1124 | if ($item->{flags} =~ /D/) { | |
1125 | print $fh <<~"EOT"; | |
54c193ae | 1126 | |
783469eb KW |
1127 | C<B<DEPRECATED!>> It is planned to remove C<$item->{name}> from a |
1128 | future release of Perl. Do not use it for new code; remove it from | |
1129 | existing code. | |
1130 | EOT | |
1131 | } | |
1132 | elsif ($item->{flags} =~ /x/) { | |
1133 | print $fh <<~"EOT"; | |
54c193ae | 1134 | |
783469eb KW |
1135 | NOTE: C<$item->{name}> is B<experimental> and may change or be |
1136 | removed without notice. | |
1137 | EOT | |
1138 | } | |
54c193ae | 1139 | } |
94bdecf9 | 1140 | |
783469eb KW |
1141 | chomp $pod; # Make sure prints pod with a single trailing \n |
1142 | print $fh "\n", $pod, "\n"; | |
3cfa66c4 | 1143 | |
783469eb KW |
1144 | for my $item (@items) { |
1145 | my $item_flags = $item->{flags}; | |
1146 | my $item_name = $item->{name}; | |
3cfa66c4 | 1147 | |
783469eb KW |
1148 | print $fh "\nNOTE: the C<perl_$item_name()> form is B<deprecated>.\n" |
1149 | if $item_flags =~ /O/; | |
1150 | # Is Perl_, but no #define foo # Perl_foo | |
1151 | if (($item_flags =~ /p/ && $item_flags =~ /o/ && $item_flags !~ /M/) | |
1152 | ||
1153 | # Can't handle threaded varargs | |
1154 | || ($item_flags =~ /f/ && $item_flags !~ /T/)) | |
1155 | { | |
1156 | $item->{name} = "Perl_$item_name"; | |
1157 | print $fh <<~"EOT"; | |
1158 | ||
1159 | NOTE: C<$item_name> must be explicitly called as | |
1160 | C<$item->{name}> | |
1161 | EOT | |
1162 | print $fh "with an C<aTHX_> parameter" if $item_flags !~ /T/; | |
1163 | print $fh ".\n"; | |
1164 | } | |
1165 | } | |
94bdecf9 | 1166 | |
9e728159 KW |
1167 | if ($flags =~ /[Uy]/) { # no usage; typedefs are considered simple enough |
1168 | # to never warrant a usage line | |
1169 | warn("U and s flags are incompatible") | |
1170 | if $flags =~ /U/ && $flags =~ /s/; | |
20046047 | 1171 | # nothing |
05ca4832 | 1172 | } else { |
783469eb KW |
1173 | |
1174 | print $fh "\n=over $usage_indent\n"; | |
1175 | ||
00d68f0d KW |
1176 | if (defined $docref->{usage}) { # An override of the usage section |
1177 | print $fh "\n", ($docref->{usage} =~ s/^/ /mrg), "\n"; | |
1178 | } | |
783469eb KW |
1179 | else { |
1180 | ||
8d1c4423 KW |
1181 | # Add the thread context formal parameter on expanded-out names |
1182 | for my $item (@items) { | |
1183 | unshift $item->{args}->@*, (($item->{args}->@*) | |
1184 | ? "pTHX_" | |
1185 | : "pTHX") | |
1186 | if $item->{flags} !~ /T/ | |
1187 | && $item->{name} =~ /^Perl_/; | |
1188 | } | |
1189 | ||
783469eb KW |
1190 | # Look through all the items in this entry. If all have the same |
1191 | # return type and arguments, only the main entry is displayed. | |
1192 | # Also, find the longest return type and longest name so that if | |
1193 | # multiple ones are shown, they can be vertically aligned nicely | |
783469eb | 1194 | my $need_individual_usage = 0; |
53c9fbf2 | 1195 | my $longest_name_length = length $items[0]->{name}; |
783469eb | 1196 | my $base_ret_type = $items[0]->{ret_type}; |
53c9fbf2 | 1197 | my $longest_ret = length $base_ret_type; |
783469eb | 1198 | my @base_args = $items[0]->{args}->@*; |
53c9fbf2 | 1199 | for (my $i = 1; $i < @items; $i++) { |
783469eb | 1200 | no warnings 'experimental::smartmatch'; |
53c9fbf2 | 1201 | my $item = $items[$i]; |
783469eb KW |
1202 | $need_individual_usage = 1 |
1203 | if $item->{ret_type} ne $base_ret_type | |
1204 | || ! ($item->{args}->@* ~~ @base_args); | |
1205 | my $ret_length = length $item->{ret_type}; | |
1206 | $longest_ret = $ret_length if $ret_length > $longest_ret; | |
1207 | my $name_length = length $item->{name}; | |
1208 | $longest_name_length = $name_length | |
1209 | if $name_length > $longest_name_length; | |
1ded1f42 | 1210 | } |
783469eb KW |
1211 | |
1212 | # If we're only showing one entry, only its length matters. | |
1213 | $longest_name_length = length($items[0]->{name}) | |
1214 | unless $need_individual_usage; | |
1215 | print $fh "\n"; | |
1216 | ||
1217 | my $indent = 1; # 1 is sufficient for verbatim; =over is used | |
1218 | # for more | |
1219 | my $ret_name_sep_length = 2; # spaces between return type and name | |
1220 | my $name_indent = $indent + $longest_ret; | |
1221 | $name_indent += $ret_name_sep_length if $longest_ret; | |
1222 | ||
1223 | # 80 column terminal - 1 for pager adding a column; -7 for nroff | |
1224 | # indent; | |
1225 | my $max_length = 80 - 1 - 7 - $description_indent - $usage_indent; | |
1226 | ||
1227 | for my $item (@items) { | |
1228 | my $ret_type = $item->{ret_type}; | |
1229 | my @args = $item->{args}->@*; | |
1230 | my $name = $item->{name}; | |
1231 | my $item_flags = $item->{flags}; | |
1232 | ||
783469eb KW |
1233 | # The return type |
1234 | print $fh (" " x $indent), $ret_type; | |
1235 | ||
1236 | print $fh " " x ( $ret_name_sep_length | |
1237 | + $longest_ret - length $ret_type); | |
1238 | print $fh $name; | |
1239 | ||
1240 | if ($item_flags =~ /n/) { # no args | |
1241 | warn("$file: $element_name: n flag without m") | |
1242 | unless $item_flags =~ /m/; | |
1243 | warn("$file: $name: n flag but apparently has args") | |
1244 | if @args; | |
1245 | } | |
1246 | else { | |
1247 | # +1 for the '(' | |
1248 | my $arg_indent = $name_indent + $longest_name_length + 1; | |
1249 | ||
1250 | # Align the argument lists of the items | |
1251 | print $fh " " x ($longest_name_length - length($name)); | |
1252 | print $fh "("; | |
1253 | ||
1254 | # Display as many of the arguments on the same line as | |
1255 | # will fit. | |
1256 | my $total_length = $arg_indent; | |
1257 | my $first_line = 1; | |
1258 | for (my $i = 0; $i < @args; $i++) { | |
1259 | my $arg = $args[$i]; | |
1260 | my $arg_length = length($arg); | |
1261 | ||
1262 | # All but the first arg are preceded by a blank | |
1263 | my $use_blank = $i > 0; | |
1264 | ||
1265 | # +1 here and below because either the argument has a | |
1266 | # trailing comma or trailing ')' | |
1267 | $total_length += $arg_length + $use_blank + 1; | |
1268 | ||
1269 | # We want none of the arguments to be positioned so | |
1270 | # they extend too far to the right. Ideally, they | |
1271 | # should all start in the same column as the arguments | |
1272 | # on the first line of the function display do. But, if | |
1273 | # necessary, outdent them so that they all start in | |
1274 | # another column, with the longest ending at the right | |
1275 | # margin, like so: | |
1276 | # void function_name(pTHX_ short1, | |
1277 | # short2, | |
1278 | # very_long_argument, | |
1279 | # short3) | |
1280 | if ($total_length > $max_length) { | |
1281 | ||
1282 | # If this is the first continuation line, | |
1283 | # calculate the longest argument; this will be the | |
1284 | # one we may have to outdent for. | |
1285 | if ($first_line) { | |
1286 | $first_line = 0; | |
1287 | ||
1288 | # We will need at least as much as the current | |
1289 | # argument | |
1290 | my $longest_arg_length = $arg_length | |
1291 | + $use_blank + 1; | |
1292 | ||
1293 | # Look through the rest of the args to see if | |
1294 | # any are longer than this one. | |
1295 | for (my $j = $i + 1; $j < @args; $j++) { | |
1296 | ||
1297 | # Include the trailing ',' or ')' in the | |
1298 | # length. No need to concern ourselves | |
1299 | # with a leading blank, as the argument | |
1300 | # would be positioned first on the next | |
1301 | # line | |
1302 | my $peek_arg_length = length ($args[$j]) | |
1303 | + 1; | |
1304 | $longest_arg_length = $peek_arg_length | |
1305 | if $peek_arg_length > $longest_arg_length; | |
1306 | } | |
1307 | ||
1308 | # Calculate the new indent if necessary. | |
1309 | $arg_indent = $max_length - $longest_arg_length | |
1310 | if $arg_indent + $longest_arg_length | |
1311 | > $max_length; | |
1312 | } | |
1313 | ||
1314 | print $fh "\n", (" " x $arg_indent); | |
1315 | $total_length = $arg_indent + $arg_length + 1; | |
1316 | $use_blank = 0; | |
1317 | } | |
1318 | ||
1319 | # Display this argument | |
1320 | print $fh " " if $use_blank; | |
1321 | print $fh $arg; | |
1322 | print $fh "," if $i < @args - 1 && $args[$i] ne 'pTHX_'; | |
1323 | ||
1324 | } # End of loop through args | |
1325 | ||
1326 | print $fh ")"; | |
1ded1f42 | 1327 | } |
783469eb KW |
1328 | |
1329 | print $fh ";" if $item_flags =~ /s/; # semicolon: "dTHR;" | |
1330 | print $fh "\n"; | |
1331 | ||
1332 | # Only the first entry is normally displayed | |
1333 | last unless $need_individual_usage; | |
1ded1f42 | 1334 | } |
1ded1f42 | 1335 | } |
783469eb KW |
1336 | |
1337 | print $fh "\n=back\n"; | |
94bdecf9 | 1338 | } |
783469eb KW |
1339 | |
1340 | print $fh "\n=back\n"; | |
63da8140 | 1341 | print $fh "\n=for hackers\nFound in file $file\n"; |
94bdecf9 JH |
1342 | } |
1343 | ||
02c5183e KW |
1344 | sub construct_missings_section { |
1345 | my ($pod_name, $missings_ref) = @_; | |
1346 | my $text = ""; | |
1347 | ||
1348 | return $text unless $missings_ref->@*; | |
1349 | ||
1350 | $text .= <<~EOT; | |
d5c04b1e | 1351 | |
02c5183e KW |
1352 | =head1 Undocumented functions |
1353 | ||
1354 | EOT | |
1355 | if ($pod_name eq 'perlapi') { | |
1356 | $text .= <<~'EOT'; | |
1357 | The following functions have been flagged as part of the public | |
1358 | API, but are currently undocumented. Use them at your own risk, | |
1359 | as the interfaces are subject to change. Functions that are not | |
1360 | listed in this document are not intended for public use, and | |
1361 | should NOT be used under any circumstances. | |
1362 | ||
1363 | If you feel you need to use one of these functions, first send | |
1364 | email to L<perl5-porters@perl.org|mailto:perl5-porters@perl.org>. | |
1365 | It may be that there is a good reason for the function not being | |
1366 | documented, and it should be removed from this list; or it may | |
1367 | just be that no one has gotten around to documenting it. In the | |
1368 | latter case, you will be asked to submit a patch to document the | |
1369 | function. Once your patch is accepted, it will indicate that the | |
1370 | interface is stable (unless it is explicitly marked otherwise) and | |
1371 | usable by you. | |
1372 | EOT | |
1373 | } | |
1374 | else { | |
1375 | $text .= <<~'EOT'; | |
1376 | The following functions are currently undocumented. If you use | |
1377 | one of them, you may wish to consider creating and submitting | |
1378 | documentation for it. | |
1379 | EOT | |
1380 | } | |
1381 | ||
e4f5ddf4 KW |
1382 | $text .= "\n"; |
1383 | ||
1384 | # Sort the elements. | |
1385 | my @missings = sort dictionary_order $missings_ref->@*; | |
1386 | ||
1387 | # Output the X<> references to the names, packed since they don't get | |
1388 | # displayed, but not too many per line so that when someone is editing the | |
1389 | # file, it doesn't run on | |
1390 | my $line_length = 0; | |
1391 | for my $missing (sort dictionary_order @missings) { | |
1392 | my $entry = "X<$missing>"; | |
1393 | my $entry_length = length $entry; | |
1394 | ||
1395 | # Don't loop forever if we have a verrry long name, and don't go too | |
1396 | # far to the right. | |
1397 | if ($line_length > 0 && $line_length + $entry_length > $max_width) { | |
1398 | $text .= "\n"; | |
1399 | $line_length = 0; | |
1400 | } | |
1401 | ||
1402 | $text .= $entry; | |
1403 | $line_length += $entry_length; | |
1404 | } | |
1405 | ||
1406 | use integer; | |
1407 | ||
1408 | # Look through all the elements in the list and see how many columns we | |
1409 | # could place them in the output what will fit in the available width. | |
1410 | my $min_spacer = 2; # Need this much space between columns | |
1411 | my $columns; | |
1412 | my $rows; | |
1413 | my @col_widths; | |
1414 | ||
1415 | COLUMN: | |
1416 | # We start with more columns, and work down until we find a number that | |
1417 | # can accommodate all the data. This algorithm doesn't require the | |
1418 | # resulting columns to all have the same width. This can allow for | |
1419 | # as tight of packing as the data will possibly allow. | |
1420 | for ($columns = 7; $columns > 1; $columns--) { | |
1421 | ||
1422 | # For this many columns, we will need this many rows (final row might | |
1423 | # not be completely filled) | |
1424 | $rows = (@missings + $columns - 1) / $columns; | |
1425 | ||
1426 | my $row_width = 0; | |
1427 | my $i = 0; # Which missing element | |
1428 | ||
1429 | # For each column ... | |
1430 | for my $col (0 .. $columns - 1) { | |
1431 | ||
1432 | # Calculate how wide the column needs to be, which is based on the | |
1433 | # widest element in it | |
1434 | $col_widths[$col] = 0; | |
1435 | ||
1436 | # Look through all the rows to find the widest element | |
1437 | for my $row (0 .. $rows - 1) { | |
1438 | ||
1439 | # Skip if this row doesn't have an entry for this column | |
1440 | last if $i >= @missings; | |
1441 | ||
1442 | # This entry occupies this many bytes. | |
1443 | my $this_width = length $missings[$i]; | |
1444 | ||
1445 | # All but the final column need a spacer between it and the | |
1446 | # next column over. | |
1447 | $this_width += $min_spacer if $col < $columns - 1; | |
1448 | ||
1449 | ||
1450 | # This column will need to have enough width to accommodate | |
1451 | # this element | |
1452 | if ($this_width > $col_widths[$col]) { | |
1453 | ||
1454 | # We can't have this many columns if the total width | |
1455 | # exceeds the available; bail now and try fewer columns | |
1456 | next COLUMN if $row_width + $this_width > $max_width; | |
1457 | ||
1458 | $col_widths[$col] = $this_width; | |
1459 | } | |
1460 | ||
1461 | $i++; # The next row will contain the next item | |
1462 | } | |
1463 | ||
1464 | $row_width += $col_widths[$col]; | |
1465 | next COLUMN if $row_width > $max_width; | |
1466 | } | |
1467 | ||
1468 | # If we get this far, this many columns works | |
1469 | last; | |
1470 | } | |
1471 | ||
1472 | # Here, have calculated the number of rows ($rows) and columns ($columns) | |
1473 | # required to list the elements. @col_widths contains the width of each | |
1474 | # column. | |
1475 | ||
1476 | $text .= "\n\n=over $description_indent\n\n"; | |
1477 | ||
1478 | # Assemble the output | |
1479 | for my $row (0 .. $rows - 1) { | |
1480 | for my $col (0 .. $columns - 1) { | |
1481 | $text .= " " if $col == 0; # Indent one to mark as verbatim | |
1482 | ||
1483 | my $index = $row + $rows * $col; # Convert 2 dimensions to 1 | |
1484 | ||
1485 | # Skip if this row doesn't have an entry for this column | |
1486 | next if $index >= @missings; | |
1487 | ||
1488 | my $element = $missings[$index]; | |
1489 | $text .= $element; | |
1490 | ||
1491 | # Add alignment spaces for all but final column | |
1492 | $text .= " " x ($col_widths[$col] - length $element) | |
1493 | if $col < $columns - 1; | |
1494 | } | |
02c5183e | 1495 | |
e4f5ddf4 | 1496 | $text .= "\n"; # End of row |
02c5183e KW |
1497 | } |
1498 | ||
1499 | $text .= "\n=back\n"; | |
1500 | ||
1501 | return $text; | |
1502 | } | |
1503 | ||
1504 | sub dictionary_order { | |
f83c6033 KW |
1505 | # Do a case-insensitive dictionary sort, with only alphabetics |
1506 | # significant, falling back to using everything for determinancy | |
1354d57e | 1507 | return (uc($a =~ s/[[:^alpha:]]//r) cmp uc($b =~ s/[[:^alpha:]]//r)) |
f83c6033 KW |
1508 | || uc($a) cmp uc($b) |
1509 | || $a cmp $b; | |
1510 | } | |
1511 | ||
7b73ff98 | 1512 | sub output { |
d5c04b1e | 1513 | my ($podname, $header, $dochash, $missings_ref, $footer) = @_; |
6a4c4cd4 DM |
1514 | # |
1515 | # strip leading '|' from each line which had been used to hide | |
1516 | # pod from pod checkers. | |
1517 | s/^\|//gm for $header, $footer; | |
1518 | ||
7882b24a | 1519 | my $fh = open_new("pod/$podname.pod", undef, |
20046047 | 1520 | {by => "$0 extracting documentation", |
f1f44974 | 1521 | from => 'the C source files'}, 1); |
e0492643 | 1522 | |
63da8140 | 1523 | print $fh $header, "\n"; |
e0492643 | 1524 | |
02c5183e | 1525 | for my $section_name (sort dictionary_order keys %valid_sections) { |
7a997878 | 1526 | my $section_info = $dochash->{$section_name}; |
36ea2d2f | 1527 | |
5781dcac KW |
1528 | # We allow empty sections in perlintern. |
1529 | if (! $section_info && $podname eq 'perlapi') { | |
36ea2d2f | 1530 | warn "Empty section '$section_name'; skipped"; |
36ea2d2f KW |
1531 | next; |
1532 | } | |
1533 | ||
63da8140 KW |
1534 | print $fh "\n=head1 $section_name\n"; |
1535 | ||
5781dcac KW |
1536 | if ($podname eq 'perlapi') { |
1537 | print $fh "\n", $valid_sections{$section_name}{header}, "\n" | |
1538 | if defined $valid_sections{$section_name}{header}; | |
151c3fe5 | 1539 | |
5781dcac KW |
1540 | # Output any heading-level documentation and delete so won't get in |
1541 | # the way later | |
1542 | if (exists $section_info->{""}) { | |
1543 | print $fh "\n", $section_info->{""}, "\n"; | |
1544 | delete $section_info->{""}; | |
1545 | } | |
151c3fe5 | 1546 | } |
151c3fe5 | 1547 | |
5781dcac KW |
1548 | |
1549 | if ($section_info) { | |
02c5183e | 1550 | for my $function_name (sort dictionary_order keys %$section_info) { |
5781dcac KW |
1551 | docout($fh, $function_name, $section_info->{$function_name}); |
1552 | } | |
1553 | } | |
1554 | else { | |
1555 | print $fh "\nThere are only public API items currently in $section_name\n"; | |
20046047 | 1556 | } |
00b0fb83 KW |
1557 | |
1558 | print $fh "\n", $valid_sections{$section_name}{footer}, "\n" | |
5781dcac KW |
1559 | if $podname eq 'perlapi' |
1560 | && defined $valid_sections{$section_name}{footer}; | |
7b73ff98 NC |
1561 | } |
1562 | ||
d5c04b1e | 1563 | print $fh construct_missings_section($podname, $missings_ref); |
63da8140 KW |
1564 | |
1565 | print $fh "\n$footer\n=cut\n"; | |
5a0155e6 | 1566 | |
7882b24a | 1567 | read_only_bottom_close_and_rename($fh); |
cd093254 MM |
1568 | } |
1569 | ||
e8e591c9 NC |
1570 | foreach (@{(setup_embed())[0]}) { |
1571 | next if @$_ < 2; | |
b1cb8a49 | 1572 | my ($flags, $ret_type, $func, @args) = @$_; |
e8e591c9 | 1573 | s/\b(?:NN|NULLOK)\b\s+//g for @args; |
bc350081 | 1574 | |
5ce57792 | 1575 | $funcflags{$func} = { |
20046047 | 1576 | flags => $flags, |
b1cb8a49 | 1577 | ret_type => $ret_type, |
20046047 KE |
1578 | args => \@args, |
1579 | }; | |
5ce57792 NC |
1580 | } |
1581 | ||
5ce57792 NC |
1582 | # glob() picks up docs from extra .c or .h files that may be in unclean |
1583 | # development trees. | |
741c0772 NC |
1584 | open my $fh, '<', 'MANIFEST' |
1585 | or die "Can't open MANIFEST: $!"; | |
1586 | while (my $line = <$fh>) { | |
b87d9527 | 1587 | next unless my ($file) = $line =~ /^(\S+\.(?:[ch]|pod))\t/; |
5ce57792 | 1588 | |
7cb0b6c4 KW |
1589 | # Don't pick up pods from these. (We may pick up generated stuff from |
1590 | # /lib though) | |
1591 | next if $file =~ m! ^ ( cpan | dist | ext ) / !x; | |
1592 | ||
1ae6ead9 | 1593 | open F, '<', $file or die "Cannot open $file for docs: $!\n"; |
5ce57792 NC |
1594 | autodoc(\*F,$file); |
1595 | close F or die "Error closing $file: $!\n"; | |
1596 | } | |
741c0772 | 1597 | close $fh or die "Error whilst reading MANIFEST: $!"; |
5ce57792 | 1598 | |
00d68f0d KW |
1599 | parse_config_h(); |
1600 | ||
5ce57792 NC |
1601 | for (sort keys %funcflags) { |
1602 | next unless $funcflags{$_}{flags} =~ /d/; | |
6523e108 | 1603 | next if $funcflags{$_}{flags} =~ /h/; |
573cf2be | 1604 | warn "no docs for $_\n"; |
bc350081 | 1605 | } |
94bdecf9 | 1606 | |
5ce57792 | 1607 | foreach (sort keys %missing) { |
5ce57792 | 1608 | warn "Function '$_', documented in $missing{$_}, not listed in embed.fnc"; |
94bdecf9 JH |
1609 | } |
1610 | ||
ff5af78d KW |
1611 | # List of funcs in the public API that aren't also marked as core-only, |
1612 | # experimental nor deprecated. | |
b87d9527 KW |
1613 | my @missing_api = grep $funcflags{$_}{flags} =~ /A/ |
1614 | && $funcflags{$_}{flags} !~ /[xD]/ | |
1615 | && !$docs{api}{$_}, keys %funcflags; | |
00d68f0d | 1616 | push @missing_api, keys %missing_macros; |
a026e430 | 1617 | |
02c5183e | 1618 | my $other_places = join ", ", map { "L<$_>" } sort dictionary_order qw( perlclib perlxs), |
a026e430 KW |
1619 | keys %described_elsewhere; |
1620 | ||
00d68f0d | 1621 | # The S< > makes things less densely packed, hence more readable |
02c5183e KW |
1622 | my $has_defs_text .= join ",S< > ", map { "C<$_>" } sort dictionary_order @has_defs; |
1623 | my $has_r_defs_text .= join ",S< > ", map { "C<$_>" } sort dictionary_order @has_r_defs; | |
3951f999 KW |
1624 | $valid_sections{$genconfig_scn}{footer} =~ s/__HAS_LIST__/$has_defs_text/; |
1625 | $valid_sections{$genconfig_scn}{footer} =~ s/__HAS_R_LIST__/$has_r_defs_text/; | |
00d68f0d | 1626 | |
02c5183e | 1627 | my $include_defs_text .= join ",S< > ", map { "C<$_>" } sort dictionary_order @include_defs; |
3951f999 | 1628 | $valid_sections{$genconfig_scn}{footer} =~ s/__INCLUDE_LIST__/$include_defs_text/; |
00d68f0d | 1629 | |
02c5183e | 1630 | my $section_list = join "\n\n", map { "=item L</$_>" } sort dictionary_order keys %valid_sections; |
7260d4ba | 1631 | |
b87d9527 | 1632 | output('perlapi', <<"_EOB_", $docs{api}, \@missing_api, <<"_EOE_"); |
6a4c4cd4 DM |
1633 | |=encoding UTF-8 |
1634 | | | |
1635 | |=head1 NAME | |
1636 | | | |
1637 | |perlapi - autogenerated documentation for the perl public API | |
1638 | | | |
1639 | |=head1 DESCRIPTION | |
1640 | |X<Perl API> X<API> X<api> | |
1641 | | | |
b87d9527 KW |
1642 | |This file contains most of the documentation of the perl public API, as |
1643 | |generated by F<embed.pl>. Specifically, it is a listing of functions, | |
a026e430 KW |
1644 | |macros, flags, and variables that may be used by extension writers. Besides |
1645 | |L<perlintern> and F<config.h>, some items are listed here as being actually | |
1646 | |documented in another pod. | |
b87d9527 KW |
1647 | | |
1648 | |L<At the end|/Undocumented functions> is a list of functions which have yet | |
1649 | |to be documented. Patches welcome! The interfaces of these are subject to | |
1650 | |change without notice. | |
1651 | | | |
6927cd44 KW |
1652 | |Some of the functions documented here are consolidated so that a single entry |
1653 | |serves for multiple functions which all do basically the same thing, but have | |
1654 | |some slight differences. For example, one form might process magic, while | |
1655 | |another doesn't. The name of each variation is listed at the top of the | |
1656 | |single entry. But if all have the same signature (arguments and return type) | |
1657 | |except for their names, only the usage for the base form is shown. If any | |
1658 | |one of the forms has a different signature (such as returning C<const> or | |
1659 | |not) every function's signature is explicitly displayed. | |
7260d4ba | 1660 | | |
a026e430 KW |
1661 | |Anything not listed here or in the other mentioned pods is not part of the |
1662 | |public API, and should not be used by extension writers at all. For these | |
1663 | |reasons, blindly using functions listed in F<proto.h> is to be avoided when | |
1664 | |writing extensions. | |
6a4c4cd4 DM |
1665 | | |
1666 | |In Perl, unlike C, a string of characters may generally contain embedded | |
1667 | |C<NUL> characters. Sometimes in the documentation a Perl string is referred | |
1668 | |to as a "buffer" to distinguish it from a C string, but sometimes they are | |
1669 | |both just referred to as strings. | |
1670 | | | |
1671 | |Note that all Perl API global variables must be referenced with the C<PL_> | |
1672 | |prefix. Again, those not listed here are not to be used by extension writers, | |
1673 | |and can be changed or removed without notice; same with macros. | |
1674 | |Some macros are provided for compatibility with the older, | |
1675 | |unadorned names, but this support may be disabled in a future release. | |
1676 | | | |
1677 | |Perl was originally written to handle US-ASCII only (that is characters | |
1678 | |whose ordinal numbers are in the range 0 - 127). | |
1679 | |And documentation and comments may still use the term ASCII, when | |
1680 | |sometimes in fact the entire range from 0 - 255 is meant. | |
1681 | | | |
1682 | |The non-ASCII characters below 256 can have various meanings, depending on | |
1683 | |various things. (See, most notably, L<perllocale>.) But usually the whole | |
1684 | |range can be referred to as ISO-8859-1. Often, the term "Latin-1" (or | |
1685 | |"Latin1") is used as an equivalent for ISO-8859-1. But some people treat | |
1686 | |"Latin1" as referring just to the characters in the range 128 through 255, or | |
2de859ce | 1687 | |sometimes from 160 through 255. |
6a4c4cd4 DM |
1688 | |This documentation uses "Latin1" and "Latin-1" to refer to all 256 characters. |
1689 | | | |
1690 | |Note that Perl can be compiled and run under either ASCII or EBCDIC (See | |
1691 | |L<perlebcdic>). Most of the documentation (and even comments in the code) | |
1692 | |ignore the EBCDIC possibility. | |
1693 | |For almost all purposes the differences are transparent. | |
1694 | |As an example, under EBCDIC, | |
1695 | |instead of UTF-8, UTF-EBCDIC is used to encode Unicode strings, and so | |
1696 | |whenever this documentation refers to C<utf8> | |
1697 | |(and variants of that name, including in function names), | |
1698 | |it also (essentially transparently) means C<UTF-EBCDIC>. | |
1699 | |But the ordinals of characters differ between ASCII, EBCDIC, and | |
1700 | |the UTF- encodings, and a string encoded in UTF-EBCDIC may occupy a different | |
1701 | |number of bytes than in UTF-8. | |
1702 | | | |
6927cd44 KW |
1703 | |The organization of this document is tentative and subject to change. |
1704 | |Suggestions and patches welcome | |
1705 | |L<perl5-porters\@perl.org|mailto:perl5-porters\@perl.org>. | |
1706 | | | |
1707 | |The sections in this document currently are | |
1708 | | | |
1709 | |=over | |
1710 | ||
1711 | |$section_list | |
1712 | | | |
1713 | |=back | |
1714 | | | |
6a4c4cd4 | 1715 | |The listing below is alphabetical, case insensitive. |
94bdecf9 | 1716 | _EOB_ |
6a4c4cd4 DM |
1717 | |=head1 AUTHORS |
1718 | | | |
1719 | |Until May 1997, this document was maintained by Jeff Okamoto | |
7b1f0a98 | 1720 | |<okamoto\@corp.hp.com>. It is now maintained as part of Perl itself. |
6a4c4cd4 DM |
1721 | | |
1722 | |With lots of help and suggestions from Dean Roehrich, Malcolm Beattie, | |
1723 | |Andreas Koenig, Paul Hudson, Ilya Zakharevich, Paul Marquess, Neil | |
1724 | |Bowers, Matthew Green, Tim Bunce, Spider Boardman, Ulrich Pfeifer, | |
1725 | |Stephen McCamant, and Gurusamy Sarathy. | |
1726 | | | |
7b1f0a98 | 1727 | |API Listing originally by Dean Roehrich <roehrich\@cray.com>. |
6a4c4cd4 DM |
1728 | | |
1729 | |Updated to be autogenerated from comments in the source by Benjamin Stuhl. | |
1730 | | | |
1731 | |=head1 SEE ALSO | |
1732 | | | |
a026e430 | 1733 | |F<config.h>, L<perlintern>, $other_places |
94bdecf9 JH |
1734 | _EOE_ |
1735 | ||
79fc8511 FC |
1736 | # List of non-static internal functions |
1737 | my @missing_guts = | |
9f589e47 | 1738 | grep $funcflags{$_}{flags} !~ /[AS]/ && !$docs{guts}{$_}, keys %funcflags; |
5a0155e6 | 1739 | |
7b1f0a98 | 1740 | output('perlintern', <<'_EOB_', $docs{guts}, \@missing_guts, <<"_EOE_"); |
6a4c4cd4 DM |
1741 | |=head1 NAME |
1742 | | | |
1743 | |perlintern - autogenerated documentation of purely B<internal> | |
20046047 | 1744 | |Perl functions |
6a4c4cd4 DM |
1745 | | |
1746 | |=head1 DESCRIPTION | |
1747 | |X<internal Perl functions> X<interpreter functions> | |
1748 | | | |
1749 | |This file is the autogenerated documentation of functions in the | |
1750 | |Perl interpreter that are documented using Perl's internal documentation | |
1751 | |format but are not marked as part of the Perl API. In other words, | |
1752 | |B<they are not for use in extensions>! | |
5781dcac KW |
1753 | |
1754 | |It has the same sections as L<perlapi>, though some may be empty. | |
6a4c4cd4 | 1755 | | |
7b1f0a98 | 1756 | _EOB_ |
6a4c4cd4 DM |
1757 | | |
1758 | |=head1 AUTHORS | |
1759 | | | |
1760 | |The autodocumentation system was originally added to the Perl core by | |
1761 | |Benjamin Stuhl. Documentation is by whoever was kind enough to | |
1762 | |document their functions. | |
1763 | | | |
1764 | |=head1 SEE ALSO | |
1765 | | | |
a026e430 | 1766 | |F<config.h>, L<perlapi>, $other_places |
7b1f0a98 | 1767 | _EOE_ |