3 This document is in Pod format. To read this, use a Pod formatter,
4 like "perldoc perlpod".
7 X<POD> X<plain old documentation>
9 perlpod - the Plain Old Documentation format
13 Pod is a simple-to-use markup language used for writing documentation
14 for Perl, Perl programs, and Perl modules.
16 Translators are available for converting Pod to various formats
17 like plain text, HTML, man pages, and more.
19 Pod markup consists of three basic kinds of paragraphs:
20 L<ordinary|/"Ordinary Paragraph">,
21 L<verbatim|/"Verbatim Paragraph">, and
22 L<command|/"Command Paragraph">.
25 =head2 Ordinary Paragraph
26 X<POD, ordinary paragraph>
28 Most paragraphs in your documentation will be ordinary blocks
29 of text, like this one. You can simply type in your text without
30 any markup whatsoever, and with just a blank line before and
31 after. When it gets formatted, it will undergo minimal formatting,
32 like being rewrapped, probably put into a proportionally spaced
33 font, and maybe even justified.
35 You can use formatting codes in ordinary paragraphs, for B<bold>,
36 I<italic>, C<code-style>, L<hyperlinks|perlfaq>, and more. Such
37 codes are explained in the "L<Formatting Codes|/"Formatting Codes">"
41 =head2 Verbatim Paragraph
42 X<POD, verbatim paragraph> X<verbatim>
44 Verbatim paragraphs are usually used for presenting a codeblock or
45 other text which does not require any special parsing or formatting,
46 and which shouldn't be wrapped.
48 A verbatim paragraph is distinguished by having its first character
49 be a space or a tab. (And commonly, all its lines begin with spaces
50 and/or tabs.) It should be reproduced exactly, with tabs assumed to
51 be on 8-column boundaries. There are no special formatting codes,
52 so you can't italicize or anything like that. A \ means \, and
56 =head2 Command Paragraph
59 A command paragraph is used for special treatment of whole chunks
60 of text, usually as headings or parts of lists.
62 All command paragraphs (which are typically only one line long) start
63 with "=", followed by an identifier, followed by arbitrary text that
64 the command can use however it pleases. Currently recognized commands
83 To explain them each in detail:
87 =item C<=head1 I<Heading Text>>
88 X<=head1> X<=head2> X<=head3> X<=head4> X<=head5> X<=head6>
89 X<head1> X<head2> X<head3> X<head4> X<head5> X<head6>
91 =item C<=head2 I<Heading Text>>
93 =item C<=head3 I<Heading Text>>
95 =item C<=head4 I<Heading Text>>
97 =item C<=head5 I<Heading Text>>
99 =item C<=head6 I<Heading Text>>
101 Head1 through head6 produce headings, head1 being the highest
102 level. The text in the rest of this paragraph is the content of the
103 heading. For example:
105 =head2 Object Attributes
107 The text "Object Attributes" comprises the heading there.
108 The text in these heading commands can use formatting codes, as seen here:
110 =head2 Possible Values for C<$/>
112 Such commands are explained in the
113 "L<Formatting Codes|/"Formatting Codes">" section, below.
115 Note that C<head5> and C<head6> were introduced in 2020 and in
116 L<Pod::Simple> 3.41, released in October 2020, so they might not be
117 supported on the Pod parser you use.
119 =item C<=over I<indentlevel>>
120 X<=over> X<=item> X<=back> X<over> X<item> X<back>
122 =item C<=item I<stuff...>>
126 Item, over, and back require a little more explanation: "=over" starts
127 a region specifically for the generation of a list using "=item"
128 commands, or for indenting (groups of) normal paragraphs. At the end
129 of your list, use "=back" to end it. The I<indentlevel> option to
130 "=over" indicates how far over to indent, generally in ems (where
131 one em is the width of an "M" in the document's base font) or roughly
132 comparable units; if there is no I<indentlevel> option, it defaults
133 to four. (And some formatters may just ignore whatever I<indentlevel>
134 you provide.) In the I<stuff> in C<=item I<stuff...>>, you may
135 use formatting codes, as seen here:
137 =item Using C<$|> to Control Buffering
139 Such commands are explained in the
140 "L<Formatting Codes|/"Formatting Codes">" section, below.
142 Note also that there are some basic rules to using "=over" ...
149 Don't use "=item"s outside of an "=over" ... "=back" region.
153 The first thing after the "=over" command should be an "=item", unless
154 there aren't going to be any items at all in this "=over" ... "=back"
159 Don't put "=headI<n>" commands inside an "=over" ... "=back" region.
163 And perhaps most importantly, keep the items consistent: either use
164 "=item *" for all of them, to produce bullets; or use "=item 1.",
165 "=item 2.", etc., to produce numbered lists; or use "=item foo",
166 "=item bar", etc.--namely, things that look nothing like bullets or
167 numbers. (If you have a list that contains both: 1) things that don't
168 look like bullets nor numbers, plus 2) things that do, you should
169 preface the bullet- or number-like items with C<ZE<lt>E<gt>>. See
170 L<ZE<lt>E<gt>|/ZE<lt>E<gt> -- a null (zero-effect) formatting code>
171 below for an example.)
173 If you start with bullets or numbers, stick with them, as
174 formatters use the first "=item" type to decide how to format the
182 To end a Pod block, use a blank line,
183 then a line beginning with "=cut", and a blank
184 line after it. This lets Perl (and the Pod formatter) know that
185 this is where Perl code is resuming. (The blank line before the "=cut"
186 is not technically necessary, but many older Pod processors require it.)
191 The "=pod" command by itself doesn't do much of anything, but it
192 signals to Perl (and Pod formatters) that a Pod block starts here. A
193 Pod block starts with I<any> command paragraph, so a "=pod" command is
194 usually used just when you want to start a Pod block with an ordinary
195 paragraph or a verbatim paragraph. For example:
199 This function does stuff.
209 Remember to check its return value, as in:
211 stuff() || die "Couldn't do stuff!";
215 =item C<=begin I<formatname>>
216 X<=begin> X<=end> X<=for> X<begin> X<end> X<for>
218 =item C<=end I<formatname>>
220 =item C<=for I<formatname> I<text...>>
222 For, begin, and end will let you have regions of text/code/data that
223 are not generally interpreted as normal Pod text, but are passed
224 directly to particular formatters, or are otherwise special. A
225 formatter that can use that format will use the region, otherwise it
226 will be completely ignored.
228 A command "=begin I<formatname>", some paragraphs, and a
229 command "=end I<formatname>", mean that the text/data in between
230 is meant for formatters that understand the special format
231 called I<formatname>. For example,
235 <hr> <img src="thang.png">
236 <p> This is a raw HTML paragraph </p>
240 The command "=for I<formatname> I<text...>"
241 specifies that the remainder of just this paragraph (starting
242 right after I<formatname>) is in that special format.
244 =for html <hr> <img src="thang.png">
245 <p> This is a raw HTML paragraph </p>
247 This means the same thing as the above "=begin html" ... "=end html"
250 That is, with "=for", you can have only one paragraph's worth
251 of text (i.e., the text in "=foo targetname text..."), but with
252 "=begin targetname" ... "=end targetname", you can have any amount
253 of stuff in between. (Note that there still must be a blank line
254 after the "=begin" command and a blank line before the "=end"
257 Here are some examples of how to use these:
261 <br>Figure 1.<br><IMG SRC="figure1.png"><br>
276 Some format names that formatters currently are known to accept
277 include "roff", "man", "latex", "tex", "text", and "html". (Some
278 formatters will treat some of these as synonyms.)
280 A format name of "comment" is common for just making notes (presumably
281 to yourself) that won't appear in any formatted version of the Pod
285 Make sure that all the available options are documented!
287 Some I<formatnames> will require a leading colon (as in
288 C<"=for :formatname">, or
289 C<"=begin :formatname" ... "=end :formatname">),
290 to signal that the text is not raw data, but instead I<is> Pod text
291 (i.e., possibly containing formatting codes) that's just not for
292 normal formatting (e.g., may not be a normal-use paragraph, but might
293 be for formatting as a footnote).
295 =item C<=encoding I<encodingname>>
296 X<=encoding> X<encoding>
298 This command is used for declaring the encoding of a document. Most
299 users won't need this; but if your encoding isn't US-ASCII,
300 then put a C<=encoding I<encodingname>> command very early in the document so
301 that pod formatters will know how to decode the document. For
302 I<encodingname>, use a name recognized by the L<Encode::Supported>
303 module. Some pod formatters may try to guess between a Latin-1 or
305 UTF-8 encoding, but they may guess wrong. It's best to be explicit if
306 you use anything besides strict ASCII. Examples:
318 C<=encoding> affects the whole document, and must occur only once.
322 And don't forget, all commands but C<=encoding> last up
323 until the end of its I<paragraph>, not its line. So in the
324 examples below, you can see that every command needs the blank
325 line after it, to end its paragraph. (And some older Pod translators
326 may require the C<=encoding> line to have a following blank line as
327 well, even though it should be legal to omit.)
329 Some examples of lists include:
347 Description of Foo function
351 Description of Bar function
356 =head2 Formatting Codes
357 X<POD, formatting code> X<formatting code>
358 X<POD, interior sequence> X<interior sequence>
360 In ordinary paragraphs and in some command paragraphs, various
361 formatting codes (a.k.a. "interior sequences") can be used:
364 "interior sequences" is such an opaque term.
365 Prefer "formatting codes" instead.
369 =item C<IE<lt>textE<gt>> -- italic text
370 X<I> X<< IZ<><> >> X<POD, formatting code, italic> X<italic>
372 Used for emphasis ("C<be IE<lt>careful!E<gt>>") and parameters
373 ("C<redo IE<lt>LABELE<gt>>")
375 =item C<BE<lt>textE<gt>> -- bold text
376 X<B> X<< BZ<><> >> X<POD, formatting code, bold> X<bold>
378 Used for switches ("C<perl's BE<lt>-nE<gt> switch>"), programs
379 ("C<some systems provide a BE<lt>chfnE<gt> for that>"),
380 emphasis ("C<be BE<lt>careful!E<gt>>"), and so on
381 ("C<and that feature is known as BE<lt>autovivificationE<gt>>").
383 =item C<CE<lt>codeE<gt>> -- code text
384 X<C> X<< CZ<><> >> X<POD, formatting code, code> X<code>
386 Renders code in a typewriter font, or gives some other indication that
387 this represents program text ("C<CE<lt>gmtime($^T)E<gt>>") or some other
388 form of computerese ("C<CE<lt>drwxr-xr-xE<gt>>").
390 =item C<LE<lt>nameE<gt>> -- a hyperlink
391 X<L> X<< LZ<><> >> X<POD, formatting code, hyperlink> X<hyperlink>
393 There are various syntaxes, listed below. In the syntaxes given,
394 C<text>, C<name>, and C<section> cannot contain the characters
395 '/' and '|'; and any '<' or '>' should be matched.
403 Link to a Perl manual page (e.g., C<LE<lt>Net::PingE<gt>>). Note
404 that C<name> should not contain spaces. This syntax
405 is also occasionally used for references to Unix man pages, as in
406 C<LE<lt>crontab(5)E<gt>>.
410 C<LE<lt>name/"sec"E<gt>> or C<LE<lt>name/secE<gt>>
412 Link to a section in other manual page. E.g.,
413 C<LE<lt>perlsyn/"For Loops"E<gt>>
417 C<LE<lt>/"sec"E<gt>> or C<LE<lt>/secE<gt>>
419 Link to a section in this manual page. E.g.,
420 C<LE<lt>/"Object Methods"E<gt>>
424 A section is started by the named heading or item. For
425 example, C<LE<lt>perlvar/$.E<gt>> or C<LE<lt>perlvar/"$."E<gt>> both
426 link to the section started by "C<=item $.>" in perlvar. And
427 C<LE<lt>perlsyn/For LoopsE<gt>> or C<LE<lt>perlsyn/"For Loops"E<gt>>
428 both link to the section started by "C<=head2 For Loops>"
431 To control what text is used for display, you
432 use "C<LE<lt>text|...E<gt>>", as in:
438 C<LE<lt>text|nameE<gt>>
440 Link this text to that manual page. E.g.,
441 C<LE<lt>Perl Error Messages|perldiagE<gt>>
445 C<LE<lt>text|name/"sec"E<gt>> or C<LE<lt>text|name/secE<gt>>
447 Link this text to that section in that manual page. E.g.,
448 C<LE<lt>postfix "if"|perlsyn/"Statement Modifiers"E<gt>>
452 C<LE<lt>text|/"sec"E<gt>> or C<LE<lt>text|/secE<gt>>
453 or C<LE<lt>text|"sec"E<gt>>
455 Link this text to that section in this manual page. E.g.,
456 C<LE<lt>the various attributes|/"Member Data"E<gt>>
460 Or you can link to a web page:
466 C<LE<lt>scheme:...E<gt>>
468 C<LE<lt>text|scheme:...E<gt>>
470 Links to an absolute URL. For example, C<LE<lt>http://www.perl.org/E<gt>> or
471 C<LE<lt>The Perl Home Page|http://www.perl.org/E<gt>>.
475 =item C<EE<lt>escapeE<gt>> -- a character escape
476 X<E> X<< EZ<><> >> X<POD, formatting code, escape> X<escape>
478 Very similar to HTML/XML C<&I<foo>;> "entity references":
484 C<EE<lt>ltE<gt>> -- a literal E<lt> (less than)
488 C<EE<lt>gtE<gt>> -- a literal E<gt> (greater than)
492 C<EE<lt>verbarE<gt>> -- a literal | (I<ver>tical I<bar>)
496 C<EE<lt>solE<gt>> -- a literal / (I<sol>idus)
498 The above four are optional except in other formatting codes,
499 notably C<LE<lt>...E<gt>>, and when preceded by a
504 C<EE<lt>htmlnameE<gt>>
506 Some non-numeric HTML entity name, such as C<EE<lt>eacuteE<gt>>,
507 meaning the same thing as C<é> in HTML -- i.e., a lowercase
508 e with an acute (/-shaped) accent.
514 The ASCII/Latin-1/Unicode character with that number. A
515 leading "0x" means that I<number> is hex, as in
516 C<EE<lt>0x201EE<gt>>. A leading "0" means that I<number> is octal,
517 as in C<EE<lt>075E<gt>>. Otherwise I<number> is interpreted as being
518 in decimal, as in C<EE<lt>181E<gt>>.
520 Note that older Pod formatters might not recognize octal or
521 hex numeric escapes, and that many formatters cannot reliably
522 render characters above 255. (Some formatters may even have
523 to use compromised renderings of Latin-1/CP-1252 characters, like
524 rendering C<EE<lt>eacuteE<gt>> as just a plain "e".)
528 =item C<FE<lt>filenameE<gt>> -- used for filenames
529 X<F> X<< FZ<><> >> X<POD, formatting code, filename> X<filename>
531 Typically displayed in italics. Example: "C<FE<lt>.cshrcE<gt>>"
533 =item C<SE<lt>textE<gt>> -- text contains non-breaking spaces
534 X<S> X<< SZ<><> >> X<POD, formatting code, non-breaking space>
535 X<non-breaking space>
537 This means that the words in I<text> should not be broken
538 across lines. Example: S<C<SE<lt>$x ? $y : $zE<gt>>>.
540 =item C<XE<lt>topic nameE<gt>> -- an index entry
541 X<X> X<< XZ<><> >> X<POD, formatting code, index entry> X<index entry>
543 This is ignored by most formatters, but some may use it for building
544 indexes. It always renders as empty-string.
545 Example: C<XE<lt>absolutizing relative URLsE<gt>>
547 =item C<ZE<lt>E<gt>> -- a null (zero-effect) formatting code
548 X<Z> X<< ZZ<><> >> X<POD, formatting code, null> X<null>
550 This is rarely used. It's one way to get around using an
551 EE<lt>...E<gt> code sometimes. For example, instead of
552 "C<NEE<lt>ltE<gt>3>" (for "NE<lt>3") you could write
553 "C<NZE<lt>E<gt>E<lt>3>" (the "ZE<lt>E<gt>" breaks up the "N" and
554 the "E<lt>" so they can't be considered
555 the part of a (fictitious) "NE<lt>...E<gt>" code).
557 Another use is to indicate that I<stuff> in C<=item ZE<lt>E<gt>I<stuff...>>
558 is not to be considered to be a bullet or number. For example,
559 without the C<ZE<lt>E<gt>>, the line
561 =item Z<>500 Server error
563 could possibly be parsed as an item in a numbered list when it isn't
566 Still another use is to maintain visual space between C<=item> lines.
573 it will typically get rendered as
578 That may be what you want, but if what you really want is
584 you can use C<ZE<lt>E<gt>> to accomplish that
593 This was formerly explained as a "zero-width character". But it in
594 most parser models, it parses to nothing at all, as opposed to parsing
595 as if it were a E<zwnj> or E<zwj>, which are REAL zero-width characters.
596 So "width" and "character" are exactly the wrong words.
600 Most of the time, you will need only a single set of angle brackets to
601 delimit the beginning and end of formatting codes. However,
602 sometimes you will want to put a real right angle bracket (a
603 greater-than sign, '>') inside of a formatting code. This is particularly
604 common when using a formatting code to provide a different font-type for a
605 snippet of code. As with all things in Perl, there is more than
606 one way to do it. One way is to simply escape the closing bracket
611 This will produce: "C<$a E<lt>=E<gt> $b>"
613 A more readable, and perhaps more "plain" way is to use an alternate
614 set of delimiters that doesn't require a single ">" to be escaped.
615 Doubled angle brackets ("<<" and ">>") may be used I<if and only if there is
616 whitespace right after the opening delimiter and whitespace right
617 before the closing delimiter!> For example, the following will
619 X<POD, formatting code, escaping with multiple brackets>
623 In fact, you can use as many repeated angle-brackets as you like so
624 long as you have the same number of them in the opening and closing
625 delimiters, and make sure that whitespace immediately follows the last
626 '<' of the opening delimiter, and immediately precedes the first '>'
627 of the closing delimiter. (The whitespace is ignored.) So the
628 following will also work:
629 X<POD, formatting code, escaping with multiple brackets>
634 And they all mean exactly the same as this:
638 The multiple-bracket form does not affect the interpretation of the contents of
639 the formatting code, only how it must end. That means that the examples above
640 are also exactly the same as this:
642 C<< $a E<lt>=E<gt> $b >>
644 As a further example, this means that if you wanted to put these bits of
645 code in C<C> (code) style:
647 open(X, ">>thing.dat") || die $!
650 you could do it like so:
652 C<<< open(X, ">>thing.dat") || die $! >>>
655 which is presumably easier to read than the old way:
657 C<open(X, "E<gt>E<gt>thing.dat") || die $!>
660 This is currently supported by pod2text (Pod::Text), pod2man (Pod::Man),
661 and any other pod2xxx or Pod::Xxxx translators that use
662 Pod::Parser 1.093 or later, or Pod::Tree 1.02 or later.
667 The intent is simplicity of use, not power of expression. Paragraphs
668 look like paragraphs (block format), so that they stand out
669 visually, and so that I could run them through C<fmt> easily to reformat
670 them (that's F7 in my version of B<vi>, or Esc Q in my version of
671 B<emacs>). I wanted the translator to always leave the C<'> and C<`> and
672 C<"> quotes alone, in verbatim mode, so I could slurp in a
673 working program, shift it over four spaces, and have it print out, er,
674 verbatim. And presumably in a monospace font.
676 The Pod format is not necessarily sufficient for writing a book. Pod
677 is just meant to be an idiot-proof common source for nroff, HTML,
678 TeX, and other markup languages, as used for online
679 documentation. Translators exist for B<pod2text>, B<pod2html>,
680 B<pod2man> (that's for nroff(1) and troff(1)), B<pod2latex>, and
681 B<pod2fm>. Various others are available in CPAN.
684 =head2 Embedding Pods in Perl Modules
687 You can embed Pod documentation in your Perl modules and scripts. Start
688 your documentation with an empty line, a "=head1" command at the
689 beginning, and end it with a "=cut" command and an empty line. The
690 B<perl> executable will ignore the Pod text. You can place a Pod
691 statement where B<perl> expects the beginning of a new statement, but
692 not within a statement, as that would result in an error. See any of
693 the supplied library modules for examples.
695 If you're going to put your Pod at the end of the file, and you're using
696 an C<__END__> or C<__DATA__> cut mark, make sure to put an empty line there
697 before the first Pod command.
703 Time::Local - efficiently compute time from local and GMT time
705 Without that empty line before the "=head1", many translators wouldn't
706 have recognized the "=head1" as starting a Pod block.
708 =head2 Hints for Writing Pod
713 X<podchecker> X<POD, validating>
715 The B<podchecker> command is provided for checking Pod syntax for errors
716 and warnings. For example, it checks for completely blank lines in
717 Pod blocks and for unknown commands and formatting codes. You should
718 still also pass your document through one or more translators and proofread
719 the result, or print out the result and proofread that. Some of the
720 problems found may be bugs in the translators, which you may or may not
725 If you're more familiar with writing in HTML than with writing in Pod, you
726 can try your hand at writing documentation in simple HTML, and converting
727 it to Pod with the experimental L<Pod::HTML2Pod|Pod::HTML2Pod> module,
728 (available in CPAN), and looking at the resulting code. The experimental
729 L<Pod::PXML|Pod::PXML> module in CPAN might also be useful.
733 Many older Pod translators require the lines before every Pod
734 command and after every Pod command (including "=cut"!) to be a blank
735 line. Having something like this:
737 # - - - - - - - - - - - -
738 =item $firecracker->boom()
740 This noisily detonates the firecracker object.
745 ...will make such Pod translators completely fail to see the Pod block
748 Instead, have it like this:
750 # - - - - - - - - - - - -
752 =item $firecracker->boom()
754 This noisily detonates the firecracker object.
763 Some older Pod translators require paragraphs (including command
764 paragraphs like "=head2 Functions") to be separated by I<completely>
765 empty lines. If you have an apparently empty line with some spaces
766 on it, this might not count as a separator for those translators, and
767 that could cause odd formatting.
771 Older translators might add wording around an LE<lt>E<gt> link, so that
772 C<LE<lt>Foo::BarE<gt>> may become "the Foo::Bar manpage", for example.
773 So you shouldn't write things like C<the LE<lt>fooE<gt>
774 documentation>, if you want the translated document to read sensibly.
775 Instead, write C<the LE<lt>Foo::Bar|Foo::BarE<gt> documentation> or
776 C<LE<lt>the Foo::Bar documentation|Foo::BarE<gt>>, to control how the
781 Going past the 70th column in a verbatim block might be ungracefully
782 wrapped by some formatters.
788 L<perlpodspec>, L<perlsyn/"PODs: Embedded Documentation">,
789 L<perlnewmod>, L<perldoc>, L<pod2html>, L<pod2man>, L<podchecker>.
793 Larry Wall, Sean M. Burke