6 my $arg_string = undef;
8 # give Macs a chance to provide command line parameters
11 MacPerl::Ask( 'Please enter @ARGV (-h for help)',
12 defined $ARGV[0] ? "\"$ARGV[0]\"" : "" );
15 Perl::Tidy::perltidy(argv => $arg_string);
21 perltidy - a perl script indenter and reformatter
25 perltidy [ options ] file1 file2 file3 ...
26 (output goes to file1.tdy, file2.tdy, file3.tdy, ...)
27 perltidy [ options ] file1 -o outfile
28 perltidy [ options ] file1 -st >outfile
29 perltidy [ options ] <infile >outfile
33 Perltidy reads a perl script and writes an indented, reformatted script.
35 Many users will find enough information in L<"EXAMPLES"> to get
36 started. New users may benefit from the short tutorial
38 http://perltidy.sourceforge.net/tutorial.html
40 A convenient aid to systematically defining a set of style parameters
42 http://perltidy.sourceforge.net/stylekey.html
44 Perltidy can produce output on either of two modes, depending on the
45 existence of an B<-html> flag. Without this flag, the output is passed
46 through a formatter. The default formatting tries to follow the
47 recommendations in perlstyle(1), but it can be controlled in detail with
48 numerous input parameters, which are described in L<"FORMATTING
51 When the B<-html> flag is given, the output is passed through an HTML
52 formatter which is described in L<"HTML OPTIONS">.
58 This will produce a file F<somefile.pl.tdy> containing the script reformatted
59 using the default options, which approximate the style suggested in
60 perlstyle(1). Perltidy never changes the input file.
64 Execute perltidy on all F<.pl> files in the current directory with the
65 default options. The output will be in files with an appended F<.tdy>
66 extension. For any file with an error, there will be a file with extension
69 perltidy -b file1.pl file2.pl
71 Modify F<file1.pl> and F<file1.pl> in place, and backup the originals to
72 F<file1.pl.bak> and F<file2.pl.bak>. If F<file1.pl.bak> and/or F<file2.pl.bak>
73 already exist, they will be overwritten.
75 perltidy -gnu somefile.pl
77 Execute perltidy on file F<somefile.pl> with a style which approximates the
78 GNU Coding Standards for C programs. The output will be F<somefile.pl.tdy>.
80 perltidy -i=3 somefile.pl
82 Execute perltidy on file F<somefile.pl>, with 3 columns for each level of
83 indentation (B<-i=3>) instead of the default 4 columns. There will not be any
84 tabs in the reformatted script, except for any which already exist in comments,
85 pod documents, quotes, and here documents. Output will be F<somefile.pl.tdy>.
87 perltidy -i=3 -et=8 somefile.pl
89 Same as the previous example, except that leading whitespace will
90 be entabbed with one tab character per 8 spaces.
92 perltidy -ce -l=72 somefile.pl
94 Execute perltidy on file F<somefile.pl> with all defaults except use "cuddled
95 elses" (B<-ce>) and a maximum line length of 72 columns (B<-l=72>) instead of
96 the default 80 columns.
98 perltidy -g somefile.pl
100 Execute perltidy on file F<somefile.pl> and save a log file F<somefile.pl.LOG>
101 which shows the nesting of braces, parentheses, and square brackets at
102 the start of every line.
104 perltidy -html somefile.pl
106 This will produce a file F<somefile.pl.html> containing the script with
107 html markup. The output file will contain an embedded style sheet in
108 the <HEAD> section which may be edited to change the appearance.
110 perltidy -html -css=mystyle.css somefile.pl
112 This will produce a file F<somefile.pl.html> containing the script with
113 html markup. This output file will contain a link to a separate style
114 sheet file F<mystyle.css>. If the file F<mystyle.css> does not exist,
115 it will be created. If it exists, it will not be overwritten.
117 perltidy -html -pre somefile.pl
119 Write an html snippet with only the PRE section to F<somefile.pl.html>.
120 This is useful when code snippets are being formatted for inclusion in a
121 larger web page. No style sheet will be written in this case.
123 perltidy -html -ss >mystyle.css
125 Write a style sheet to F<mystyle.css> and exit.
127 perltidy -html -frm mymodule.pm
129 Write html with a frame holding a table of contents and the source code. The
130 output files will be F<mymodule.pm.html> (the frame), F<mymodule.pm.toc.html>
131 (the table of contents), and F<mymodule.pm.src.html> (the source code).
133 =head1 OPTIONS - OVERVIEW
135 The entire command line is scanned for options, and they are processed
136 before any files are processed. As a result, it does not matter
137 whether flags are before or after any filenames. However, the relative
138 order of parameters is important, with later parameters overriding the
139 values of earlier parameters.
141 For each parameter, there is a long name and a short name. The short
142 names are convenient for keyboard input, while the long names are
143 self-documenting and therefore useful in scripts. It is customary to
144 use two leading dashes for long names, but one may be used.
146 Most parameters which serve as on/off flags can be negated with a
147 leading "n" (for the short name) or a leading "no" or "no-" (for the
148 long name). For example, the flag to outdent long quotes is is B<-olq>
149 or B<--outdent-long-quotes>. The flag to skip this is B<-nolq>
150 or B<--nooutdent-long-quotes> or B<--no-outdent-long-quotes>.
152 Options may not be bundled together. In other words, options B<-q> and
153 B<-g> may NOT be entered as B<-qg>.
155 Option names may be terminated early as long as they are uniquely identified.
156 For example, instead of B<--dump-token-types>, it would be sufficient to enter
157 B<--dump-tok>, or even B<--dump-t>, to uniquely identify this command.
161 The following parameters concern the files which are read and written.
165 =item B<-h>, B<--help>
167 Show summary of usage and exit.
169 =item B<-o>=filename, B<--outfile>=filename
171 Name of the output file (only if a single input file is being
172 processed). If no output file is specified, and output is not
173 redirected to the standard output (see B<-st>), the output will go to
174 F<filename.tdy>. [Note: - does not redirect to standard output. Use
177 =item B<-st>, B<--standard-output>
179 Perltidy must be able to operate on an arbitrarily large number of files
180 in a single run, with each output being directed to a different output
181 file. Obviously this would conflict with outputting to the single
182 standard output device, so a special flag, B<-st>, is required to
183 request outputting to the standard output. For example,
185 perltidy somefile.pl -st >somefile.new.pl
187 This option may only be used if there is just a single input file.
188 The default is B<-nst> or B<--nostandard-output>.
190 =item B<-se>, B<--standard-error-output>
192 If perltidy detects an error when processing file F<somefile.pl>, its
193 default behavior is to write error messages to file F<somefile.pl.ERR>.
194 Use B<-se> to cause all error messages to be sent to the standard error
195 output stream instead. This directive may be negated with B<-nse>.
196 Thus, you may place B<-se> in a F<.perltidyrc> and override it when
197 desired with B<-nse> on the command line.
199 =item B<-oext>=ext, B<--output-file-extension>=ext
201 Change the extension of the output file to be F<ext> instead of the
202 default F<tdy> (or F<html> in case the -B<-html> option is used).
203 See L<Specifying File Extensions>.
205 =item B<-opath>=path, B<--output-path>=path
207 When perltidy creates a filename for an output file, by default it merely
208 appends an extension to the path and basename of the input file. This
209 parameter causes the path to be changed to F<path> instead.
211 The path should end in a valid path separator character, but perltidy will try
212 to add one if it is missing.
216 perltidy somefile.pl -opath=/tmp/
218 will produce F</tmp/somefile.pl.tdy>. Otherwise, F<somefile.pl.tdy> will
219 appear in whatever directory contains F<somefile.pl>.
221 If the path contains spaces, it should be placed in quotes.
223 This parameter will be ignored if output is being directed to standard output,
224 or if it is being specified explicitly with the B<-o=s> parameter.
226 =item B<-b>, B<--backup-and-modify-in-place>
228 Modify the input file or files in-place and save the original with the
229 extension F<.bak>. Any existing F<.bak> file will be deleted. See next item
230 for changing the default backup extension.
232 A B<-b> flag will be ignored if input is from standard input, or
233 if the B<-html> flag is set.
235 =item B<-bext>=ext, B<--backup-file-extension>=ext
237 Change the extension of the backup file to be something other than the
238 default F<.bak>. See L<Specifying File Extensions>.
240 =item B<-w>, B<--warning-output>
242 Setting B<-w> causes any non-critical warning
243 messages to be reported as errors. These include messages
244 about possible pod problems, possibly bad starting indentation level,
245 and cautions about indirect object usage. The default, B<-nw> or
246 B<--nowarning-output>, is not to include these warnings.
248 =item B<-q>, B<--quiet>
250 Deactivate error messages and syntax checking (for running under
253 For example, if you use a vi-style editor, such as vim, you may execute
254 perltidy as a filter from within the editor using something like
258 where C<n1,n2> represents the selected text. Without the B<-q> flag,
259 any error message may mess up your screen, so be prepared to use your
262 =item B<-log>, B<--logfile>
264 Save the F<.LOG> file, which has many useful diagnostics. Perltidy always
265 creates a F<.LOG> file, but by default it is deleted unless a program bug is
266 suspected. Setting the B<-log> flag forces the log file to be saved.
268 =item B<-g=n>, B<--logfile-gap=n>
270 Set maximum interval between input code lines in the logfile. This purpose of
271 this flag is to assist in debugging nesting errors. The value of C<n> is
272 optional. If you set the flag B<-g> without the value of C<n>, it will be
273 taken to be 1, meaning that every line will be written to the log file. This
274 can be helpful if you are looking for a brace, paren, or bracket nesting error.
276 Setting B<-g> also causes the logfile to be saved, so it is not necessary to
277 also include B<-log>.
279 If no B<-g> flag is given, a value of 50 will be used, meaning that at least
280 every 50th line will be recorded in the logfile. This helps prevent
281 excessively long log files.
283 Setting a negative value of C<n> is the same as not setting B<-g> at all.
285 =item B<-npro> B<--noprofile>
287 Ignore any F<.perltidyrc> command file. Normally, perltidy looks first in
288 your current directory for a F<.perltidyrc> file of parameters. (The format
289 is described below). If it finds one, it applies those options to the
290 initial default values, and then it applies any that have been defined
291 on the command line. If no F<.perltidyrc> file is found, it looks for one
292 in your home directory.
294 If you set the B<-npro> flag, perltidy will not look for this file.
296 =item B<-pro=filename> or B<--profile=filename>
298 To simplify testing and switching .perltidyrc files, this command may be
299 used to specify a configuration file which will override the default
300 name of .perltidyrc. There must not be a space on either side of the
301 '=' sign. For example, the line
303 perltidy -pro=testcfg
305 would cause file F<testcfg> to be used instead of the
306 default F<.perltidyrc>.
308 =item B<-opt>, B<--show-options>
310 Write a list of all options used to the F<.LOG> file.
311 Please see B<--dump-options> for a simpler way to do this.
313 =item B<-f>, B<--force-read-binary>
315 Force perltidy to process binary files. To avoid producing excessive
316 error messages, perltidy skips files identified by the system as non-text.
317 However, valid perl scripts containing binary data may sometimes be identified
318 as non-text, and this flag forces perltidy to process them.
322 =head1 FORMATTING OPTIONS
328 =item B<-l=n>, B<--maximum-line-length=n>
330 The default maximum line length is n=80 characters. Perltidy will try
331 to find line break points to keep lines below this length. However, long
332 quotes and side comments may cause lines to exceed this length.
333 Setting B<-l=0> is equivalent to setting B<-l=(a large number)>.
335 =item B<-i=n>, B<--indent-columns=n>
337 Use n columns per indentation level (default n=4).
341 Using tab characters will almost certainly lead to future portability
342 and maintenance problems, so the default and recommendation is not to
343 use them. For those who prefer tabs, however, there are two different
346 Except for possibly introducing tab indentation characters, as outlined
347 below, perltidy does not introduce any tab characters into your file,
348 and it removes any tabs from the code (unless requested not to do so
349 with B<-fws>). If you have any tabs in your comments, quotes, or
350 here-documents, they will remain.
354 =item B<-et=n>, B<--entab-leading-whitespace>
356 This flag causes each B<n> initial space characters to be replaced by
357 one tab character. Note that the integer B<n> is completely independent
358 of the integer specified for indentation parameter, B<-i=n>.
360 =item B<-t>, B<--tabs>
362 This flag causes one leading tab character to be inserted for each level
363 of indentation. Certain other features are incompatible with this
364 option, and if these options are also given, then a warning message will
365 be issued and this flag will be unset. One example is the B<-lp>
370 =item B<-syn>, B<--check-syntax>
372 This flag causes perltidy to run C<perl -c -T> to check syntax of input
373 and output. (To change the flags passed to perl, see the next
374 item, B<-pscf>). The results are written to the F<.LOG> file, which
375 will be saved if an error is detected in the output script. The output
376 script is not checked if the input script has a syntax error. Perltidy
377 does its own checking, but this option employs perl to get a "second
380 If perl reports errors in the input file, they will not be reported in
381 the error output unless the B<--warning-output> flag is given.
383 The default is B<not> to do this type of syntax checking (although
384 perltidy will still do as much self-checking as possible). The reason
385 is that it causes all code in BEGIN blocks to be executed, for all
386 modules being used, and this opens the door to security issues and
387 infinite loops when running perltidy.
389 =item B<-pscf=s>, B<-perl-syntax-check-flags=s>
391 When perl is invoked to check syntax, the normal flags are C<-c -T>. In
392 addition, if the B<-x> flag is given to perltidy, then perl will also be
393 passed a B<-x> flag. It should not normally be necessary to change
394 these flags, but it can be done with the B<-pscf=s> flag. For example,
395 if the taint flag, C<-T>, is not wanted, the flag could be set to be just
398 Perltidy will pass your string to perl with the exception that it will
399 add a B<-c> and B<-x> if appropriate. The F<.LOG> file will show
400 exactly what flags were passed to perl.
402 =item B<-io>, B<--indent-only>
404 This flag is used to deactivate all formatting and line break changes.
405 When it is in effect, the only change to the script will be indentation.
406 And any flags controlling whitespace and newlines will be ignored. You
407 might want to use this if you are perfectly happy with your whitespace
408 and line breaks, and merely want perltidy to handle the indentation.
409 (This also speeds up perltidy by well over a factor of two, so it might be
410 useful when perltidy is merely being used to help find a brace error in
413 Setting this flag is equivalent to setting B<--freeze-newlines> and
414 B<--freeze-whitespace>.
416 =item B<-ole=s>, B<--output-line-ending=s>
418 where s=C<win>, C<dos>, C<unix>, or C<mac>. This flag tells perltidy
419 to output line endings for a specific system. Normally,
420 perltidy writes files with the line separator character of the host
421 system. The C<win> and C<dos> flags have an identical result.
423 =item B<-ple>, B<--preserve-line-endings>
425 This flag tells perltidy to write its output files with the same line
426 endings as the input file, if possible. It should work for
427 B<dos>, B<unix>, and B<mac> line endings. It will only work if perltidy
428 input comes from a filename (rather than stdin, for example). If
429 perltidy has trouble determining the input file line ending, it will
430 revert to the default behavior of using the line ending of the host system.
434 =head2 Code Indentation Control
438 =item B<-ci=n>, B<--continuation-indentation=n>
440 Continuation indentation is extra indentation spaces applied when
441 a long line is broken. The default is n=2, illustrated here:
444 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
446 The same example, with n=0, is a little harder to read:
449 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
451 The value given to B<-ci> is also used by some commands when a small
452 space is required. Examples are commands for outdenting labels,
453 B<-ola>, and control keywords, B<-okw>.
455 When default values are not used, it is suggested that the value B<n>
456 given with B<-ci=n> be no more than about one-half of the number of
457 spaces assigned to a full indentation level on the B<-i=n> command.
459 =item B<-sil=n> B<--starting-indentation-level=n>
461 By default, perltidy examines the input file and tries to determine the
462 starting indentation level. While it is often zero, it may not be
463 zero for a code snippet being sent from an editing session. If the
464 default method does not work correctly, or you want to change the
465 starting level, use B<-sil=n>, to force the starting level to be n.
467 =item List indentation using B<-lp>, B<--line-up-parentheses>
469 By default, perltidy indents lists with 4 spaces, or whatever value
470 is specified with B<-i=n>. Here is a small list formatted in this way:
474 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
475 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
478 Use the B<-lp> flag to add extra indentation to cause the data to begin
479 past the opening parentheses of a sub call or list, or opening square
480 bracket of an anonymous array, or opening curly brace of an anonymous
481 hash. With this option, the above list would become:
485 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
486 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
489 If the available line length (see B<-l=n> ) does not permit this much
490 space, perltidy will use less. For alternate placement of the
491 closing paren, see the next section.
493 This option has no effect on code BLOCKS, such as if/then/else blocks,
494 which always use whatever is specified with B<-i=n>. Also, the
495 existence of line breaks and/or block comments between the opening and
496 closing parens may cause perltidy to temporarily revert to its default
499 Note: The B<-lp> option may not be used together with the B<-t> tabs option.
500 It may, however, be used with the B<-et=n> tab method.
502 In addition, any parameter which significantly restricts the ability of
503 perltidy to choose newlines will conflict with B<-lp> and will cause
504 B<-lp> to be deactivated. These include B<-io>, B<-fnl>, B<-nanl>, and
505 B<-ndnl>. The reason is that the B<-lp> indentation style can require
506 the careful coordination of an arbitrary number of break points in
507 hierarchical lists, and these flags may prevent that.
509 =item B<-cti=n>, B<--closing-token-indentation>
511 The B<-cti=n> flag controls the indentation of a line beginning with
512 a C<)>, C<]>, or a non-block C<}>. Such a line receives:
514 -cti = 0 no extra indentation (default)
515 -cti = 1 extra indentation such that the closing token
516 aligns with its opening token.
517 -cti = 2 one extra indentation level if the line looks like:
519 -cti = 3 one extra indentation level always
521 The flags B<-cti=1> and B<-cti=2> work well with the B<-lp> flag (previous
524 # perltidy -lp -cti=1
526 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
527 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
530 # perltidy -lp -cti=2
532 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
533 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
536 These flags are merely hints to the formatter and they may not always be
537 followed. In particular, if -lp is not being used, the indentation for
538 B<cti=1> is constrained to be no more than one indentation level.
540 If desired, this control can be applied independently to each of the
541 closing container token types. In fact, B<-cti=n> is merely an
542 abbreviation for B<-cpi=n -csbi=n -cbi=n>, where:
543 B<-cpi> or B<--closing-paren-indentation> controls B<)>'s,
544 B<-csbi> or B<--closing-square-bracket-indentation> controls B<]>'s,
545 B<-cbi> or B<--closing-brace-indentation> controls non-block B<}>'s.
547 =item B<-icp>, B<--indent-closing-paren>
549 The B<-icp> flag is equivalent to
550 B<-cti=2>, described in the previous section. The B<-nicp> flag is
551 equivalent B<-cti=0>. They are included for backwards compatability.
553 =item B<-icb>, B<--indent-closing-brace>
555 The B<-icb> option gives one extra level of indentation to a brace which
556 terminates a code block . For example,
565 The default is not to do this, indicated by B<-nicb>.
567 =item B<-olq>, B<--outdent-long-quotes>
569 When B<-olq> is set, lines which is a quoted string longer than the
570 value B<maximum-line-length> will have their indentation removed to make
571 them more readable. This is the default. To prevent such out-denting,
572 use B<-nolq> or B<--nooutdent-long-lines>.
574 =item B<-oll>, B<--outdent-long-lines>
576 This command is equivalent to B<--outdent-long-quotes> and
577 B<--outdent-long-comments>, and it is included for compatibility with previous
578 versions of perltidy. The negation of this also works, B<-noll> or
579 B<--nooutdent-long-lines>, and is equivalent to setting B<-nolq> and B<-nolc>.
581 =item Outdenting Labels: B<-ola>, B<--outdent-labels>
583 This command will cause labels to be outdented by 2 spaces (or whatever B<-ci>
584 has been set to), if possible. This is the default. For example:
587 LOOP: while ( $i = <FOTOS> ) {
593 Use B<-nola> to not outdent labels.
595 =item Outdenting Keywords
599 =item B<-okw>, B<--outdent-keywords>
601 The command B<-okw> will will cause certain leading control keywords to
602 be outdented by 2 spaces (or whatever B<-ci> has been set to), if
603 possible. By default, these keywords are C<redo>, C<next>, C<last>,
604 C<goto>, and C<return>. The intention is to make these control keywords
605 easier to see. To change this list of keywords being outdented, see
608 For example, using C<perltidy -okw> on the previous example gives:
611 LOOP: while ( $i = <FOTOS> ) {
617 The default is not to do this.
619 =item Specifying Outdented Keywords: B<-okwl=string>, B<--outdent-keyword-list=string>
621 This command can be used to change the keywords which are outdented with
622 the B<-okw> command. The parameter B<string> is a required list of perl
623 keywords, which should be placed in quotes if there are more than one.
624 By itself, it does not cause any outdenting to occur, so the B<-okw>
625 command is still required.
627 For example, the commands C<-okwl="next last redo goto" -okw> will cause
628 those four keywords to be outdented. It is probably simplest to place
629 any B<-okwl> command in a F<.perltidyrc> file.
635 =head2 Whitespace Control
637 Whitespace refers to the blank space between variables, operators,
638 and other code tokens.
642 =item B<-fws>, B<--freeze-whitespace>
644 This flag causes your original whitespace to remain unchanged, and
645 causes the rest of the whitespace commands in this section, the
646 Code Indentation section, and
647 the Comment Control section to be ignored.
649 =item Tightness of curly braces, parentheses, and square brackets.
651 Here the term "tightness" will mean the closeness with which
652 pairs of enclosing tokens, such as parentheses, contain the quantities
653 within. A numerical value of 0, 1, or 2 defines the tightness, with
654 0 being least tight and 2 being most tight. Spaces within containers
655 are always symmetric, so if there is a space after a C<(> then there
656 will be a space before the corresponding C<)>.
658 The B<-pt=n> or B<--paren-tightness=n> parameter controls the space within
659 parens. The example below shows the effect of the three possible
662 if ( ( my $len_tab = length( $tabstr ) ) > 0 ) { # -pt=0
663 if ( ( my $len_tab = length($tabstr) ) > 0 ) { # -pt=1 (default)
664 if ((my $len_tab = length($tabstr)) > 0) { # -pt=2
666 When n is 0, there is always a space to the right of a '(' and to the left
667 of a ')'. For n=2 there is never a space. For n=1, the default, there
668 is a space unless the quantity within the parens is a single token, such
669 as an identifier or quoted string.
671 Likewise, the parameter B<-sbt=n> or B<--square-bracket-tightness=n>
672 controls the space within square brackets, as illustrated below.
674 $width = $col[ $j + $k ] - $col[ $j ]; # -sbt=0
675 $width = $col[ $j + $k ] - $col[$j]; # -sbt=1 (default)
676 $width = $col[$j + $k] - $col[$j]; # -sbt=2
678 Curly braces which do not contain code blocks are controlled by
679 the parameter B<-bt=n> or B<--brace-tightness=n>.
681 $obj->{ $parsed_sql->{ 'table' }[0] }; # -bt=0
682 $obj->{ $parsed_sql->{'table'}[0] }; # -bt=1 (default)
683 $obj->{$parsed_sql->{'table'}[0]}; # -bt=2
685 And finally, curly braces which contain blocks of code are controlled by the
686 parameter B<-bbt=n> or B<--block-brace-tightness=n> as illustrated in the
689 %bf = map { $_ => -M $_ } grep { /\.deb$/ } dirents '.'; # -bbt=0 (default)
690 %bf = map { $_ => -M $_ } grep {/\.deb$/} dirents '.'; # -bbt=1
691 %bf = map {$_ => -M $_} grep {/\.deb$/} dirents '.'; # -bbt=2
693 =item B<-sts>, B<--space-terminal-semicolon>
695 Some programmers prefer a space before all terminal semicolons. The
696 default is for no such space, and is indicated with B<-nsts> or
697 B<--nospace-terminal-semicolon>.
700 $i = 1; # -nsts (default)
702 =item B<-sfs>, B<--space-for-semicolon>
704 Semicolons within B<for> loops may sometimes be hard to see,
705 particularly when commas are also present. This option places spaces on
706 both sides of these special semicolons, and is the default. Use
707 B<-nsfs> or B<--nospace-for-semicolon> to deactivate it.
709 for ( @a = @$ap, $u = shift @a ; @a ; $u = $v ) { # -sfs (default)
710 for ( @a = @$ap, $u = shift @a; @a; $u = $v ) { # -nsfs
712 =item B<-asc>, B<--add-semicolons>
714 Setting B<-asc> allows perltidy to add any missing optional semicolon at the end
715 of a line which is followed by a closing curly brace on the next line. This
716 is the default, and may be deactivated with B<-nasc> or B<--noadd-semicolons>.
718 =item B<-dsm>, B<--delete-semicolons>
720 Setting B<-dsm> allows perltidy to delete extra semicolons which are
721 simply empty statements. This is the default, and may be deactivated
722 with B<-ndsm> or B<--nodelete-semicolons>. (Such semicolons are not
723 deleted, however, if they would promote a side comment to a block
726 =item B<-aws>, B<--add-whitespace>
728 Setting this option allows perltidy to add certain whitespace improve
729 code readability. This is the default. If you do not want any
730 whitespace added, but are willing to have some whitespace deleted, use
731 B<-naws>. (Use B<-fws> to leave whitespace completely unchanged).
733 =item B<-dws>, B<--delete-old-whitespace>
735 Setting this option allows perltidy to remove some old whitespace
736 between characters, if necessary. This is the default. If you
737 do not want any old whitespace removed, use B<-ndws> or
738 B<--nodelete-old-whitespace>.
740 =item Detailed whitespace controls around tokens
742 For those who want more detailed control over the whitespace around
743 tokens, there are four parameters which can directly modify the default
744 whitespace rules built into perltidy for any token. They are:
746 B<-wls=s> or B<--want-left-space=s>,
748 B<-nwls=s> or B<--nowant-left-space=s>,
750 B<-wrs=s> or B<--want-right-space=s>,
752 B<-nwrs=s> or B<--nowant-right-space=s>.
754 These parameters are each followed by a quoted string, B<s>, containing a
755 list of token types. No more than one of each of these parameters
756 should be specified, because repeating a command-line parameter
757 always overwrites the previous one before perltidy ever sees it.
759 To illustrate how these are used, suppose it is desired that there be no
760 space on either side of the token types B<= + - / *>. The following two
761 parameters would specify this desire:
763 -nwls="= + - / *" -nwrs="= + - / *"
765 (Note that the token types are in quotes, and that they are separated by
766 spaces). With these modified whitespace rules, the following line of math:
768 $root = -$b + sqrt( $b * $b - 4. * $a * $c ) / ( 2. * $a );
772 $root=-$b+sqrt( $b*$b-4.*$a*$c )/( 2.*$a );
774 These parameters should be considered to be hints to perltidy rather
775 than fixed rules, because perltidy must try to resolve conflicts that
776 arise between them and all of the other rules that it uses. One
777 conflict that can arise is if, between two tokens, the left token wants
778 a space and the right one doesn't. In this case, the token not wanting
779 a space takes priority.
781 It is necessary to have a list of all token types in order to create
782 this type of input. Such a list can be obtained by the command
783 B<--dump-token-types>. Also try the B<-D> flag on a short snippet of code
784 and look at the .DEBUG file to see the tokenization.
786 B<WARNING> Be sure to put these tokens in quotes to avoid having them
787 misinterpreted by your command shell.
789 =item Space between specific keywords and opening paren
791 When an opening paren follows a Perl keyword, no space is introduced after the
792 keyword, unless it is (by default) one of these:
794 my local our and or eq ne if else elsif until unless
795 while for foreach return switch case given when
797 These defaults can be modified with two commands:
799 B<-sak=s> or B<--space-after-keyword=s> adds keywords.
801 B<-nsak=s> or B<--nospace-after-keyword=s> removes keywords.
803 where B<s> is a list of keywords (in quotes if necessary). For example,
805 my ( $a, $b, $c ) = @_; # default
806 my( $a, $b, $c ) = @_; # -nsak="my local our"
808 To put a space after all keywords, see the next item.
810 =item Space between all keywords and opening parens
812 When an opening paren follows a function or keyword, no space is introduced
813 after the keyword except for the keywords noted in the previous item. To
814 always put a space between a function or keyword and its opening paren,
817 B<-skp> or B<--space-keyword-paren>
819 You will probably also want to use the flag B<-sfp> (next item) too.
821 =item Space between all function names and opening parens
823 When an opening paren follows a function the default is not to introduce
824 a space. To cause a space to be introduced use:
826 B<-sfp> or B<--space-function-paren>
828 myfunc( $a, $b, $c ); # default
829 myfunc ( $a, $b, $c ); # -sfp
831 You will probably also want to use the flag B<-skp> (previous item) too.
833 =item Trimming whitespace around C<qw> quotes
835 B<-tqw> or B<--trim-qw> provide the default behavior of trimming
836 spaces around multi-line C<qw> quotes and indenting them appropriately.
838 B<-ntqw> or B<--notrim-qw> cause leading and trailing whitespace around
839 multi-line C<qw> quotes to be left unchanged. This option will not
840 normally be necessary, but was added for testing purposes, because in
841 some versions of perl, trimming C<qw> quotes changes the syntax tree.
845 =head2 Comment Controls
847 Perltidy has a number of ways to control the appearance of both block comments
848 and side comments. The term B<block comment> here refers to a full-line
849 comment, whereas B<side comment> will refer to a comment which appears on a
850 line to the right of some code.
854 =item B<-ibc>, B<--indent-block-comments>
856 Block comments normally look best when they are indented to the same
857 level as the code which follows them. This is the default behavior, but
858 you may use B<-nibc> to keep block comments left-justified. Here is an
861 # this comment is indented (-ibc, default)
862 if ($task) { yyy(); }
864 The alternative is B<-nibc>:
866 # this comment is not indented (-nibc)
867 if ($task) { yyy(); }
869 See also the next item, B<-isbc>, as well as B<-sbc>, for other ways to
870 have some indented and some outdented block comments.
872 =item B<-isbc>, B<--indent-spaced-block-comments>
874 If there is no leading space on the line, then the comment will not be
875 indented, and otherwise it may be.
877 If both B<-ibc> and B<-isbc> are set, then B<-isbc> takes priority.
879 =item B<-olc>, B<--outdent-long-comments>
881 When B<-olc> is set, lines which are full-line (block) comments longer
882 than the value B<maximum-line-length> will have their indentation
883 removed. This is the default; use B<-nolc> to prevent outdenting.
885 =item B<-msc=n>, B<--minimum-space-to-comment=n>
887 Side comments look best when lined up several spaces to the right of
888 code. Perltidy will try to keep comments at least n spaces to the
889 right. The default is n=4 spaces.
891 =item B<-hsc>, B<--hanging-side-comments>
893 By default, perltidy tries to identify and align "hanging side
894 comments", which are something like this:
896 my $IGNORE = 0; # This is a side comment
897 # This is a hanging side comment
900 A comment is considered to be a hanging side comment if (1) it immediately
901 follows a line with a side comment, or another hanging side comment, and
902 (2) there is some leading whitespace on the line.
903 To deactivate this feature, use B<-nhsc> or B<--nohanging-side-comments>.
904 If block comments are preceded by a blank line, or have no leading
905 whitespace, they will not be mistaken as hanging side comments.
907 =item Closing Side Comments
909 A closing side comment is a special comment which perltidy can
910 automatically create and place after the closing brace of a code block.
911 They can be useful for code maintenance and debugging. The command
912 B<-csc> (or B<--closing-side-comments>) adds or updates closing side
913 comments. For example, here is a small code snippet
916 if ( !defined( $_[0] ) ) {
917 print("Hello, World\n");
920 print( $_[0], "\n" );
924 And here is the result of processing with C<perltidy -csc>:
927 if ( !defined( $_[0] ) ) {
928 print("Hello, World\n");
931 print( $_[0], "\n" );
935 A closing side comment was added for C<sub message> in this case, but not
936 for the C<if> and C<else> blocks, because they were below the 6 line
937 cutoff limit for adding closing side comments. This limit may be
938 changed with the B<-csci> command, described below.
940 The command B<-dcsc> (or B<--delete-closing-side-comments>) reverses this
941 process and removes these comments.
943 Several commands are available to modify the behavior of these two basic
944 commands, B<-csc> and B<-dcsc>:
948 =item B<-csci=n>, or B<--closing-side-comment-interval=n>
950 where C<n> is the minimum number of lines that a block must have in
951 order for a closing side comment to be added. The default value is
952 C<n=6>. To illustrate:
954 # perltidy -csci=2 -csc
956 if ( !defined( $_[0] ) ) {
957 print("Hello, World\n");
958 } ## end if ( !defined( $_[0] ))
960 print( $_[0], "\n" );
961 } ## end else [ if ( !defined( $_[0] ))
964 Now the C<if> and C<else> blocks are commented. However, now this has
965 become very cluttered.
967 =item B<-cscp=string>, or B<--closing-side-comment-prefix=string>
969 where string is the prefix used before the name of the block type. The
970 default prefix, shown above, is C<## end>. This string will be added to
971 closing side comments, and it will also be used to recognize them in
972 order to update, delete, and format them. Any comment identified as a
973 closing side comment will be placed just a single space to the right of
976 =item B<-cscl=string>, or B<--closing-side-comment-list-string>
978 where C<string> is a list of block types to be tagged with closing side
979 comments. By default, all code block types preceded by a keyword or
980 label (such as C<if>, C<sub>, and so on) will be tagged. The B<-cscl>
981 command changes the default list to be any selected block types; see
982 L<Specifying Block Types>.
983 For example, the following command
984 requests that only C<sub>'s, labels, C<BEGIN>, and C<END> blocks be
985 affected by any B<-csc> or B<-dcsc> operation:
987 -cscl="sub : BEGIN END"
989 =item B<-csct=n>, or B<--closing-side-comment-maximum-text=n>
991 The text appended to certain block types, such as an C<if> block, is
992 whatever lies between the keyword introducing the block, such as C<if>,
993 and the opening brace. Since this might be too much text for a side
994 comment, there needs to be a limit, and that is the purpose of this
995 parameter. The default value is C<n=20>, meaning that no additional
996 tokens will be appended to this text after its length reaches 20
997 characters. Omitted text is indicated with C<...>. (Tokens, including
998 sub names, are never truncated, however, so actual lengths may exceed
999 this). To illustrate, in the above example, the appended text of the
1000 first block is C< ( !defined( $_[0] )...>. The existing limit of
1001 C<n=20> caused this text to be truncated, as indicated by the C<...>.
1003 =item B<-csce=n>, or B<--closing-side-comment-else-flag=n>
1005 The default, B<n=0>, places the text of the opening C<if> statement after any
1008 If B<n=2> is used, then each C<elsif> is also given the text of the opening
1009 C<if> statement. Also, an C<else> will include the text of a preceding
1010 C<elsif> statement. Note that this may result some long closing
1013 If B<n=1> is used, the results will be the same as B<n=2> whenever the
1014 resulting line length is less than the maximum allowed.
1016 =item B<-cscw>, or B<--closing-side-comment-warnings>
1018 This parameter is intended to help make the initial transition to the use of
1019 closing side comments.
1021 things to happen if a closing side comment replaces an existing, different
1022 closing side comment: first, an error message will be issued, and second, the
1023 original side comment will be placed alone on a new specially marked comment
1024 line for later attention.
1026 The intent is to avoid clobbering existing hand-written side comments
1027 which happen to match the pattern of closing side comments. This flag
1028 should only be needed on the first run with B<-csc>.
1032 B<Important Notes on Closing Side Comments:>
1038 Closing side comments are only placed on lines terminated with a closing
1039 brace. Certain closing styles, such as the use of cuddled elses
1040 (B<-ce>), preclude the generation of some closing side comments.
1044 Please note that adding or deleting of closing side comments takes
1045 place only through the commands B<-csc> or B<-dcsc>. The other commands,
1046 if used, merely modify the behavior of these two commands.
1050 It is recommended that the B<-cscw> flag be used along with B<-csc> on
1051 the first use of perltidy on a given file. This will prevent loss of
1052 any existing side comment data which happens to have the csc prefix.
1056 Once you use B<-csc>, you should continue to use it so that any
1057 closing side comments remain correct as code changes. Otherwise, these
1058 comments will become incorrect as the code is updated.
1062 If you edit the closing side comments generated by perltidy, you must also
1063 change the prefix to be different from the closing side comment prefix.
1064 Otherwise, your edits will be lost when you rerun perltidy with B<-csc>. For
1065 example, you could simply change C<## end> to be C<## End>, since the test is
1066 case sensitive. You may also want to use the B<-ssc> flag to keep these
1067 modified closing side comments spaced the same as actual closing side comments.
1071 Temporarily generating closing side comments is a useful technique for
1072 exploring and/or debugging a perl script, especially one written by someone
1073 else. You can always remove them with B<-dcsc>.
1077 =item Static Block Comments
1079 Static block comments are block comments with a special leading pattern,
1080 C<##> by default, which will be treated slightly differently from other
1081 block comments. They effectively behave as if they had glue along their
1082 left and top edges, because they stick to the left edge and previous line
1083 when there is no blank spaces in those places. This option is
1084 particularly useful for controlling how commented code is displayed.
1088 =item B<-sbc>, B<--static-block-comments>
1090 When B<-sbc> is used, a block comment with a special leading pattern, C<##> by
1091 default, will be treated specially.
1093 Comments so identified are treated as follows:
1099 If there is no leading space on the line, then the comment will not
1100 be indented, and otherwise it may be,
1104 no new blank line will be
1105 inserted before such a comment, and
1109 such a comment will never become
1110 a hanging side comment.
1114 For example, assuming C<@month_of_year> is
1117 @month_of_year = ( # -sbc (default)
1118 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1122 Without this convention, the above code would become
1124 @month_of_year = ( # -nsbc
1125 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1131 which is not as clear.
1132 The default is to use B<-sbc>. This may be deactivated with B<-nsbc>.
1134 =item B<-sbcp=string>, B<--static-block-comment-prefix=string>
1136 This parameter defines the prefix used to identify static block comments
1137 when the B<-sbc> parameter is set. The default prefix is C<##>,
1138 corresponding to C<-sbcp=##>. The prefix is actually part of a perl
1139 pattern used to match lines and it must either begin with C<#> or C<^#>.
1140 In the first case a prefix ^\s* will be added to match any leading
1141 whitespace, while in the second case the pattern will match only
1142 comments with no leading whitespace. For example, to
1143 identify all comments as static block comments, one would use C<-sbcp=#>.
1144 To identify all left-adjusted comments as static block comments, use C<-sbcp='^#'>.
1146 Please note that B<-sbcp> merely defines the pattern used to identify static
1147 block comments; it will not be used unless the switch B<-sbc> is set. Also,
1148 please be aware that since this string is used in a perl regular expression
1149 which identifies these comments, it must enable a valid regular expression to
1152 =item B<-osbc>, B<--outdent-static-block-comments>
1154 The command B<-osbc> will will cause static block comments to be outdented by 2
1155 spaces (or whatever B<-ci=n> has been set to), if possible.
1159 =item Static Side Comments
1161 Static side comments are side comments with a special leading pattern.
1162 This option can be useful for controlling how commented code is displayed
1163 when it is a side comment.
1167 =item B<-ssc>, B<--static-side-comments>
1169 When B<-ssc> is used, a side comment with a static leading pattern, which is
1170 C<##> by default, will be be spaced only a single space from previous
1171 character, and it will not be vertically aligned with other side comments.
1173 The default is B<-nssc>.
1175 =item B<-sscp=string>, B<--static-side-comment-prefix=string>
1177 This parameter defines the prefix used to identify static side comments
1178 when the B<-ssc> parameter is set. The default prefix is C<##>,
1179 corresponding to C<-sscp=##>.
1181 Please note that B<-sscp> merely defines the pattern used to identify
1182 static side comments; it will not be used unless the switch B<-ssc> is
1183 set. Also, note that this string is used in a perl regular expression
1184 which identifies these comments, so it must enable a valid regular
1185 expression to be formed.
1192 =head2 Skipping Selected Sections of Code
1194 Selected lines of code may be passed verbatim to the output without any
1195 formatting. This feature is enabled by default but can be disabled with
1196 the B<--noformat-skipping> or B<-nfs> flag. It should be used sparingly to
1197 avoid littering code with markers, but it might be helpful for working
1198 around occasional problems. For example it might be useful for keeping
1199 the indentation of old commented code unchanged, keeping indentation of
1200 long blocks of aligned comments unchanged, keeping certain list
1201 formatting unchanged, or working around a glitch in perltidy.
1205 =item B<-fs>, B<--format-skipping>
1207 This flag, which is enabled by default, causes any code between
1208 special beginning and ending comment markers to be passed to the
1209 output without formatting. The default beginning marker is #<<<
1210 and the default ending marker is #>>> but they
1211 may be changed (see next items below). Additional text may appear on
1212 these special comment lines provided that it is separated from the
1213 marker by at least one space. For example
1215 #<<< do not let perltidy touch this
1223 The comment markers may be placed at any location that a block comment may
1224 appear. If they do not appear to be working, use the -log flag and examine the
1225 F<.LOG> file. Use B<-nfs> to disable this feature.
1227 =item B<-fsb=string>, B<--format-skipping-begin=string>
1229 The B<-fsb=string> parameter may be used to change the beginning marker for
1230 format skipping. The default is equivalent to -fsb='#<<<'. The string that
1231 you enter must begin with a # and should be in quotes as necessary to get past
1232 the command shell of your system. It is actually the leading text of a pattern
1233 that is constructed by appending a '\s', so you must also include backslashes
1234 for characters to be taken literally rather than as patterns.
1236 Some examples show how example strings become patterns:
1238 -fsb='#\{\{\{' becomes /^#\{\{\{\s/ which matches #{{{ but not #{{{{
1239 -fsb='#\*\*' becomes /^#\*\*\s/ which matches #** but not #***
1240 -fsb='#\*{2,}' becomes /^#\*{2,}\s/ which matches #** and #*****
1242 =item B<-fse=string>, B<--format-skipping-end=string>
1244 The B<-fsb=string> is the corresponding parameter used to change the
1245 ending marker for format skipping. The default is equivalent to
1250 =head2 Line Break Control
1254 =item B<-fnl>, B<--freeze-newlines>
1256 If you do not want any changes to the line breaks in your script, set
1257 B<-fnl>, and they will remain fixed, and the rest of the commands in
1258 this section and sections
1259 L<Controlling List Formatting>,
1260 L<Retaining or Ignoring Existing Line Breaks>, and
1261 L<Blank Line Control> will be ignored. You may want to use B<-noll>
1264 =item B<-ce>, B<--cuddled-else>
1266 Enable the "cuddled else" style, in which C<else> and C<elsif> are
1267 follow immediately after the curly brace closing the previous block.
1268 The default is not to use cuddled elses, and is indicated with the flag
1269 B<-nce> or B<--nocuddled-else>. Here is a comparison of the
1281 else { # -nce (default)
1285 =item B<-bl>, B<--opening-brace-on-new-line>
1287 Use the flag B<-bl> to place the opening brace on a new line:
1289 if ( $input_file eq '-' ) # -bl
1291 important_function();
1294 This flag applies to all structural blocks, including sub's (unless
1295 the B<-sbl> flag is set -- see next item).
1297 The default style, B<-nbl>, places an opening brace on the same line as
1298 the keyword introducing it. For example,
1300 if ( $input_file eq '-' ) { # -nbl (default)
1302 =item B<-sbl>, B<--opening-sub-brace-on-new-line>
1304 The flag B<-sbl> can be used to override the value of B<-bl> for
1305 opening sub braces. For example,
1309 produces this result:
1313 if (!defined($_[0])) {
1314 print("Hello, World\n");
1321 This flag is negated with B<-nsbl>. If B<-sbl> is not specified,
1322 the value of B<-bl> is used.
1324 =item B<-bli>, B<--brace-left-and-indent>
1326 The flag B<-bli> is the same as B<-bl> but in addition it causes one
1327 unit of continuation indentation ( see B<-ci> ) to be placed before
1328 an opening and closing block braces.
1332 if ( $input_file eq '-' ) # -bli
1334 important_function();
1337 By default, this extra indentation occurs for blocks of type:
1338 B<if>, B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>,
1339 B<while>, B<until>, and also with a preceding label. The next item
1340 shows how to change this.
1342 =item B<-blil=s>, B<--brace-left-and-indent-list=s>
1344 Use this parameter to change the types of block braces for which the
1345 B<-bli> flag applies; see L<Specifying Block Types>. For example,
1346 B<-blil='if elsif else'> would apply it to only C<if/elsif/else> blocks.
1348 =item B<-bar>, B<--opening-brace-always-on-right>
1350 The default style, B<-nbl> places the opening code block brace on a new
1351 line if it does not fit on the same line as the opening keyword, like
1354 if ( $bigwasteofspace1 && $bigwasteofspace2
1355 || $bigwasteofspace3 && $bigwasteofspace4 )
1357 big_waste_of_time();
1360 To force the opening brace to always be on the right, use the B<-bar>
1361 flag. In this case, the above example becomes
1363 if ( $bigwasteofspace1 && $bigwasteofspace2
1364 || $bigwasteofspace3 && $bigwasteofspace4 ) {
1365 big_waste_of_time();
1368 A conflict occurs if both B<-bl> and B<-bar> are specified.
1370 =item B<-otr>, B<--opening-token-right> and related flags
1372 The B<-otr> flag is a hint that perltidy should not place a break between a
1373 comma and an opening token. For example:
1375 # default formatting
1376 push @{ $self->{$module}{$key} },
1378 accno => $ref->{accno},
1379 description => $ref->{description}
1383 push @{ $self->{$module}{$key} }, {
1384 accno => $ref->{accno},
1385 description => $ref->{description}
1388 The flag B<-otr> is actually a synonym for three other flags
1389 which can be used to control parens, hash braces, and square brackets
1390 separately if desired:
1392 -opr or --opening-paren-right
1393 -ohbr or --opening-hash-brace-right
1394 -osbr or --opening-square-bracket-right
1396 =item Vertical tightness of non-block curly braces, parentheses, and square brackets.
1398 These parameters control what shall be called vertical tightness. Here are the
1405 Opening tokens (except for block braces) are controlled by B<-vt=n>, or
1406 B<--vertical-tightness=n>, where
1408 -vt=0 always break a line after opening token (default).
1409 -vt=1 do not break unless this would produce more than one
1410 step in indentation in a line.
1411 -vt=2 never break a line after opening token
1415 You must also use the B<-lp> flag when you use the B<-vt> flag; the
1416 reason is explained below.
1420 Closing tokens (except for block braces) are controlled by B<-vtc=n>, or
1421 B<--vertical-tightness-closing=n>, where
1423 -vtc=0 always break a line before a closing token (default),
1424 -vtc=1 do not break before a closing token which is followed
1425 by a semicolon or another closing token, and is not in
1427 -vtc=2 never break before a closing token.
1429 The rules for B<-vtc=1> are designed to maintain a reasonable balance
1430 between tightness and readability in complex lists.
1434 Different controls may be applied to to different token types,
1435 and it is also possible to control block braces; see below.
1439 Finally, please note that these vertical tightness flags are merely
1440 hints to the formatter, and it cannot always follow them. Things which
1441 make it difficult or impossible include comments, blank lines, blocks of
1442 code within a list, and possibly the lack of the B<-lp> parameter.
1443 Also, these flags may be ignored for very small lists (2 or 3 lines in
1448 Here are some examples:
1450 # perltidy -lp -vt=0 -vtc=0
1458 # perltidy -lp -vt=1 -vtc=0
1459 %romanNumerals = ( one => 'I',
1465 # perltidy -lp -vt=1 -vtc=1
1466 %romanNumerals = ( one => 'I',
1471 The difference between B<-vt=1> and B<-vt=2> is shown here:
1473 # perltidy -lp -vt=1
1475 mysprintf( "(void)find_threadsv(%s);",
1476 cstring( $threadsv_names[ $op->targ ] )
1480 # perltidy -lp -vt=2
1481 $init->add( mysprintf( "(void)find_threadsv(%s);",
1482 cstring( $threadsv_names[ $op->targ ] )
1486 With B<-vt=1>, the line ending in C<add(> does not combine with the next
1487 line because the next line is not balanced. This can help with
1488 readability, but B<-vt=2> can be used to ignore this rule.
1490 The tightest, and least readable, code is produced with both C<-vt=2> and
1493 # perltidy -lp -vt=2 -vtc=2
1494 $init->add( mysprintf( "(void)find_threadsv(%s);",
1495 cstring( $threadsv_names[ $op->targ ] ) ) );
1497 Notice how the code in all of these examples collapses vertically as
1498 B<-vt> increases, but the indentation remains unchanged. This is
1499 because perltidy implements the B<-vt> parameter by first formatting as
1500 if B<-vt=0>, and then simply overwriting one output line on top of the
1501 next, if possible, to achieve the desired vertical tightness. The
1502 B<-lp> indentation style has been designed to allow this vertical
1503 collapse to occur, which is why it is required for the B<-vt> parameter.
1505 The B<-vt=n> and B<-vtc=n> parameters apply to each type of container
1506 token. If desired, vertical tightness controls can be applied
1507 independently to each of the closing container token types.
1509 The parameters for controlling parentheses are B<-pvt=n> or
1510 B<--paren-vertical-tightness=n>, and B<-pcvt=n> or
1511 B<--paren-vertical-tightness-closing=n>.
1513 Likewise, the parameters for square brackets are B<-sbvt=n> or
1514 B<--square-bracket-vertical-tightness=n>, and B<-sbcvt=n> or
1515 B<--square-bracket-vertical-tightness-closing=n>.
1517 Finally, the parameters for controlling non-code block braces are
1518 B<-bvt=n> or B<--brace-vertical-tightness=n>, and B<-bcvt=n> or
1519 B<--brace-vertical-tightness-closing=n>.
1521 In fact, the parameter B<-vt=n> is actually just an abbreviation for
1522 B<-pvt=n -bvt=n sbvt=n>, and likewise B<-vtc=n> is an abbreviation
1523 for B<-pvtc=n -bvtc=n sbvtc=n>.
1525 =item B<-bbvt=n> or B<--block-brace-vertical-tightness=n>
1527 The B<-bbvt=n> flag is just like the B<-vt=n> flag but applies
1528 to opening code block braces.
1530 -bbvt=0 break after opening block brace (default).
1531 -bbvt=1 do not break unless this would produce more than one
1532 step in indentation in a line.
1533 -bbvt=2 do not break after opening block brace.
1535 It is necessary to also use either B<-bl> or B<-bli> for this to work,
1536 because, as with other vertical tightness controls, it is implemented by
1537 simply overwriting a line ending with an opening block brace with the
1538 subsequent line. For example:
1540 # perltidy -bli -bbvt=0
1541 if ( open( FILE, "< $File" ) )
1543 while ( $File = <FILE> )
1551 # perltidy -bli -bbvt=1
1552 if ( open( FILE, "< $File" ) )
1553 { while ( $File = <FILE> )
1560 By default this applies to blocks associated with keywords B<if>,
1561 B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>, B<while>,
1562 B<until>, and also with a preceding label. This can be changed with
1563 the parameter B<-bbvtl=string>, or
1564 B<--block-brace-vertical-tightness-list=string>, where B<string> is a
1565 space-separated list of block types. For more information on the
1566 possible values of this string, see L<Specifying Block Types>
1568 For example, if we want to just apply this style to C<if>,
1569 C<elsif>, and C<else> blocks, we could use
1570 C<perltidy -bli -bbvt=1 -bbvtl='if elsif else'>.
1572 There is no vertical tightness control for closing block braces; with
1573 the exception of one-line blocks, they will normally remain on a
1576 =item B<-sot>, B<--stack-opening-token> and related flags
1578 The B<-sot> flag tells perltidy to "stack" opening tokens
1579 when possible to avoid lines with isolated opening tokens.
1584 $opt_c = Text::CSV_XS->new(
1593 $opt_c = Text::CSV_XS->new( {
1600 For detailed control of individual closing tokens the following
1601 controls can be used:
1603 -sop or --stack-opening-paren
1604 -sohb or --stack-opening-hash-brace
1605 -sosb or --stack-opening-square-bracket
1607 The flag B<-sot> is a synonym for B<-sop -sohb -sosb>.
1609 =item B<-sct>, B<--stack-closing-token> and related flags
1611 The B<-sct> flag tells perltidy to "stack" closing tokens
1612 when possible to avoid lines with isolated closing tokens.
1617 $opt_c = Text::CSV_XS->new(
1626 $opt_c = Text::CSV_XS->new(
1633 The B<-sct> flag is somewhat similar to the B<-vtc> flags, and in some
1634 cases it can give a similar result. The difference is that the B<-vtc>
1635 flags try to avoid lines with leading opening tokens by "hiding" them at
1636 the end of a previous line, whereas the B<-sct> flag merely tries to
1637 reduce the number of lines with isolated closing tokens by stacking them
1638 but does not try to hide them. For example:
1641 $opt_c = Text::CSV_XS->new(
1645 always_quote => 1, } );
1647 For detailed control of the stacking of individual closing tokens the
1648 following controls can be used:
1650 -scp or --stack-closing-paren
1651 -schb or --stack-closing-hash-brace
1652 -scsb or --stack-closing-square-bracket
1654 The flag B<-sct> is a synonym for B<-scp -schb -scsb>.
1656 =item B<-dnl>, B<--delete-old-newlines>
1658 By default, perltidy first deletes all old line break locations, and then it
1659 looks for good break points to match the desired line length. Use B<-ndnl>
1660 or B<--nodelete-old-newlines> to force perltidy to retain all old line break
1663 =item B<-anl>, B<--add-newlines>
1665 By default, perltidy will add line breaks when necessary to create
1666 continuations of long lines and to improve the script appearance. Use
1667 B<-nanl> or B<--noadd-newlines> to prevent any new line breaks.
1669 This flag does not prevent perltidy from eliminating existing line
1670 breaks; see B<--freeze-newlines> to completely prevent changes to line
1673 =item Controlling whether perltidy breaks before or after operators
1675 Two command line parameters provide some control over whether
1676 a line break should be before or after specific token types.
1678 B<-wba=s> or B<--want-break-after=s>, and
1680 B<-wbb=s> or B<--want-break-before=s>.
1682 These parameters are each followed by a quoted string, B<s>, containing
1683 a list of token types (separated only by spaces). No more than one of each
1684 of these parameters should be specified, because repeating a
1685 command-line parameter always overwrites the previous one before
1686 perltidy ever sees it.
1688 By default, perltidy breaks B<after> these token types:
1689 % + - * / x != == >= <= =~ !~ < > | &
1690 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
1692 And perltidy breaks B<before> these token types by default:
1695 To illustrate, to cause a break after a concatenation operator, C<'.'>,
1696 rather than before it, the command line would be
1700 As another example, the following command would cause a break before
1701 math operators C<'+'>, C<'-'>, C<'/'>, and C<'*'>:
1705 These commands should work well for most of the token types that perltidy uses
1706 (use B<--dump-token-types> for a list). Also try the B<-D> flag on a short
1707 snippet of code and look at the .DEBUG file to see the tokenization. However,
1708 for a few token types there may be conflicts with hardwired logic which cause
1709 unexpected results. One example is curly braces, which should be controlled
1710 with the parameter B<bl> provided for that purpose.
1712 B<WARNING> Be sure to put these tokens in quotes to avoid having them
1713 misinterpreted by your command shell.
1717 =head2 Controlling List Formatting
1719 Perltidy attempts to place comma-separated arrays of values in tables
1720 which look good. Its default algorithms usually work well, and they
1721 have been improving with each release, but several parameters are
1722 available to control list formatting.
1726 =item B<-boc>, B<--break-at-old-comma-breakpoints>
1728 This flag tells perltidy to try to break at all old commas. This is not
1729 the default. Normally, perltidy makes a best guess at list formatting,
1730 and seldom uses old comma breakpoints. Usually this works well,
1739 The default formatting will flatten this down to one line:
1741 # perltidy (default)
1742 my @list = ( 1, 1, 1, 1, 2, 1, 1, 3, 3, 1, 1, 4, 6, 4, 1, );
1744 which hides the structure. Using B<-boc>, plus additional flags
1745 to retain the original style, yields
1747 # perltidy -boc -lp -pt=2 -vt=1 -vtc=1
1754 A disadvantage of this flag is that all tables in the file
1755 must already be nicely formatted. For another possibility see
1756 the -fs flag in L<Skipping Selected Sections of Code>.
1758 =item B<-mft=n>, B<--maximum-fields-per-table=n>
1760 If the computed number of fields for any table exceeds B<n>, then it
1761 will be reduced to B<n>. The default value for B<n> is a large number,
1762 40. While this value should probably be left unchanged as a general
1763 rule, it might be used on a small section of code to force a list to
1764 have a particular number of fields per line, and then either the B<-boc>
1765 flag could be used to retain this formatting, or a single comment could
1766 be introduced somewhere to freeze the formatting in future applications
1779 =item B<-cab=n>, B<--comma-arrow-breakpoints=n>
1781 A comma which follows a comma arrow, '=>', requires special
1782 consideration. In a long list, it is common to break at all such
1783 commas. This parameter can be used to control how perltidy breaks at
1784 these commas. (However, it will have no effect if old comma breaks are
1785 being forced because B<-boc> is used). The possible values of B<n> are:
1787 n=0 break at all commas after =>
1788 n=1 stable: break at all commas after => unless this would break
1789 an existing one-line container (default)
1790 n=2 break at all commas after =>, but try to form the maximum
1791 maximum one-line container lengths
1792 n=3 do not treat commas after => specially at all
1794 For example, given the following single line, perltidy by default will
1795 not add any line breaks because it would break the existing one-line
1798 bless { B => $B, Root => $Root } => $package;
1800 Using B<-cab=0> will force a break after each comma-arrow item:
1808 If perltidy is subsequently run with this container broken, then by
1809 default it will break after each '=>' because the container is now
1810 broken. To reform a one-line container, the parameter B<-cab=2> would
1813 The flag B<-cab=3> can be used to prevent these commas from being
1814 treated specially. In this case, an item such as "01" => 31 is
1815 treated as a single item in a table. The number of fields in this table
1816 will be determined by the same rules that are used for any other table.
1821 "01" => 31, "02" => 29, "03" => 31, "04" => 30,
1822 "05" => 31, "06" => 30, "07" => 31, "08" => 31,
1823 "09" => 30, "10" => 31, "11" => 30, "12" => 31
1828 =head2 Retaining or Ignoring Existing Line Breaks
1830 Several additional parameters are available for controlling the extent
1831 to which line breaks in the input script influence the output script.
1832 In most cases, the default parameter values are set so that, if a choice
1833 is possible, the output style follows the input style. For example, if
1834 a short logical container is broken in the input script, then the
1835 default behavior is for it to remain broken in the output script.
1837 Most of the parameters in this section would only be required for a
1838 one-time conversion of a script from short container lengths to longer
1839 container lengths. The opposite effect, of converting long container
1840 lengths to shorter lengths, can be obtained by temporarily using a short
1841 maximum line length.
1845 =item B<-bol>, B<--break-at-old-logical-breakpoints>
1847 By default, if a logical expression is broken at a C<&&>, C<||>, C<and>,
1848 or C<or>, then the container will remain broken. Also, breaks
1849 at internal keywords C<if> and C<unless> will normally be retained.
1850 To prevent this, and thus form longer lines, use B<-nbol>.
1852 =item B<-bok>, B<--break-at-old-keyword-breakpoints>
1854 By default, perltidy will retain a breakpoint before keywords which may
1855 return lists, such as C<sort> and <map>. This allows chains of these
1856 operators to be displayed one per line. Use B<-nbok> to prevent
1857 retaining these breakpoints.
1859 =item B<-bot>, B<--break-at-old-ternary-breakpoints>
1861 By default, if a conditional (ternary) operator is broken at a C<:>,
1862 then it will remain broken. To prevent this, and thereby
1863 form longer lines, use B<-nbot>.
1865 =item B<-iob>, B<--ignore-old-breakpoints>
1867 Use this flag to tell perltidy to ignore existing line breaks to the
1868 maximum extent possible. This will tend to produce the longest possible
1869 containers, regardless of type, which do not exceed the line length
1874 =head2 Blank Line Control
1876 Blank lines can improve the readability of a script if they are carefully
1877 placed. Perltidy has several commands for controlling the insertion,
1878 retention, and removal of blank lines.
1882 =item B<-bbc>, B<--blanks-before-comments>
1884 A blank line will be introduced before a full-line comment. This is the
1885 default. Use B<-nbbc> or B<--noblanks-before-comments> to prevent
1886 such blank lines from being introduced.
1888 =item B<-bbs>, B<--blanks-before-subs>
1890 A blank line will be introduced before a B<sub> definition, unless it is a
1891 one-liner or preceded by a comment. A blank line will also be introduced
1892 before a B<package> statement and a B<BEGIN> and B<END> block. This is the
1893 default. The intention is to help display the structure of a program by
1894 setting off certain key sections of code. This is negated with B<-nbbs> or
1895 B<--noblanks-before-subs>.
1897 =item B<-bbb>, B<--blanks-before-blocks>
1899 A blank line will be introduced before blocks of coding delimited by
1900 B<for>, B<foreach>, B<while>, B<until>, and B<if>, B<unless>, in the following
1907 The block is not preceded by a comment.
1911 The block is not a one-line block.
1915 The number of consecutive non-blank lines at the current indentation depth is at least B<-lbl>
1920 This is the default. The intention of this option is to introduce
1921 some space within dense coding.
1922 This is negated with B<-nbbb> or B<--noblanks-before-blocks>.
1924 =item B<-lbl=n> B<--long-block-line-count=n>
1926 This controls how often perltidy is allowed to add blank lines before
1927 certain block types (see previous section). The default is 8. Entering
1928 a value of B<0> is equivalent to entering a very large number.
1930 =item B<-mbl=n> B<--maximum-consecutive-blank-lines=n>
1932 This parameter specifies the maximum number of consecutive blank lines
1933 in the output script. The default is n=1. If the input file has more
1934 than n consecutive blank lines, the number will be reduced to n.
1935 (This obviously does not apply to pod sections, here-documents, and quotes).
1937 =item B<-sob>, B<--swallow-optional-blank-lines>
1939 All blank lines not required by the above flags, B<-bbb>, B<-bbs>, and B<-bbc>,
1940 will be deleted. (But essential blank lines above pod documents will be
1941 retained). This is NOT the default.
1943 =item B<-nsob>, B<--noswallow-optional-blank-lines>
1945 Retain blank lines, including those which do not corresponding to flags
1946 B<-bbb>, B<-bbs>, and B<-bbc>. This is the default. The number of
1947 blanks retained is subject to the limit imposed by
1948 B<--maximum-consecutive-blank-lines>, however.
1954 A style refers to a convenient collection of existing parameters.
1958 =item B<-gnu>, B<--gnu-style>
1960 B<-gnu> gives an approximation to the GNU Coding Standards (which do
1961 not apply to perl) as they are sometimes implemented. At present, this
1962 style overrides the default style with the following parameters:
1964 -lp -bl -noll -pt=2 -bt=2 -sbt=2 -icp
1966 =item B<-pbp>, B<--perl-best-practices>
1968 B<-pbp> is an abbreviation for the parameters in the book B<Perl Best Practices>
1971 -l=78 -i=4 -ci=4 -st -se -vt=2 -cti=0 -pt=1 -bt=1 -sbt=1 -bbt=1 -nsfs -nolq
1972 -wbb="% + - * / x != == >= <= =~ !~ < > | & =
1973 **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x="
1975 Note that the -st and -se flags make perltidy act as a filter on one file only.
1976 These can be overridden with -nst and -nse if necessary.
1980 =head2 Other Controls
1984 =item Deleting selected text
1986 Perltidy can selectively delete comments and/or pod documentation. The
1987 command B<-dac> or B<--delete-all-comments> will delete all comments
1988 B<and> all pod documentation, leaving just code and any leading system
1991 The command B<-dp> or B<--delete-pod> will remove all pod documentation
1994 Two commands which remove comments (but not pod) are: B<-dbc> or
1995 B<--delete-block-comments> and B<-dsc> or B<--delete-side-comments>.
1996 (Hanging side comments will be deleted with block comments here.)
1998 The negatives of these commands also work, and are the defaults. When
1999 block comments are deleted, any leading 'hash-bang' will be retained.
2000 Also, if the B<-x> flag is used, any system commands before a leading
2001 hash-bang will be retained (even if they are in the form of comments).
2003 =item Writing selected text to a file
2005 When perltidy writes a formatted text file, it has the ability to also
2006 send selected text to a file with a F<.TEE> extension. This text can
2007 include comments and pod documentation.
2009 The command B<-tac> or B<--tee-all-comments> will write all comments
2010 B<and> all pod documentation.
2012 The command B<-tp> or B<--tee-pod> will write all pod documentation (but
2015 The commands which write comments (but not pod) are: B<-tbc> or
2016 B<--tee-block-comments> and B<-tsc> or B<--tee-side-comments>.
2017 (Hanging side comments will be written with block comments here.)
2019 The negatives of these commands also work, and are the defaults.
2021 =item Using a F<.perltidyrc> command file
2023 If you use perltidy frequently, you probably won't be happy until you
2024 create a F<.perltidyrc> file to avoid typing commonly-used parameters.
2025 Perltidy will first look in your current directory for a command file
2026 named F<.perltidyrc>. If it does not find one, it will continue looking
2027 for one in other standard locations.
2029 These other locations are system-dependent, and may be displayed with
2030 the command C<perltidy -dpro>. Under Unix systems, it will look for a
2031 F<.perltidyrc> file in the home directory, and then for a system-wide
2032 file F</usr/local/etc/perltidyrc>, and then it will look for
2033 F</etc/perltidyrc>. Note that these last two system-wide files do not
2034 have a leading dot. Further system-dependent information will be found
2035 in the INSTALL file distributed with perltidy.
2037 This file is free format, and simply a list of parameters, just as they
2038 would be entered on a command line. Any number of lines may be used,
2039 with any number of parameters per line, although it may be easiest to
2040 read with one parameter per line. Blank lines are ignored, and text
2041 after a '#' is ignored to the end of a line.
2043 Here is an example of a F<.perltidyrc> file:
2045 # This is a simple of a .perltidyrc configuration file
2046 # This implements a highly spaced style
2047 -se # errors to standard error output
2048 -w # show all warnings
2049 -bl # braces on new lines
2050 -pt=0 # parens not tight at all
2051 -bt=0 # braces not tight
2052 -sbt=0 # square brackets not tight
2054 The parameters in the F<.perltidyrc> file are installed first, so any
2055 parameters given on the command line will have priority over them.
2057 To avoid confusion, perltidy ignores any command in the .perltidyrc
2058 file which would cause some kind of dump and an exit. These are:
2060 -h -v -ddf -dln -dop -dsn -dtt -dwls -dwrs -ss
2062 There are several options may be helpful in debugging a F<.perltidyrc>
2069 A very helpful command is B<--dump-profile> or B<-dpro>. It writes a
2070 list of all configuration filenames tested to standard output, and
2071 if a file is found, it dumps the content to standard output before
2072 exiting. So, to find out where perltidy looks for its configuration
2073 files, and which one if any it selects, just enter
2079 It may be simplest to develop and test configuration files with
2080 alternative names, and invoke them with B<-pro=filename> on the command
2081 line. Then rename the desired file to F<.perltidyrc> when finished.
2085 The parameters in the F<.perltidyrc> file can be switched off with
2086 the B<-npro> option.
2090 The commands B<--dump-options>, B<--dump-defaults>, B<--dump-long-names>,
2091 and B<--dump-short-names>, all described below, may all be helpful.
2095 =item Creating a new abbreviation
2097 A special notation is available for use in a F<.perltidyrc> file
2098 for creating an abbreviation for a group
2099 of options. This can be used to create a
2100 shorthand for one or more styles which are frequently, but not always,
2101 used. The notation is to group the options within curly braces which
2102 are preceded by the name of the alias (without leading dashes), like this:
2109 where B<newword> is the abbreviation, and B<opt1>, etc, are existing parameters
2110 I<or other abbreviations>. The main syntax requirement is that
2111 the new abbreviation must begin on a new line.
2112 Space before and after the curly braces is optional.
2114 specific example, the following line
2116 airy {-bl -pt=0 -bt=0 -sbt=0}
2118 could be placed in a F<.perltidyrc> file, and then invoked at will with
2120 perltidy -airy somefile.pl
2122 (Either C<-airy> or C<--airy> may be used).
2124 =item Skipping leading non-perl commands with B<-x> or B<--look-for-hash-bang>
2126 If your script has leading lines of system commands or other text which
2127 are not valid perl code, and which are separated from the start of the
2128 perl code by a "hash-bang" line, ( a line of the form C<#!...perl> ),
2129 you must use the B<-x> flag to tell perltidy not to parse and format any
2130 lines before the "hash-bang" line. This option also invokes perl with a
2131 -x flag when checking the syntax. This option was originally added to
2132 allow perltidy to parse interactive VMS scripts, but it should be used
2133 for any script which is normally invoked with C<perl -x>.
2135 =item Making a file unreadable
2137 The goal of perltidy is to improve the readability of files, but there
2138 are two commands which have the opposite effect, B<--mangle> and
2139 B<--extrude>. They are actually
2140 merely aliases for combinations of other parameters. Both of these
2141 strip all possible whitespace, but leave comments and pod documents,
2142 so that they are essentially reversible. The
2143 difference between these is that B<--mangle> puts the fewest possible
2144 line breaks in a script while B<--extrude> puts the maximum possible.
2145 Note that these options do not provided any meaningful obfuscation, because
2146 perltidy can be used to reformat the files. They were originally
2147 developed to help test the tokenization logic of perltidy, but they
2149 One use for B<--mangle> is the following:
2151 perltidy --mangle myfile.pl -st | perltidy -o myfile.pl.new
2153 This will form the maximum possible number of one-line blocks (see next
2154 section), and can sometimes help clean up a badly formatted script.
2156 A similar technique can be used with B<--extrude> instead of B<--mangle>
2157 to make the minimum number of one-line blocks.
2159 Another use for B<--mangle> is to combine it with B<-dac> to reduce
2160 the file size of a perl script.
2162 =item One-line blocks
2164 There are a few points to note regarding one-line blocks. A one-line
2165 block is something like this,
2167 if ($x > 0) { $y = 1 / $x }
2169 where the contents within the curly braces is short enough to fit
2172 With few exceptions, perltidy retains existing one-line blocks, if it
2173 is possible within the line-length constraint, but it does not attempt
2174 to form new ones. In other words, perltidy will try to follow the
2175 one-line block style of the input file.
2177 If an existing one-line block is longer than the maximum line length,
2178 however, it will be broken into multiple lines. When this happens, perltidy
2179 checks for and adds any optional terminating semicolon (unless the B<-nasc>
2180 option is used) if the block is a code block.
2182 The main exception is that perltidy will attempt to form new one-line
2183 blocks following the keywords C<map>, C<eval>, and C<sort>, because
2184 these code blocks are often small and most clearly displayed in a single
2187 One-line block rules can conflict with the cuddled-else option. When
2188 the cuddled-else option is used, perltidy retains existing one-line
2189 blocks, even if they do not obey cuddled-else formatting.
2191 Occasionally, when one-line blocks get broken because they exceed the
2192 available line length, the formatting will violate the requested brace style.
2193 If this happens, reformatting the script a second time should correct
2198 The following flags are available for debugging:
2200 B<--dump-defaults> or B<-ddf> will write the default option set to standard output and quit
2202 B<--dump-profile> or B<-dpro> will write the name of the current
2203 configuration file and its contents to standard output and quit.
2205 B<--dump-options> or B<-dop> will write current option set to standard
2208 B<--dump-long-names> or B<-dln> will write all command line long names (passed
2209 to Get_options) to standard output and quit.
2211 B<--dump-short-names> or B<-dsn> will write all command line short names
2212 to standard output and quit.
2214 B<--dump-token-types> or B<-dtt> will write a list of all token types
2215 to standard output and quit.
2217 B<--dump-want-left-space> or B<-dwls> will write the hash %want_left_space
2218 to standard output and quit. See the section on controlling whitespace
2221 B<--dump-want-right-space> or B<-dwrs> will write the hash %want_right_space
2222 to standard output and quit. See the section on controlling whitespace
2225 B<-DEBUG> will write a file with extension F<.DEBUG> for each input file
2226 showing the tokenization of all lines of code.
2228 =item Working with MakeMaker, AutoLoader and SelfLoader
2230 The first $VERSION line of a file which might be eval'd by MakeMaker
2231 is passed through unchanged except for indentation.
2232 Use B<--nopass-version-line>, or B<-npvl>, to deactivate this feature.
2234 If the AutoLoader module is used, perltidy will continue formatting
2235 code after seeing an __END__ line.
2236 Use B<--nolook-for-autoloader>, or B<-nlal>, to deactivate this feature.
2238 Likewise, if the SelfLoader module is used, perltidy will continue formatting
2239 code after seeing a __DATA__ line.
2240 Use B<--nolook-for-selfloader>, or B<-nlsl>, to deactivate this feature.
2242 =item Working around problems with older version of Perl
2244 Perltidy contains a number of rules which help avoid known subtleties
2245 and problems with older versions of perl, and these rules always
2246 take priority over whatever formatting flags have been set. For example,
2247 perltidy will usually avoid starting a new line with a bareword, because
2248 this might cause problems if C<use strict> is active.
2250 There is no way to override these rules.
2258 =item The B<-html> master switch
2260 The flag B<-html> causes perltidy to write an html file with extension
2261 F<.html>. So, for example, the following command
2263 perltidy -html somefile.pl
2265 will produce a syntax-colored html file named F<somefile.pl.html>
2266 which may be viewed with a browser.
2268 B<Please Note>: In this case, perltidy does not do any formatting to the
2269 input file, and it does not write a formatted file with extension
2270 F<.tdy>. This means that two perltidy runs are required to create a
2271 fully reformatted, html copy of a script.
2273 =item The B<-pre> flag for code snippets
2275 When the B<-pre> flag is given, only the pre-formatted section, within
2276 the <PRE> and </PRE> tags, will be output. This simplifies inclusion
2277 of the output in other files. The default is to output a complete
2280 =item The B<-nnn> flag for line numbering
2282 When the B<-nnn> flag is given, the output lines will be numbered.
2284 =item The B<-toc>, or B<--html-table-of-contents> flag
2286 By default, a table of contents to packages and subroutines will be
2287 written at the start of html output. Use B<-ntoc> to prevent this.
2288 This might be useful, for example, for a pod document which contains a
2289 number of unrelated code snippets. This flag only influences the code
2290 table of contents; it has no effect on any table of contents produced by
2291 pod2html (see next item).
2293 =item The B<-pod>, or B<--pod2html> flag
2295 There are two options for formatting pod documentation. The default is
2296 to pass the pod through the Pod::Html module (which forms the basis of
2297 the pod2html utility). Any code sections are formatted by perltidy, and
2298 the results then merged. Note: perltidy creates a temporary file when
2299 Pod::Html is used; see L<"FILES">. Also, Pod::Html creates temporary
2300 files for its cache.
2302 NOTE: Perltidy counts the number of C<=cut> lines, and either moves the
2303 pod text to the top of the html file if there is one C<=cut>, or leaves
2304 the pod text in its original order (interleaved with code) otherwise.
2306 Most of the flags accepted by pod2html may be included in the perltidy
2307 command line, and they will be passed to pod2html. In some cases,
2308 the flags have a prefix C<pod> to emphasize that they are for the
2309 pod2html, and this prefix will be removed before they are passed to
2310 pod2html. The flags which have the additional C<pod> prefix are:
2312 --[no]podheader --[no]podindex --[no]podrecurse --[no]podquiet
2313 --[no]podverbose --podflush
2315 The flags which are unchanged from their use in pod2html are:
2317 --backlink=s --cachedir=s --htmlroot=s --libpods=s --title=s
2318 --podpath=s --podroot=s
2320 where 's' is an appropriate character string. Not all of these flags are
2321 available in older versions of Pod::Html. See your Pod::Html documentation for
2324 The alternative, indicated with B<-npod>, is not to use Pod::Html, but
2325 rather to format pod text in italics (or whatever the stylesheet
2326 indicates), without special html markup. This is useful, for example,
2327 if pod is being used as an alternative way to write comments.
2329 =item The B<-frm>, or B<--frames> flag
2331 By default, a single html output file is produced. This can be changed
2332 with the B<-frm> option, which creates a frame holding a table of
2333 contents in the left panel and the source code in the right side. This
2334 simplifies code browsing. Assume, for example, that the input file is
2335 F<MyModule.pm>. Then, for default file extension choices, these three
2336 files will be created:
2338 MyModule.pm.html - the frame
2339 MyModule.pm.toc.html - the table of contents
2340 MyModule.pm.src.html - the formatted source code
2342 Obviously this file naming scheme requires that output be directed to a real
2343 file (as opposed to, say, standard output). If this is not the
2344 case, or if the file extension is unknown, the B<-frm> option will be
2347 =item The B<-text=s>, or B<--html-toc-extension> flag
2349 Use this flag to specify the extra file extension of the table of contents file
2350 when html frames are used. The default is "toc".
2351 See L<Specifying File Extensions>.
2353 =item The B<-sext=s>, or B<--html-src-extension> flag
2355 Use this flag to specify the extra file extension of the content file when html
2356 frames are used. The default is "src".
2357 See L<Specifying File Extensions>.
2359 =item The B<-hent>, or B<--html-entities> flag
2361 This flag controls the use of Html::Entities for html formatting. By
2362 default, the module Html::Entities is used to encode special symbols.
2363 This may not be the right thing for some browser/language
2364 combinations. Use --nohtml-entities or -nhent to prevent this.
2368 Style sheets make it very convenient to control and adjust the
2369 appearance of html pages. The default behavior is to write a page of
2370 html with an embedded style sheet.
2372 An alternative to an embedded style sheet is to create a page with a
2373 link to an external style sheet. This is indicated with the
2374 B<-css=filename>, where the external style sheet is F<filename>. The
2375 external style sheet F<filename> will be created if and only if it does
2376 not exist. This option is useful for controlling multiple pages from a
2379 To cause perltidy to write a style sheet to standard output and exit,
2380 use the B<-ss>, or B<--stylesheet>, flag. This is useful if the style
2381 sheet could not be written for some reason, such as if the B<-pre> flag
2382 was used. Thus, for example,
2384 perltidy -html -ss >mystyle.css
2386 will write a style sheet with the default properties to file
2389 The use of style sheets is encouraged, but a web page without a style
2390 sheets can be created with the flag B<-nss>. Use this option if you
2391 must to be sure that older browsers (roughly speaking, versions prior to
2392 4.0 of Netscape Navigator and Internet Explorer) can display the
2393 syntax-coloring of the html files.
2395 =item Controlling HTML properties
2397 Note: It is usually more convenient to accept the default properties
2398 and then edit the stylesheet which is produced. However, this section
2399 shows how to control the properties with flags to perltidy.
2401 Syntax colors may be changed from their default values by flags of the either
2402 the long form, B<-html-color-xxxxxx=n>, or more conveniently the short form,
2403 B<-hcx=n>, where B<xxxxxx> is one of the following words, and B<x> is the
2404 corresponding abbreviation:
2407 ---------- -------- --
2410 identifier identifier i
2411 bareword, function bareword w
2413 quite, pattern quote q
2414 here doc text here-doc-text h
2415 here doc target here-doc-target hh
2416 punctuation punctuation pu
2418 structural braces structure s
2419 semicolon semicolon sc
2423 sub definition name subroutine m
2424 pod text pod-text pd
2426 A default set of colors has been defined, but they may be changed by providing
2427 values to any of the following parameters, where B<n> is either a 6 digit
2428 hex RGB color value or an ascii name for a color, such as 'red'.
2430 To illustrate, the following command will produce an html
2431 file F<somefile.pl.html> with "aqua" keywords:
2433 perltidy -html -hck=00ffff somefile.pl
2435 and this should be equivalent for most browsers:
2437 perltidy -html -hck=aqua somefile.pl
2439 Perltidy merely writes any non-hex names that it sees in the html file.
2440 The following 16 color names are defined in the HTML 3.2 standard:
2459 Many more names are supported in specific browsers, but it is safest
2460 to use the hex codes for other colors. Helpful color tables can be
2461 located with an internet search for "HTML color tables".
2463 Besides color, two other character attributes may be set: bold, and italics.
2464 To set a token type to use bold, use the flag
2465 B<--html-bold-xxxxxx> or B<-hbx>, where B<xxxxxx> or B<x> are the long
2466 or short names from the above table. Conversely, to set a token type to
2467 NOT use bold, use B<--nohtml-bold-xxxxxx> or B<-nhbx>.
2469 Likewise, to set a token type to use an italic font, use the flag
2470 B<--html-italic-xxxxxx> or B<-hix>, where again B<xxxxxx> or B<x> are the
2471 long or short names from the above table. And to set a token type to
2472 NOT use italics, use B<--nohtml-italic-xxxxxx> or B<-nhix>.
2474 For example, to use bold braces and lime color, non-bold, italics keywords the
2475 following command would be used:
2477 perltidy -html -hbs -hck=00FF00 -nhbk -hik somefile.pl
2479 The background color can be specified with B<--html-color-background=n>,
2480 or B<-hcbg=n> for short, where n is a 6 character hex RGB value. The
2481 default color of text is the value given to B<punctuation>, which is
2484 Here are some notes and hints:
2486 1. If you find a preferred set of these parameters, you may want
2487 to create a F<.perltidyrc> file containing them. See the perltidy man
2488 page for an explanation.
2490 2. Rather than specifying values for these parameters, it is probably
2491 easier to accept the defaults and then edit a style sheet. The style
2492 sheet contains comments which should make this easy.
2494 3. The syntax-colored html files can be very large, so it may be best to
2495 split large files into smaller pieces to improve download times.
2499 =head1 SOME COMMON INPUT CONVENTIONS
2501 =head2 Specifying Block Types
2503 Several parameters which refer to code block types may be customized by also
2504 specifying an associated list of block types. The type of a block is the name
2505 of the keyword which introduces that block, such as B<if>, B<else>, or B<sub>.
2506 An exception is a labeled block, which has no keyword, and should be specified
2509 For example, the following parameter specifies C<sub>, labels, C<BEGIN>, and
2512 -cscl="sub : BEGIN END"
2514 (the meaning of the -cscl parameter is described above.) Note that
2515 quotes are required around the list of block types because of the
2518 =head2 Specifying File Extensions
2520 Several parameters allow default file extensions to be overridden. For
2521 example, a backup file extension may be specified with B<-bext=ext>,
2522 where B<ext> is some new extension. In order to provides the user some
2523 flexibility, the following convention is used in all cases to decide if
2524 a leading '.' should be used. If the extension C<ext> begins with
2525 C<A-Z>, C<a-z>, or C<0-9>, then it will be appended to the filename with
2526 an intermediate '.' (or perhaps an '_' on VMS systems). Otherwise, it
2527 will be appended directly.
2529 For example, suppose the file is F<somefile.pl>. For C<-bext=old>, a '.' is
2530 added to give F<somefile.pl.old>. For C<-bext=.old>, no additional '.' is
2531 added, so again the backup file is F<somefile.pl.old>. For C<-bext=~>, then no
2532 dot is added, and the backup file will be F<somefile.pl~> .
2534 =head1 SWITCHES WHICH MAY BE NEGATED
2536 The following list shows all short parameter names which allow a prefix
2537 'n' to produce the negated form:
2539 D anl asc aws b bbb bbc bbs bl bli boc bok bol bot ce
2540 csc dac dbc dcsc ddf dln dnl dop dp dpro dsc dsm dsn dtt dwls
2541 dwrs dws f fll frm fs hsc html ibc icb icp iob isbc lal log
2542 lp lsl ohbr okw ola oll opr opt osbr otr ple ple pod pvl q
2543 sbc sbl schb scp scsb sct se sfp sfs skp sob sohb sop sosb sot
2544 ssc st sts syn t tac tbc toc tp tqw tsc w x bar
2546 Equivalently, the prefix 'no' or 'no-' on the corresponding long names may be
2553 =item Parsing Limitations
2555 Perltidy should work properly on most perl scripts. It does a lot of
2556 self-checking, but still, it is possible that an error could be
2557 introduced and go undetected. Therefore, it is essential to make
2558 careful backups and to test reformatted scripts.
2560 The main current limitation is that perltidy does not scan modules
2561 included with 'use' statements. This makes it necessary to guess the
2562 context of any bare words introduced by such modules. Perltidy has good
2563 guessing algorithms, but they are not infallible. When it must guess,
2564 it leaves a message in the log file.
2566 If you encounter a bug, please report it.
2568 =item What perltidy does not parse and format
2570 Perltidy indents but does not reformat comments and C<qw> quotes.
2571 Perltidy does not in any way modify the contents of here documents or
2572 quoted text, even if they contain source code. (You could, however,
2573 reformat them separately). Perltidy does not format 'format' sections
2574 in any way. And, of course, it does not modify pod documents.
2582 =item Temporary files
2584 Under the -html option with the default --pod2html flag, a temporary file is
2585 required to pass text to Pod::Html. Unix systems will try to use the POSIX
2586 tmpnam() function. Otherwise the file F<perltidy.TMP> will be temporarily
2587 created in the current working directory.
2589 =item Special files when standard input is used
2591 When standard input is used, the log file, if saved, is F<perltidy.LOG>,
2592 and any errors are written to F<perltidy.ERR> unless the B<-se> flag is
2593 set. These are saved in the current working directory.
2595 =item Files overwritten
2597 The following file extensions are used by perltidy, and files with these
2598 extensions may be overwritten or deleted: F<.ERR>, F<.LOG>, F<.TEE>,
2599 and/or F<.tdy>, F<.html>, and F<.bak>, depending on the run type and
2602 =item Files extensions limitations
2604 Perltidy does not operate on files for which the run could produce a file with
2605 a duplicated file extension. These extensions include F<.LOG>, F<.ERR>,
2606 F<.TEE>, and perhaps F<.tdy> and F<.bak>, depending on the run type. The
2607 purpose of this rule is to prevent generating confusing filenames such as
2608 F<somefile.tdy.tdy.tdy>.
2614 perlstyle(1), Perl::Tidy(3)
2618 This man page documents perltidy version 20070424.
2622 Michael Cartmell supplied code for adaptation to VMS and helped with
2625 Yves Orton supplied code for adaptation to the various versions
2628 Axel Rose supplied a patch for MacPerl.
2630 Hugh S. Myers designed and implemented the initial Perl::Tidy module interface.
2632 Many others have supplied key ideas, suggestions, and bug reports;
2633 see the CHANGES file.
2638 email: perltidy at users.sourceforge.net
2639 http://perltidy.sourceforge.net
2643 Copyright (c) 2000-2006 by Steve Hancock
2647 This package is free software; you can redistribute it and/or modify it
2648 under the terms of the "GNU General Public License".
2650 Please refer to the file "COPYING" for details.
2654 This package is distributed in the hope that it will be useful,
2655 but WITHOUT ANY WARRANTY; without even the implied warranty of
2656 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
2658 See the "GNU General Public License" for more details.
projects / perltidy.git / blob