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<-fpsc=n>, B<--fixed-position-side-comment=n>
893 This parameter tells perltidy to line up side comments in column number B<n>
894 whenever possible. The default, n=0, is not do do this.
896 =item B<-hsc>, B<--hanging-side-comments>
898 By default, perltidy tries to identify and align "hanging side
899 comments", which are something like this:
901 my $IGNORE = 0; # This is a side comment
902 # This is a hanging side comment
905 A comment is considered to be a hanging side comment if (1) it immediately
906 follows a line with a side comment, or another hanging side comment, and
907 (2) there is some leading whitespace on the line.
908 To deactivate this feature, use B<-nhsc> or B<--nohanging-side-comments>.
909 If block comments are preceded by a blank line, or have no leading
910 whitespace, they will not be mistaken as hanging side comments.
912 =item Closing Side Comments
914 A closing side comment is a special comment which perltidy can
915 automatically create and place after the closing brace of a code block.
916 They can be useful for code maintenance and debugging. The command
917 B<-csc> (or B<--closing-side-comments>) adds or updates closing side
918 comments. For example, here is a small code snippet
921 if ( !defined( $_[0] ) ) {
922 print("Hello, World\n");
925 print( $_[0], "\n" );
929 And here is the result of processing with C<perltidy -csc>:
932 if ( !defined( $_[0] ) ) {
933 print("Hello, World\n");
936 print( $_[0], "\n" );
940 A closing side comment was added for C<sub message> in this case, but not
941 for the C<if> and C<else> blocks, because they were below the 6 line
942 cutoff limit for adding closing side comments. This limit may be
943 changed with the B<-csci> command, described below.
945 The command B<-dcsc> (or B<--delete-closing-side-comments>) reverses this
946 process and removes these comments.
948 Several commands are available to modify the behavior of these two basic
949 commands, B<-csc> and B<-dcsc>:
953 =item B<-csci=n>, or B<--closing-side-comment-interval=n>
955 where C<n> is the minimum number of lines that a block must have in
956 order for a closing side comment to be added. The default value is
957 C<n=6>. To illustrate:
959 # perltidy -csci=2 -csc
961 if ( !defined( $_[0] ) ) {
962 print("Hello, World\n");
963 } ## end if ( !defined( $_[0] ))
965 print( $_[0], "\n" );
966 } ## end else [ if ( !defined( $_[0] ))
969 Now the C<if> and C<else> blocks are commented. However, now this has
970 become very cluttered.
972 =item B<-cscp=string>, or B<--closing-side-comment-prefix=string>
974 where string is the prefix used before the name of the block type. The
975 default prefix, shown above, is C<## end>. This string will be added to
976 closing side comments, and it will also be used to recognize them in
977 order to update, delete, and format them. Any comment identified as a
978 closing side comment will be placed just a single space to the right of
981 =item B<-cscl=string>, or B<--closing-side-comment-list-string>
983 where C<string> is a list of block types to be tagged with closing side
984 comments. By default, all code block types preceded by a keyword or
985 label (such as C<if>, C<sub>, and so on) will be tagged. The B<-cscl>
986 command changes the default list to be any selected block types; see
987 L<Specifying Block Types>.
988 For example, the following command
989 requests that only C<sub>'s, labels, C<BEGIN>, and C<END> blocks be
990 affected by any B<-csc> or B<-dcsc> operation:
992 -cscl="sub : BEGIN END"
994 =item B<-csct=n>, or B<--closing-side-comment-maximum-text=n>
996 The text appended to certain block types, such as an C<if> block, is
997 whatever lies between the keyword introducing the block, such as C<if>,
998 and the opening brace. Since this might be too much text for a side
999 comment, there needs to be a limit, and that is the purpose of this
1000 parameter. The default value is C<n=20>, meaning that no additional
1001 tokens will be appended to this text after its length reaches 20
1002 characters. Omitted text is indicated with C<...>. (Tokens, including
1003 sub names, are never truncated, however, so actual lengths may exceed
1004 this). To illustrate, in the above example, the appended text of the
1005 first block is C< ( !defined( $_[0] )...>. The existing limit of
1006 C<n=20> caused this text to be truncated, as indicated by the C<...>.
1008 =item B<-csce=n>, or B<--closing-side-comment-else-flag=n>
1010 The default, B<n=0>, places the text of the opening C<if> statement after any
1013 If B<n=2> is used, then each C<elsif> is also given the text of the opening
1014 C<if> statement. Also, an C<else> will include the text of a preceding
1015 C<elsif> statement. Note that this may result some long closing
1018 If B<n=1> is used, the results will be the same as B<n=2> whenever the
1019 resulting line length is less than the maximum allowed.
1021 =item B<-cscw>, or B<--closing-side-comment-warnings>
1023 This parameter is intended to help make the initial transition to the use of
1024 closing side comments.
1026 things to happen if a closing side comment replaces an existing, different
1027 closing side comment: first, an error message will be issued, and second, the
1028 original side comment will be placed alone on a new specially marked comment
1029 line for later attention.
1031 The intent is to avoid clobbering existing hand-written side comments
1032 which happen to match the pattern of closing side comments. This flag
1033 should only be needed on the first run with B<-csc>.
1037 B<Important Notes on Closing Side Comments:>
1043 Closing side comments are only placed on lines terminated with a closing
1044 brace. Certain closing styles, such as the use of cuddled elses
1045 (B<-ce>), preclude the generation of some closing side comments.
1049 Please note that adding or deleting of closing side comments takes
1050 place only through the commands B<-csc> or B<-dcsc>. The other commands,
1051 if used, merely modify the behavior of these two commands.
1055 It is recommended that the B<-cscw> flag be used along with B<-csc> on
1056 the first use of perltidy on a given file. This will prevent loss of
1057 any existing side comment data which happens to have the csc prefix.
1061 Once you use B<-csc>, you should continue to use it so that any
1062 closing side comments remain correct as code changes. Otherwise, these
1063 comments will become incorrect as the code is updated.
1067 If you edit the closing side comments generated by perltidy, you must also
1068 change the prefix to be different from the closing side comment prefix.
1069 Otherwise, your edits will be lost when you rerun perltidy with B<-csc>. For
1070 example, you could simply change C<## end> to be C<## End>, since the test is
1071 case sensitive. You may also want to use the B<-ssc> flag to keep these
1072 modified closing side comments spaced the same as actual closing side comments.
1076 Temporarily generating closing side comments is a useful technique for
1077 exploring and/or debugging a perl script, especially one written by someone
1078 else. You can always remove them with B<-dcsc>.
1082 =item Static Block Comments
1084 Static block comments are block comments with a special leading pattern,
1085 C<##> by default, which will be treated slightly differently from other
1086 block comments. They effectively behave as if they had glue along their
1087 left and top edges, because they stick to the left edge and previous line
1088 when there is no blank spaces in those places. This option is
1089 particularly useful for controlling how commented code is displayed.
1093 =item B<-sbc>, B<--static-block-comments>
1095 When B<-sbc> is used, a block comment with a special leading pattern, C<##> by
1096 default, will be treated specially.
1098 Comments so identified are treated as follows:
1104 If there is no leading space on the line, then the comment will not
1105 be indented, and otherwise it may be,
1109 no new blank line will be
1110 inserted before such a comment, and
1114 such a comment will never become
1115 a hanging side comment.
1119 For example, assuming C<@month_of_year> is
1122 @month_of_year = ( # -sbc (default)
1123 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1127 Without this convention, the above code would become
1129 @month_of_year = ( # -nsbc
1130 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1136 which is not as clear.
1137 The default is to use B<-sbc>. This may be deactivated with B<-nsbc>.
1139 =item B<-sbcp=string>, B<--static-block-comment-prefix=string>
1141 This parameter defines the prefix used to identify static block comments
1142 when the B<-sbc> parameter is set. The default prefix is C<##>,
1143 corresponding to C<-sbcp=##>. The prefix is actually part of a perl
1144 pattern used to match lines and it must either begin with C<#> or C<^#>.
1145 In the first case a prefix ^\s* will be added to match any leading
1146 whitespace, while in the second case the pattern will match only
1147 comments with no leading whitespace. For example, to
1148 identify all comments as static block comments, one would use C<-sbcp=#>.
1149 To identify all left-adjusted comments as static block comments, use C<-sbcp='^#'>.
1151 Please note that B<-sbcp> merely defines the pattern used to identify static
1152 block comments; it will not be used unless the switch B<-sbc> is set. Also,
1153 please be aware that since this string is used in a perl regular expression
1154 which identifies these comments, it must enable a valid regular expression to
1157 =item B<-osbc>, B<--outdent-static-block-comments>
1159 The command B<-osbc> will will cause static block comments to be outdented by 2
1160 spaces (or whatever B<-ci=n> has been set to), if possible.
1164 =item Static Side Comments
1166 Static side comments are side comments with a special leading pattern.
1167 This option can be useful for controlling how commented code is displayed
1168 when it is a side comment.
1172 =item B<-ssc>, B<--static-side-comments>
1174 When B<-ssc> is used, a side comment with a static leading pattern, which is
1175 C<##> by default, will be be spaced only a single space from previous
1176 character, and it will not be vertically aligned with other side comments.
1178 The default is B<-nssc>.
1180 =item B<-sscp=string>, B<--static-side-comment-prefix=string>
1182 This parameter defines the prefix used to identify static side comments
1183 when the B<-ssc> parameter is set. The default prefix is C<##>,
1184 corresponding to C<-sscp=##>.
1186 Please note that B<-sscp> merely defines the pattern used to identify
1187 static side comments; it will not be used unless the switch B<-ssc> is
1188 set. Also, note that this string is used in a perl regular expression
1189 which identifies these comments, so it must enable a valid regular
1190 expression to be formed.
1197 =head2 Skipping Selected Sections of Code
1199 Selected lines of code may be passed verbatim to the output without any
1200 formatting. This feature is enabled by default but can be disabled with
1201 the B<--noformat-skipping> or B<-nfs> flag. It should be used sparingly to
1202 avoid littering code with markers, but it might be helpful for working
1203 around occasional problems. For example it might be useful for keeping
1204 the indentation of old commented code unchanged, keeping indentation of
1205 long blocks of aligned comments unchanged, keeping certain list
1206 formatting unchanged, or working around a glitch in perltidy.
1210 =item B<-fs>, B<--format-skipping>
1212 This flag, which is enabled by default, causes any code between
1213 special beginning and ending comment markers to be passed to the
1214 output without formatting. The default beginning marker is #<<<
1215 and the default ending marker is #>>> but they
1216 may be changed (see next items below). Additional text may appear on
1217 these special comment lines provided that it is separated from the
1218 marker by at least one space. For example
1220 #<<< do not let perltidy touch this
1228 The comment markers may be placed at any location that a block comment may
1229 appear. If they do not appear to be working, use the -log flag and examine the
1230 F<.LOG> file. Use B<-nfs> to disable this feature.
1232 =item B<-fsb=string>, B<--format-skipping-begin=string>
1234 The B<-fsb=string> parameter may be used to change the beginning marker for
1235 format skipping. The default is equivalent to -fsb='#<<<'. The string that
1236 you enter must begin with a # and should be in quotes as necessary to get past
1237 the command shell of your system. It is actually the leading text of a pattern
1238 that is constructed by appending a '\s', so you must also include backslashes
1239 for characters to be taken literally rather than as patterns.
1241 Some examples show how example strings become patterns:
1243 -fsb='#\{\{\{' becomes /^#\{\{\{\s/ which matches #{{{ but not #{{{{
1244 -fsb='#\*\*' becomes /^#\*\*\s/ which matches #** but not #***
1245 -fsb='#\*{2,}' becomes /^#\*{2,}\s/ which matches #** and #*****
1247 =item B<-fse=string>, B<--format-skipping-end=string>
1249 The B<-fsb=string> is the corresponding parameter used to change the
1250 ending marker for format skipping. The default is equivalent to
1255 =head2 Line Break Control
1259 =item B<-fnl>, B<--freeze-newlines>
1261 If you do not want any changes to the line breaks in your script, set
1262 B<-fnl>, and they will remain fixed, and the rest of the commands in
1263 this section and sections
1264 L<Controlling List Formatting>,
1265 L<Retaining or Ignoring Existing Line Breaks>, and
1266 L<Blank Line Control> will be ignored. You may want to use B<-noll>
1269 =item B<-ce>, B<--cuddled-else>
1271 Enable the "cuddled else" style, in which C<else> and C<elsif> are
1272 follow immediately after the curly brace closing the previous block.
1273 The default is not to use cuddled elses, and is indicated with the flag
1274 B<-nce> or B<--nocuddled-else>. Here is a comparison of the
1286 else { # -nce (default)
1290 =item B<-bl>, B<--opening-brace-on-new-line>
1292 Use the flag B<-bl> to place the opening brace on a new line:
1294 if ( $input_file eq '-' ) # -bl
1296 important_function();
1299 This flag applies to all structural blocks, including sub's (unless
1300 the B<-sbl> flag is set -- see next item).
1302 The default style, B<-nbl>, places an opening brace on the same line as
1303 the keyword introducing it. For example,
1305 if ( $input_file eq '-' ) { # -nbl (default)
1307 =item B<-sbl>, B<--opening-sub-brace-on-new-line>
1309 The flag B<-sbl> can be used to override the value of B<-bl> for
1310 opening sub braces. For example,
1314 produces this result:
1318 if (!defined($_[0])) {
1319 print("Hello, World\n");
1326 This flag is negated with B<-nsbl>. If B<-sbl> is not specified,
1327 the value of B<-bl> is used.
1329 =item B<-bli>, B<--brace-left-and-indent>
1331 The flag B<-bli> is the same as B<-bl> but in addition it causes one
1332 unit of continuation indentation ( see B<-ci> ) to be placed before
1333 an opening and closing block braces.
1337 if ( $input_file eq '-' ) # -bli
1339 important_function();
1342 By default, this extra indentation occurs for blocks of type:
1343 B<if>, B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>,
1344 B<while>, B<until>, and also with a preceding label. The next item
1345 shows how to change this.
1347 =item B<-blil=s>, B<--brace-left-and-indent-list=s>
1349 Use this parameter to change the types of block braces for which the
1350 B<-bli> flag applies; see L<Specifying Block Types>. For example,
1351 B<-blil='if elsif else'> would apply it to only C<if/elsif/else> blocks.
1353 =item B<-bar>, B<--opening-brace-always-on-right>
1355 The default style, B<-nbl> places the opening code block brace on a new
1356 line if it does not fit on the same line as the opening keyword, like
1359 if ( $bigwasteofspace1 && $bigwasteofspace2
1360 || $bigwasteofspace3 && $bigwasteofspace4 )
1362 big_waste_of_time();
1365 To force the opening brace to always be on the right, use the B<-bar>
1366 flag. In this case, the above example becomes
1368 if ( $bigwasteofspace1 && $bigwasteofspace2
1369 || $bigwasteofspace3 && $bigwasteofspace4 ) {
1370 big_waste_of_time();
1373 A conflict occurs if both B<-bl> and B<-bar> are specified.
1375 =item B<-otr>, B<--opening-token-right> and related flags
1377 The B<-otr> flag is a hint that perltidy should not place a break between a
1378 comma and an opening token. For example:
1380 # default formatting
1381 push @{ $self->{$module}{$key} },
1383 accno => $ref->{accno},
1384 description => $ref->{description}
1388 push @{ $self->{$module}{$key} }, {
1389 accno => $ref->{accno},
1390 description => $ref->{description}
1393 The flag B<-otr> is actually a synonym for three other flags
1394 which can be used to control parens, hash braces, and square brackets
1395 separately if desired:
1397 -opr or --opening-paren-right
1398 -ohbr or --opening-hash-brace-right
1399 -osbr or --opening-square-bracket-right
1401 =item Vertical tightness of non-block curly braces, parentheses, and square brackets.
1403 These parameters control what shall be called vertical tightness. Here are the
1410 Opening tokens (except for block braces) are controlled by B<-vt=n>, or
1411 B<--vertical-tightness=n>, where
1413 -vt=0 always break a line after opening token (default).
1414 -vt=1 do not break unless this would produce more than one
1415 step in indentation in a line.
1416 -vt=2 never break a line after opening token
1420 You must also use the B<-lp> flag when you use the B<-vt> flag; the
1421 reason is explained below.
1425 Closing tokens (except for block braces) are controlled by B<-vtc=n>, or
1426 B<--vertical-tightness-closing=n>, where
1428 -vtc=0 always break a line before a closing token (default),
1429 -vtc=1 do not break before a closing token which is followed
1430 by a semicolon or another closing token, and is not in
1432 -vtc=2 never break before a closing token.
1434 The rules for B<-vtc=1> are designed to maintain a reasonable balance
1435 between tightness and readability in complex lists.
1439 Different controls may be applied to to different token types,
1440 and it is also possible to control block braces; see below.
1444 Finally, please note that these vertical tightness flags are merely
1445 hints to the formatter, and it cannot always follow them. Things which
1446 make it difficult or impossible include comments, blank lines, blocks of
1447 code within a list, and possibly the lack of the B<-lp> parameter.
1448 Also, these flags may be ignored for very small lists (2 or 3 lines in
1453 Here are some examples:
1455 # perltidy -lp -vt=0 -vtc=0
1463 # perltidy -lp -vt=1 -vtc=0
1464 %romanNumerals = ( one => 'I',
1470 # perltidy -lp -vt=1 -vtc=1
1471 %romanNumerals = ( one => 'I',
1476 The difference between B<-vt=1> and B<-vt=2> is shown here:
1478 # perltidy -lp -vt=1
1480 mysprintf( "(void)find_threadsv(%s);",
1481 cstring( $threadsv_names[ $op->targ ] )
1485 # perltidy -lp -vt=2
1486 $init->add( mysprintf( "(void)find_threadsv(%s);",
1487 cstring( $threadsv_names[ $op->targ ] )
1491 With B<-vt=1>, the line ending in C<add(> does not combine with the next
1492 line because the next line is not balanced. This can help with
1493 readability, but B<-vt=2> can be used to ignore this rule.
1495 The tightest, and least readable, code is produced with both C<-vt=2> and
1498 # perltidy -lp -vt=2 -vtc=2
1499 $init->add( mysprintf( "(void)find_threadsv(%s);",
1500 cstring( $threadsv_names[ $op->targ ] ) ) );
1502 Notice how the code in all of these examples collapses vertically as
1503 B<-vt> increases, but the indentation remains unchanged. This is
1504 because perltidy implements the B<-vt> parameter by first formatting as
1505 if B<-vt=0>, and then simply overwriting one output line on top of the
1506 next, if possible, to achieve the desired vertical tightness. The
1507 B<-lp> indentation style has been designed to allow this vertical
1508 collapse to occur, which is why it is required for the B<-vt> parameter.
1510 The B<-vt=n> and B<-vtc=n> parameters apply to each type of container
1511 token. If desired, vertical tightness controls can be applied
1512 independently to each of the closing container token types.
1514 The parameters for controlling parentheses are B<-pvt=n> or
1515 B<--paren-vertical-tightness=n>, and B<-pcvt=n> or
1516 B<--paren-vertical-tightness-closing=n>.
1518 Likewise, the parameters for square brackets are B<-sbvt=n> or
1519 B<--square-bracket-vertical-tightness=n>, and B<-sbcvt=n> or
1520 B<--square-bracket-vertical-tightness-closing=n>.
1522 Finally, the parameters for controlling non-code block braces are
1523 B<-bvt=n> or B<--brace-vertical-tightness=n>, and B<-bcvt=n> or
1524 B<--brace-vertical-tightness-closing=n>.
1526 In fact, the parameter B<-vt=n> is actually just an abbreviation for
1527 B<-pvt=n -bvt=n sbvt=n>, and likewise B<-vtc=n> is an abbreviation
1528 for B<-pvtc=n -bvtc=n sbvtc=n>.
1530 =item B<-bbvt=n> or B<--block-brace-vertical-tightness=n>
1532 The B<-bbvt=n> flag is just like the B<-vt=n> flag but applies
1533 to opening code block braces.
1535 -bbvt=0 break after opening block brace (default).
1536 -bbvt=1 do not break unless this would produce more than one
1537 step in indentation in a line.
1538 -bbvt=2 do not break after opening block brace.
1540 It is necessary to also use either B<-bl> or B<-bli> for this to work,
1541 because, as with other vertical tightness controls, it is implemented by
1542 simply overwriting a line ending with an opening block brace with the
1543 subsequent line. For example:
1545 # perltidy -bli -bbvt=0
1546 if ( open( FILE, "< $File" ) )
1548 while ( $File = <FILE> )
1556 # perltidy -bli -bbvt=1
1557 if ( open( FILE, "< $File" ) )
1558 { while ( $File = <FILE> )
1565 By default this applies to blocks associated with keywords B<if>,
1566 B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>, B<while>,
1567 B<until>, and also with a preceding label. This can be changed with
1568 the parameter B<-bbvtl=string>, or
1569 B<--block-brace-vertical-tightness-list=string>, where B<string> is a
1570 space-separated list of block types. For more information on the
1571 possible values of this string, see L<Specifying Block Types>
1573 For example, if we want to just apply this style to C<if>,
1574 C<elsif>, and C<else> blocks, we could use
1575 C<perltidy -bli -bbvt=1 -bbvtl='if elsif else'>.
1577 There is no vertical tightness control for closing block braces; with
1578 the exception of one-line blocks, they will normally remain on a
1581 =item B<-sot>, B<--stack-opening-token> and related flags
1583 The B<-sot> flag tells perltidy to "stack" opening tokens
1584 when possible to avoid lines with isolated opening tokens.
1589 $opt_c = Text::CSV_XS->new(
1598 $opt_c = Text::CSV_XS->new( {
1605 For detailed control of individual closing tokens the following
1606 controls can be used:
1608 -sop or --stack-opening-paren
1609 -sohb or --stack-opening-hash-brace
1610 -sosb or --stack-opening-square-bracket
1612 The flag B<-sot> is a synonym for B<-sop -sohb -sosb>.
1614 =item B<-sct>, B<--stack-closing-token> and related flags
1616 The B<-sct> flag tells perltidy to "stack" closing tokens
1617 when possible to avoid lines with isolated closing tokens.
1622 $opt_c = Text::CSV_XS->new(
1631 $opt_c = Text::CSV_XS->new(
1638 The B<-sct> flag is somewhat similar to the B<-vtc> flags, and in some
1639 cases it can give a similar result. The difference is that the B<-vtc>
1640 flags try to avoid lines with leading opening tokens by "hiding" them at
1641 the end of a previous line, whereas the B<-sct> flag merely tries to
1642 reduce the number of lines with isolated closing tokens by stacking them
1643 but does not try to hide them. For example:
1646 $opt_c = Text::CSV_XS->new(
1650 always_quote => 1, } );
1652 For detailed control of the stacking of individual closing tokens the
1653 following controls can be used:
1655 -scp or --stack-closing-paren
1656 -schb or --stack-closing-hash-brace
1657 -scsb or --stack-closing-square-bracket
1659 The flag B<-sct> is a synonym for B<-scp -schb -scsb>.
1661 =item B<-dnl>, B<--delete-old-newlines>
1663 By default, perltidy first deletes all old line break locations, and then it
1664 looks for good break points to match the desired line length. Use B<-ndnl>
1665 or B<--nodelete-old-newlines> to force perltidy to retain all old line break
1668 =item B<-anl>, B<--add-newlines>
1670 By default, perltidy will add line breaks when necessary to create
1671 continuations of long lines and to improve the script appearance. Use
1672 B<-nanl> or B<--noadd-newlines> to prevent any new line breaks.
1674 This flag does not prevent perltidy from eliminating existing line
1675 breaks; see B<--freeze-newlines> to completely prevent changes to line
1678 =item Controlling whether perltidy breaks before or after operators
1680 Four command line parameters provide some control over whether
1681 a line break should be before or after specific token types.
1682 Two parameters give detailed control:
1684 B<-wba=s> or B<--want-break-after=s>, and
1686 B<-wbb=s> or B<--want-break-before=s>.
1688 These parameters are each followed by a quoted string, B<s>, containing
1689 a list of token types (separated only by spaces). No more than one of each
1690 of these parameters should be specified, because repeating a
1691 command-line parameter always overwrites the previous one before
1692 perltidy ever sees it.
1694 By default, perltidy breaks B<after> these token types:
1695 % + - * / x != == >= <= =~ !~ < > | &
1696 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
1698 And perltidy breaks B<before> these token types by default:
1701 To illustrate, to cause a break after a concatenation operator, C<'.'>,
1702 rather than before it, the command line would be
1706 As another example, the following command would cause a break before
1707 math operators C<'+'>, C<'-'>, C<'/'>, and C<'*'>:
1711 These commands should work well for most of the token types that perltidy uses
1712 (use B<--dump-token-types> for a list). Also try the B<-D> flag on a short
1713 snippet of code and look at the .DEBUG file to see the tokenization. However,
1714 for a few token types there may be conflicts with hardwired logic which cause
1715 unexpected results. One example is curly braces, which should be controlled
1716 with the parameter B<bl> provided for that purpose.
1718 B<WARNING> Be sure to put these tokens in quotes to avoid having them
1719 misinterpreted by your command shell.
1721 Two additional parameters are available which, though they provide no further
1722 capability, can simplify input are:
1724 B<-baao> or B<--break-after-all-operators>,
1726 B<-bbao> or B<--break-before-all-operators>.
1728 The -baao sets the default to be to break after all of the following operators:
1730 % + - * / x != == >= <= =~ !~ < > | &
1731 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
1733 and the B<-bbao> flag sets the default to break before all of these operators.
1734 These can be used to define an initial break preference which can be fine-tuned
1735 with the B<-wba> and B<-wbb> flags. For example, to break before all operators
1736 except an B<=> one could use --bbao -wba='=' rather than listing every
1737 single perl operator except B<=> on a -wbb flag.
1741 =head2 Controlling List Formatting
1743 Perltidy attempts to place comma-separated arrays of values in tables
1744 which look good. Its default algorithms usually work well, and they
1745 have been improving with each release, but several parameters are
1746 available to control list formatting.
1750 =item B<-boc>, B<--break-at-old-comma-breakpoints>
1752 This flag tells perltidy to try to break at all old commas. This is not
1753 the default. Normally, perltidy makes a best guess at list formatting,
1754 and seldom uses old comma breakpoints. Usually this works well,
1763 The default formatting will flatten this down to one line:
1765 # perltidy (default)
1766 my @list = ( 1, 1, 1, 1, 2, 1, 1, 3, 3, 1, 1, 4, 6, 4, 1, );
1768 which hides the structure. Using B<-boc>, plus additional flags
1769 to retain the original style, yields
1771 # perltidy -boc -lp -pt=2 -vt=1 -vtc=1
1778 A disadvantage of this flag is that all tables in the file
1779 must already be nicely formatted. For another possibility see
1780 the -fs flag in L<Skipping Selected Sections of Code>.
1782 =item B<-mft=n>, B<--maximum-fields-per-table=n>
1784 If the computed number of fields for any table exceeds B<n>, then it
1785 will be reduced to B<n>. The default value for B<n> is a large number,
1786 40. While this value should probably be left unchanged as a general
1787 rule, it might be used on a small section of code to force a list to
1788 have a particular number of fields per line, and then either the B<-boc>
1789 flag could be used to retain this formatting, or a single comment could
1790 be introduced somewhere to freeze the formatting in future applications
1803 =item B<-cab=n>, B<--comma-arrow-breakpoints=n>
1805 A comma which follows a comma arrow, '=>', requires special
1806 consideration. In a long list, it is common to break at all such
1807 commas. This parameter can be used to control how perltidy breaks at
1808 these commas. (However, it will have no effect if old comma breaks are
1809 being forced because B<-boc> is used). The possible values of B<n> are:
1811 n=0 break at all commas after =>
1812 n=1 stable: break at all commas after => unless this would break
1813 an existing one-line container (default)
1814 n=2 break at all commas after =>, but try to form the maximum
1815 maximum one-line container lengths
1816 n=3 do not treat commas after => specially at all
1818 For example, given the following single line, perltidy by default will
1819 not add any line breaks because it would break the existing one-line
1822 bless { B => $B, Root => $Root } => $package;
1824 Using B<-cab=0> will force a break after each comma-arrow item:
1832 If perltidy is subsequently run with this container broken, then by
1833 default it will break after each '=>' because the container is now
1834 broken. To reform a one-line container, the parameter B<-cab=2> would
1837 The flag B<-cab=3> can be used to prevent these commas from being
1838 treated specially. In this case, an item such as "01" => 31 is
1839 treated as a single item in a table. The number of fields in this table
1840 will be determined by the same rules that are used for any other table.
1845 "01" => 31, "02" => 29, "03" => 31, "04" => 30,
1846 "05" => 31, "06" => 30, "07" => 31, "08" => 31,
1847 "09" => 30, "10" => 31, "11" => 30, "12" => 31
1852 =head2 Retaining or Ignoring Existing Line Breaks
1854 Several additional parameters are available for controlling the extent
1855 to which line breaks in the input script influence the output script.
1856 In most cases, the default parameter values are set so that, if a choice
1857 is possible, the output style follows the input style. For example, if
1858 a short logical container is broken in the input script, then the
1859 default behavior is for it to remain broken in the output script.
1861 Most of the parameters in this section would only be required for a
1862 one-time conversion of a script from short container lengths to longer
1863 container lengths. The opposite effect, of converting long container
1864 lengths to shorter lengths, can be obtained by temporarily using a short
1865 maximum line length.
1869 =item B<-bol>, B<--break-at-old-logical-breakpoints>
1871 By default, if a logical expression is broken at a C<&&>, C<||>, C<and>,
1872 or C<or>, then the container will remain broken. Also, breaks
1873 at internal keywords C<if> and C<unless> will normally be retained.
1874 To prevent this, and thus form longer lines, use B<-nbol>.
1876 =item B<-bok>, B<--break-at-old-keyword-breakpoints>
1878 By default, perltidy will retain a breakpoint before keywords which may
1879 return lists, such as C<sort> and <map>. This allows chains of these
1880 operators to be displayed one per line. Use B<-nbok> to prevent
1881 retaining these breakpoints.
1883 =item B<-bot>, B<--break-at-old-ternary-breakpoints>
1885 By default, if a conditional (ternary) operator is broken at a C<:>,
1886 then it will remain broken. To prevent this, and thereby
1887 form longer lines, use B<-nbot>.
1889 =item B<-iob>, B<--ignore-old-breakpoints>
1891 Use this flag to tell perltidy to ignore existing line breaks to the
1892 maximum extent possible. This will tend to produce the longest possible
1893 containers, regardless of type, which do not exceed the line length
1896 =item B<-kis>, B<--keep-interior-semicolons>
1898 Use the B<-kis> flag to prevent breaking at a semicolon if
1899 there was no break there in the input flag. Normally
1900 perltidy places a newline after each semicolon which
1901 terminates a statement unless several statements are
1902 contained within a one-line brace block. To illustrate,
1903 consider the following input lines:
1905 dbmclose(%verb_delim); undef %verb_delim;
1906 dbmclose(%expanded); undef %expanded;
1908 The default is to break after each statement, giving
1910 dbmclose(%verb_delim);
1912 dbmclose(%expanded);
1915 With B<perltidy -kis> the multiple statements are retained:
1917 dbmclose(%verb_delim); undef %verb_delim;
1918 dbmclose(%expanded); undef %expanded;
1920 The statements are still subject to the specified value
1921 of B<maximum-line-length> and will be broken if this
1926 =head2 Blank Line Control
1928 Blank lines can improve the readability of a script if they are carefully
1929 placed. Perltidy has several commands for controlling the insertion,
1930 retention, and removal of blank lines.
1934 =item B<-bbc>, B<--blanks-before-comments>
1936 A blank line will be introduced before a full-line comment. This is the
1937 default. Use B<-nbbc> or B<--noblanks-before-comments> to prevent
1938 such blank lines from being introduced.
1940 =item B<-bbs>, B<--blanks-before-subs>
1942 A blank line will be introduced before a B<sub> definition, unless it is a
1943 one-liner or preceded by a comment. A blank line will also be introduced
1944 before a B<package> statement and a B<BEGIN> and B<END> block. This is the
1945 default. The intention is to help display the structure of a program by
1946 setting off certain key sections of code. This is negated with B<-nbbs> or
1947 B<--noblanks-before-subs>.
1949 =item B<-bbb>, B<--blanks-before-blocks>
1951 A blank line will be introduced before blocks of coding delimited by
1952 B<for>, B<foreach>, B<while>, B<until>, and B<if>, B<unless>, in the following
1959 The block is not preceded by a comment.
1963 The block is not a one-line block.
1967 The number of consecutive non-blank lines at the current indentation depth is at least B<-lbl>
1972 This is the default. The intention of this option is to introduce
1973 some space within dense coding.
1974 This is negated with B<-nbbb> or B<--noblanks-before-blocks>.
1976 =item B<-lbl=n> B<--long-block-line-count=n>
1978 This controls how often perltidy is allowed to add blank lines before
1979 certain block types (see previous section). The default is 8. Entering
1980 a value of B<0> is equivalent to entering a very large number.
1982 =item B<-mbl=n> B<--maximum-consecutive-blank-lines=n>
1984 This parameter specifies the maximum number of consecutive blank lines
1985 in the output script. The default is n=1. If the input file has more
1986 than n consecutive blank lines, the number will be reduced to n.
1987 (This obviously does not apply to pod sections, here-documents, and quotes).
1989 =item B<-sob>, B<--swallow-optional-blank-lines>
1991 All blank lines not required by the above flags, B<-bbb>, B<-bbs>, and B<-bbc>,
1992 will be deleted. (But essential blank lines above pod documents will be
1993 retained). This is NOT the default.
1995 =item B<-nsob>, B<--noswallow-optional-blank-lines>
1997 Retain blank lines, including those which do not corresponding to flags
1998 B<-bbb>, B<-bbs>, and B<-bbc>. This is the default. The number of
1999 blanks retained is subject to the limit imposed by
2000 B<--maximum-consecutive-blank-lines>, however.
2006 A style refers to a convenient collection of existing parameters.
2010 =item B<-gnu>, B<--gnu-style>
2012 B<-gnu> gives an approximation to the GNU Coding Standards (which do
2013 not apply to perl) as they are sometimes implemented. At present, this
2014 style overrides the default style with the following parameters:
2016 -lp -bl -noll -pt=2 -bt=2 -sbt=2 -icp
2018 =item B<-pbp>, B<--perl-best-practices>
2020 B<-pbp> is an abbreviation for the parameters in the book B<Perl Best Practices>
2023 -l=78 -i=4 -ci=4 -st -se -vt=2 -cti=0 -pt=1 -bt=1 -sbt=1 -bbt=1 -nsfs -nolq
2024 -wbb="% + - * / x != == >= <= =~ !~ < > | & =
2025 **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x="
2027 Note that the -st and -se flags make perltidy act as a filter on one file only.
2028 These can be overridden with -nst and -nse if necessary.
2032 =head2 Other Controls
2036 =item Deleting selected text
2038 Perltidy can selectively delete comments and/or pod documentation. The
2039 command B<-dac> or B<--delete-all-comments> will delete all comments
2040 B<and> all pod documentation, leaving just code and any leading system
2043 The command B<-dp> or B<--delete-pod> will remove all pod documentation
2046 Two commands which remove comments (but not pod) are: B<-dbc> or
2047 B<--delete-block-comments> and B<-dsc> or B<--delete-side-comments>.
2048 (Hanging side comments will be deleted with block comments here.)
2050 The negatives of these commands also work, and are the defaults. When
2051 block comments are deleted, any leading 'hash-bang' will be retained.
2052 Also, if the B<-x> flag is used, any system commands before a leading
2053 hash-bang will be retained (even if they are in the form of comments).
2055 =item Writing selected text to a file
2057 When perltidy writes a formatted text file, it has the ability to also
2058 send selected text to a file with a F<.TEE> extension. This text can
2059 include comments and pod documentation.
2061 The command B<-tac> or B<--tee-all-comments> will write all comments
2062 B<and> all pod documentation.
2064 The command B<-tp> or B<--tee-pod> will write all pod documentation (but
2067 The commands which write comments (but not pod) are: B<-tbc> or
2068 B<--tee-block-comments> and B<-tsc> or B<--tee-side-comments>.
2069 (Hanging side comments will be written with block comments here.)
2071 The negatives of these commands also work, and are the defaults.
2073 =item Using a F<.perltidyrc> command file
2075 If you use perltidy frequently, you probably won't be happy until you
2076 create a F<.perltidyrc> file to avoid typing commonly-used parameters.
2077 Perltidy will first look in your current directory for a command file
2078 named F<.perltidyrc>. If it does not find one, it will continue looking
2079 for one in other standard locations.
2081 These other locations are system-dependent, and may be displayed with
2082 the command C<perltidy -dpro>. Under Unix systems, it will look for a
2083 F<.perltidyrc> file in the home directory, and then for a system-wide
2084 file F</usr/local/etc/perltidyrc>, and then it will look for
2085 F</etc/perltidyrc>. Note that these last two system-wide files do not
2086 have a leading dot. Further system-dependent information will be found
2087 in the INSTALL file distributed with perltidy.
2089 This file is free format, and simply a list of parameters, just as they
2090 would be entered on a command line. Any number of lines may be used,
2091 with any number of parameters per line, although it may be easiest to
2092 read with one parameter per line. Blank lines are ignored, and text
2093 after a '#' is ignored to the end of a line.
2095 Here is an example of a F<.perltidyrc> file:
2097 # This is a simple of a .perltidyrc configuration file
2098 # This implements a highly spaced style
2099 -se # errors to standard error output
2100 -w # show all warnings
2101 -bl # braces on new lines
2102 -pt=0 # parens not tight at all
2103 -bt=0 # braces not tight
2104 -sbt=0 # square brackets not tight
2106 The parameters in the F<.perltidyrc> file are installed first, so any
2107 parameters given on the command line will have priority over them.
2109 To avoid confusion, perltidy ignores any command in the .perltidyrc
2110 file which would cause some kind of dump and an exit. These are:
2112 -h -v -ddf -dln -dop -dsn -dtt -dwls -dwrs -ss
2114 There are several options may be helpful in debugging a F<.perltidyrc>
2121 A very helpful command is B<--dump-profile> or B<-dpro>. It writes a
2122 list of all configuration filenames tested to standard output, and
2123 if a file is found, it dumps the content to standard output before
2124 exiting. So, to find out where perltidy looks for its configuration
2125 files, and which one if any it selects, just enter
2131 It may be simplest to develop and test configuration files with
2132 alternative names, and invoke them with B<-pro=filename> on the command
2133 line. Then rename the desired file to F<.perltidyrc> when finished.
2137 The parameters in the F<.perltidyrc> file can be switched off with
2138 the B<-npro> option.
2142 The commands B<--dump-options>, B<--dump-defaults>, B<--dump-long-names>,
2143 and B<--dump-short-names>, all described below, may all be helpful.
2147 =item Creating a new abbreviation
2149 A special notation is available for use in a F<.perltidyrc> file
2150 for creating an abbreviation for a group
2151 of options. This can be used to create a
2152 shorthand for one or more styles which are frequently, but not always,
2153 used. The notation is to group the options within curly braces which
2154 are preceded by the name of the alias (without leading dashes), like this:
2161 where B<newword> is the abbreviation, and B<opt1>, etc, are existing parameters
2162 I<or other abbreviations>. The main syntax requirement is that
2163 the new abbreviation must begin on a new line.
2164 Space before and after the curly braces is optional.
2166 specific example, the following line
2168 airy {-bl -pt=0 -bt=0 -sbt=0}
2170 could be placed in a F<.perltidyrc> file, and then invoked at will with
2172 perltidy -airy somefile.pl
2174 (Either C<-airy> or C<--airy> may be used).
2176 =item Skipping leading non-perl commands with B<-x> or B<--look-for-hash-bang>
2178 If your script has leading lines of system commands or other text which
2179 are not valid perl code, and which are separated from the start of the
2180 perl code by a "hash-bang" line, ( a line of the form C<#!...perl> ),
2181 you must use the B<-x> flag to tell perltidy not to parse and format any
2182 lines before the "hash-bang" line. This option also invokes perl with a
2183 -x flag when checking the syntax. This option was originally added to
2184 allow perltidy to parse interactive VMS scripts, but it should be used
2185 for any script which is normally invoked with C<perl -x>.
2187 =item Making a file unreadable
2189 The goal of perltidy is to improve the readability of files, but there
2190 are two commands which have the opposite effect, B<--mangle> and
2191 B<--extrude>. They are actually
2192 merely aliases for combinations of other parameters. Both of these
2193 strip all possible whitespace, but leave comments and pod documents,
2194 so that they are essentially reversible. The
2195 difference between these is that B<--mangle> puts the fewest possible
2196 line breaks in a script while B<--extrude> puts the maximum possible.
2197 Note that these options do not provided any meaningful obfuscation, because
2198 perltidy can be used to reformat the files. They were originally
2199 developed to help test the tokenization logic of perltidy, but they
2201 One use for B<--mangle> is the following:
2203 perltidy --mangle myfile.pl -st | perltidy -o myfile.pl.new
2205 This will form the maximum possible number of one-line blocks (see next
2206 section), and can sometimes help clean up a badly formatted script.
2208 A similar technique can be used with B<--extrude> instead of B<--mangle>
2209 to make the minimum number of one-line blocks.
2211 Another use for B<--mangle> is to combine it with B<-dac> to reduce
2212 the file size of a perl script.
2214 =item One-line blocks
2216 There are a few points to note regarding one-line blocks. A one-line
2217 block is something like this,
2219 if ($x > 0) { $y = 1 / $x }
2221 where the contents within the curly braces is short enough to fit
2224 With few exceptions, perltidy retains existing one-line blocks, if it
2225 is possible within the line-length constraint, but it does not attempt
2226 to form new ones. In other words, perltidy will try to follow the
2227 one-line block style of the input file.
2229 If an existing one-line block is longer than the maximum line length,
2230 however, it will be broken into multiple lines. When this happens, perltidy
2231 checks for and adds any optional terminating semicolon (unless the B<-nasc>
2232 option is used) if the block is a code block.
2234 The main exception is that perltidy will attempt to form new one-line
2235 blocks following the keywords C<map>, C<eval>, and C<sort>, because
2236 these code blocks are often small and most clearly displayed in a single
2239 One-line block rules can conflict with the cuddled-else option. When
2240 the cuddled-else option is used, perltidy retains existing one-line
2241 blocks, even if they do not obey cuddled-else formatting.
2243 Occasionally, when one-line blocks get broken because they exceed the
2244 available line length, the formatting will violate the requested brace style.
2245 If this happens, reformatting the script a second time should correct
2250 The following flags are available for debugging:
2252 B<--dump-defaults> or B<-ddf> will write the default option set to standard output and quit
2254 B<--dump-profile> or B<-dpro> will write the name of the current
2255 configuration file and its contents to standard output and quit.
2257 B<--dump-options> or B<-dop> will write current option set to standard
2260 B<--dump-long-names> or B<-dln> will write all command line long names (passed
2261 to Get_options) to standard output and quit.
2263 B<--dump-short-names> or B<-dsn> will write all command line short names
2264 to standard output and quit.
2266 B<--dump-token-types> or B<-dtt> will write a list of all token types
2267 to standard output and quit.
2269 B<--dump-want-left-space> or B<-dwls> will write the hash %want_left_space
2270 to standard output and quit. See the section on controlling whitespace
2273 B<--dump-want-right-space> or B<-dwrs> will write the hash %want_right_space
2274 to standard output and quit. See the section on controlling whitespace
2277 B<-DEBUG> will write a file with extension F<.DEBUG> for each input file
2278 showing the tokenization of all lines of code.
2280 =item Working with MakeMaker, AutoLoader and SelfLoader
2282 The first $VERSION line of a file which might be eval'd by MakeMaker
2283 is passed through unchanged except for indentation.
2284 Use B<--nopass-version-line>, or B<-npvl>, to deactivate this feature.
2286 If the AutoLoader module is used, perltidy will continue formatting
2287 code after seeing an __END__ line.
2288 Use B<--nolook-for-autoloader>, or B<-nlal>, to deactivate this feature.
2290 Likewise, if the SelfLoader module is used, perltidy will continue formatting
2291 code after seeing a __DATA__ line.
2292 Use B<--nolook-for-selfloader>, or B<-nlsl>, to deactivate this feature.
2294 =item Working around problems with older version of Perl
2296 Perltidy contains a number of rules which help avoid known subtleties
2297 and problems with older versions of perl, and these rules always
2298 take priority over whatever formatting flags have been set. For example,
2299 perltidy will usually avoid starting a new line with a bareword, because
2300 this might cause problems if C<use strict> is active.
2302 There is no way to override these rules.
2310 =item The B<-html> master switch
2312 The flag B<-html> causes perltidy to write an html file with extension
2313 F<.html>. So, for example, the following command
2315 perltidy -html somefile.pl
2317 will produce a syntax-colored html file named F<somefile.pl.html>
2318 which may be viewed with a browser.
2320 B<Please Note>: In this case, perltidy does not do any formatting to the
2321 input file, and it does not write a formatted file with extension
2322 F<.tdy>. This means that two perltidy runs are required to create a
2323 fully reformatted, html copy of a script.
2325 =item The B<-pre> flag for code snippets
2327 When the B<-pre> flag is given, only the pre-formatted section, within
2328 the <PRE> and </PRE> tags, will be output. This simplifies inclusion
2329 of the output in other files. The default is to output a complete
2332 =item The B<-nnn> flag for line numbering
2334 When the B<-nnn> flag is given, the output lines will be numbered.
2336 =item The B<-toc>, or B<--html-table-of-contents> flag
2338 By default, a table of contents to packages and subroutines will be
2339 written at the start of html output. Use B<-ntoc> to prevent this.
2340 This might be useful, for example, for a pod document which contains a
2341 number of unrelated code snippets. This flag only influences the code
2342 table of contents; it has no effect on any table of contents produced by
2343 pod2html (see next item).
2345 =item The B<-pod>, or B<--pod2html> flag
2347 There are two options for formatting pod documentation. The default is
2348 to pass the pod through the Pod::Html module (which forms the basis of
2349 the pod2html utility). Any code sections are formatted by perltidy, and
2350 the results then merged. Note: perltidy creates a temporary file when
2351 Pod::Html is used; see L<"FILES">. Also, Pod::Html creates temporary
2352 files for its cache.
2354 NOTE: Perltidy counts the number of C<=cut> lines, and either moves the
2355 pod text to the top of the html file if there is one C<=cut>, or leaves
2356 the pod text in its original order (interleaved with code) otherwise.
2358 Most of the flags accepted by pod2html may be included in the perltidy
2359 command line, and they will be passed to pod2html. In some cases,
2360 the flags have a prefix C<pod> to emphasize that they are for the
2361 pod2html, and this prefix will be removed before they are passed to
2362 pod2html. The flags which have the additional C<pod> prefix are:
2364 --[no]podheader --[no]podindex --[no]podrecurse --[no]podquiet
2365 --[no]podverbose --podflush
2367 The flags which are unchanged from their use in pod2html are:
2369 --backlink=s --cachedir=s --htmlroot=s --libpods=s --title=s
2370 --podpath=s --podroot=s
2372 where 's' is an appropriate character string. Not all of these flags are
2373 available in older versions of Pod::Html. See your Pod::Html documentation for
2376 The alternative, indicated with B<-npod>, is not to use Pod::Html, but
2377 rather to format pod text in italics (or whatever the stylesheet
2378 indicates), without special html markup. This is useful, for example,
2379 if pod is being used as an alternative way to write comments.
2381 =item The B<-frm>, or B<--frames> flag
2383 By default, a single html output file is produced. This can be changed
2384 with the B<-frm> option, which creates a frame holding a table of
2385 contents in the left panel and the source code in the right side. This
2386 simplifies code browsing. Assume, for example, that the input file is
2387 F<MyModule.pm>. Then, for default file extension choices, these three
2388 files will be created:
2390 MyModule.pm.html - the frame
2391 MyModule.pm.toc.html - the table of contents
2392 MyModule.pm.src.html - the formatted source code
2394 Obviously this file naming scheme requires that output be directed to a real
2395 file (as opposed to, say, standard output). If this is not the
2396 case, or if the file extension is unknown, the B<-frm> option will be
2399 =item The B<-text=s>, or B<--html-toc-extension> flag
2401 Use this flag to specify the extra file extension of the table of contents file
2402 when html frames are used. The default is "toc".
2403 See L<Specifying File Extensions>.
2405 =item The B<-sext=s>, or B<--html-src-extension> flag
2407 Use this flag to specify the extra file extension of the content file when html
2408 frames are used. The default is "src".
2409 See L<Specifying File Extensions>.
2411 =item The B<-hent>, or B<--html-entities> flag
2413 This flag controls the use of Html::Entities for html formatting. By
2414 default, the module Html::Entities is used to encode special symbols.
2415 This may not be the right thing for some browser/language
2416 combinations. Use --nohtml-entities or -nhent to prevent this.
2420 Style sheets make it very convenient to control and adjust the
2421 appearance of html pages. The default behavior is to write a page of
2422 html with an embedded style sheet.
2424 An alternative to an embedded style sheet is to create a page with a
2425 link to an external style sheet. This is indicated with the
2426 B<-css=filename>, where the external style sheet is F<filename>. The
2427 external style sheet F<filename> will be created if and only if it does
2428 not exist. This option is useful for controlling multiple pages from a
2431 To cause perltidy to write a style sheet to standard output and exit,
2432 use the B<-ss>, or B<--stylesheet>, flag. This is useful if the style
2433 sheet could not be written for some reason, such as if the B<-pre> flag
2434 was used. Thus, for example,
2436 perltidy -html -ss >mystyle.css
2438 will write a style sheet with the default properties to file
2441 The use of style sheets is encouraged, but a web page without a style
2442 sheets can be created with the flag B<-nss>. Use this option if you
2443 must to be sure that older browsers (roughly speaking, versions prior to
2444 4.0 of Netscape Navigator and Internet Explorer) can display the
2445 syntax-coloring of the html files.
2447 =item Controlling HTML properties
2449 Note: It is usually more convenient to accept the default properties
2450 and then edit the stylesheet which is produced. However, this section
2451 shows how to control the properties with flags to perltidy.
2453 Syntax colors may be changed from their default values by flags of the either
2454 the long form, B<-html-color-xxxxxx=n>, or more conveniently the short form,
2455 B<-hcx=n>, where B<xxxxxx> is one of the following words, and B<x> is the
2456 corresponding abbreviation:
2459 ---------- -------- --
2462 identifier identifier i
2463 bareword, function bareword w
2465 quite, pattern quote q
2466 here doc text here-doc-text h
2467 here doc target here-doc-target hh
2468 punctuation punctuation pu
2470 structural braces structure s
2471 semicolon semicolon sc
2475 sub definition name subroutine m
2476 pod text pod-text pd
2478 A default set of colors has been defined, but they may be changed by providing
2479 values to any of the following parameters, where B<n> is either a 6 digit
2480 hex RGB color value or an ascii name for a color, such as 'red'.
2482 To illustrate, the following command will produce an html
2483 file F<somefile.pl.html> with "aqua" keywords:
2485 perltidy -html -hck=00ffff somefile.pl
2487 and this should be equivalent for most browsers:
2489 perltidy -html -hck=aqua somefile.pl
2491 Perltidy merely writes any non-hex names that it sees in the html file.
2492 The following 16 color names are defined in the HTML 3.2 standard:
2511 Many more names are supported in specific browsers, but it is safest
2512 to use the hex codes for other colors. Helpful color tables can be
2513 located with an internet search for "HTML color tables".
2515 Besides color, two other character attributes may be set: bold, and italics.
2516 To set a token type to use bold, use the flag
2517 B<--html-bold-xxxxxx> or B<-hbx>, where B<xxxxxx> or B<x> are the long
2518 or short names from the above table. Conversely, to set a token type to
2519 NOT use bold, use B<--nohtml-bold-xxxxxx> or B<-nhbx>.
2521 Likewise, to set a token type to use an italic font, use the flag
2522 B<--html-italic-xxxxxx> or B<-hix>, where again B<xxxxxx> or B<x> are the
2523 long or short names from the above table. And to set a token type to
2524 NOT use italics, use B<--nohtml-italic-xxxxxx> or B<-nhix>.
2526 For example, to use bold braces and lime color, non-bold, italics keywords the
2527 following command would be used:
2529 perltidy -html -hbs -hck=00FF00 -nhbk -hik somefile.pl
2531 The background color can be specified with B<--html-color-background=n>,
2532 or B<-hcbg=n> for short, where n is a 6 character hex RGB value. The
2533 default color of text is the value given to B<punctuation>, which is
2536 Here are some notes and hints:
2538 1. If you find a preferred set of these parameters, you may want
2539 to create a F<.perltidyrc> file containing them. See the perltidy man
2540 page for an explanation.
2542 2. Rather than specifying values for these parameters, it is probably
2543 easier to accept the defaults and then edit a style sheet. The style
2544 sheet contains comments which should make this easy.
2546 3. The syntax-colored html files can be very large, so it may be best to
2547 split large files into smaller pieces to improve download times.
2551 =head1 SOME COMMON INPUT CONVENTIONS
2553 =head2 Specifying Block Types
2555 Several parameters which refer to code block types may be customized by also
2556 specifying an associated list of block types. The type of a block is the name
2557 of the keyword which introduces that block, such as B<if>, B<else>, or B<sub>.
2558 An exception is a labeled block, which has no keyword, and should be specified
2561 For example, the following parameter specifies C<sub>, labels, C<BEGIN>, and
2564 -cscl="sub : BEGIN END"
2566 (the meaning of the -cscl parameter is described above.) Note that
2567 quotes are required around the list of block types because of the
2570 =head2 Specifying File Extensions
2572 Several parameters allow default file extensions to be overridden. For
2573 example, a backup file extension may be specified with B<-bext=ext>,
2574 where B<ext> is some new extension. In order to provides the user some
2575 flexibility, the following convention is used in all cases to decide if
2576 a leading '.' should be used. If the extension C<ext> begins with
2577 C<A-Z>, C<a-z>, or C<0-9>, then it will be appended to the filename with
2578 an intermediate '.' (or perhaps an '_' on VMS systems). Otherwise, it
2579 will be appended directly.
2581 For example, suppose the file is F<somefile.pl>. For C<-bext=old>, a '.' is
2582 added to give F<somefile.pl.old>. For C<-bext=.old>, no additional '.' is
2583 added, so again the backup file is F<somefile.pl.old>. For C<-bext=~>, then no
2584 dot is added, and the backup file will be F<somefile.pl~> .
2586 =head1 SWITCHES WHICH MAY BE NEGATED
2588 The following list shows all short parameter names which allow a prefix
2589 'n' to produce the negated form:
2591 D anl asc aws b bbb bbc bbs bl bli boc bok bol bot ce
2592 csc dac dbc dcsc ddf dln dnl dop dp dpro dsc dsm dsn dtt dwls
2593 dwrs dws f fll frm fs hsc html ibc icb icp iob isbc lal log
2594 lp lsl ohbr okw ola oll opr opt osbr otr ple ple pod pvl q
2595 sbc sbl schb scp scsb sct se sfp sfs skp sob sohb sop sosb sot
2596 ssc st sts syn t tac tbc toc tp tqw tsc w x bar kis
2598 Equivalently, the prefix 'no' or 'no-' on the corresponding long names may be
2605 =item Parsing Limitations
2607 Perltidy should work properly on most perl scripts. It does a lot of
2608 self-checking, but still, it is possible that an error could be
2609 introduced and go undetected. Therefore, it is essential to make
2610 careful backups and to test reformatted scripts.
2612 The main current limitation is that perltidy does not scan modules
2613 included with 'use' statements. This makes it necessary to guess the
2614 context of any bare words introduced by such modules. Perltidy has good
2615 guessing algorithms, but they are not infallible. When it must guess,
2616 it leaves a message in the log file.
2618 If you encounter a bug, please report it.
2620 =item What perltidy does not parse and format
2622 Perltidy indents but does not reformat comments and C<qw> quotes.
2623 Perltidy does not in any way modify the contents of here documents or
2624 quoted text, even if they contain source code. (You could, however,
2625 reformat them separately). Perltidy does not format 'format' sections
2626 in any way. And, of course, it does not modify pod documents.
2634 =item Temporary files
2636 Under the -html option with the default --pod2html flag, a temporary file is
2637 required to pass text to Pod::Html. Unix systems will try to use the POSIX
2638 tmpnam() function. Otherwise the file F<perltidy.TMP> will be temporarily
2639 created in the current working directory.
2641 =item Special files when standard input is used
2643 When standard input is used, the log file, if saved, is F<perltidy.LOG>,
2644 and any errors are written to F<perltidy.ERR> unless the B<-se> flag is
2645 set. These are saved in the current working directory.
2647 =item Files overwritten
2649 The following file extensions are used by perltidy, and files with these
2650 extensions may be overwritten or deleted: F<.ERR>, F<.LOG>, F<.TEE>,
2651 and/or F<.tdy>, F<.html>, and F<.bak>, depending on the run type and
2654 =item Files extensions limitations
2656 Perltidy does not operate on files for which the run could produce a file with
2657 a duplicated file extension. These extensions include F<.LOG>, F<.ERR>,
2658 F<.TEE>, and perhaps F<.tdy> and F<.bak>, depending on the run type. The
2659 purpose of this rule is to prevent generating confusing filenames such as
2660 F<somefile.tdy.tdy.tdy>.
2666 perlstyle(1), Perl::Tidy(3)
2670 This man page documents perltidy version 20070801.
2674 Michael Cartmell supplied code for adaptation to VMS and helped with
2677 Yves Orton supplied code for adaptation to the various versions
2680 Axel Rose supplied a patch for MacPerl.
2682 Hugh S. Myers designed and implemented the initial Perl::Tidy module interface.
2684 Many others have supplied key ideas, suggestions, and bug reports;
2685 see the CHANGES file.
2690 email: perltidy at users.sourceforge.net
2691 http://perltidy.sourceforge.net
2695 Copyright (c) 2000-2006 by Steve Hancock
2699 This package is free software; you can redistribute it and/or modify it
2700 under the terms of the "GNU General Public License".
2702 Please refer to the file "COPYING" for details.
2706 This package is distributed in the hope that it will be useful,
2707 but WITHOUT ANY WARRANTY; without even the implied warranty of
2708 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
2710 See the "GNU General Public License" for more details.