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.
422 B<NOTE>: This only works under unix-like systems and is ignored under
425 =item B<-ple>, B<--preserve-line-endings>
427 This flag tells perltidy to write its output files with the same line
428 endings as the input file, if possible. It should work for
429 B<dos>, B<unix>, and B<mac> line endings. It will only work if perltidy
430 input comes from a filename (rather than stdin, for example). If
431 perltidy has trouble determining the input file line ending, it will
432 revert to the default behavior of using the line ending of the host system.
433 B<NOTE>: This only works under unix-like systems and is ignored under
438 =head2 Code Indentation Control
442 =item B<-ci=n>, B<--continuation-indentation=n>
444 Continuation indentation is extra indentation spaces applied when
445 a long line is broken. The default is n=2, illustrated here:
448 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
450 The same example, with n=0, is a little harder to read:
453 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
455 The value given to B<-ci> is also used by some commands when a small
456 space is required. Examples are commands for outdenting labels,
457 B<-ola>, and control keywords, B<-okw>.
459 When default values are not used, it is suggested that the value B<n>
460 given with B<-ci=n> be no more than about one-half of the number of
461 spaces assigned to a full indentation level on the B<-i=n> command.
463 =item B<-sil=n> B<--starting-indentation-level=n>
465 By default, perltidy examines the input file and tries to determine the
466 starting indentation level. While it is often zero, it may not be
467 zero for a code snippet being sent from an editing session. If the
468 default method does not work correctly, or you want to change the
469 starting level, use B<-sil=n>, to force the starting level to be n.
471 =item List indentation using B<-lp>, B<--line-up-parentheses>
473 By default, perltidy indents lists with 4 spaces, or whatever value
474 is specified with B<-i=n>. Here is a small list formatted in this way:
478 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
479 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
482 Use the B<-lp> flag to add extra indentation to cause the data to begin
483 past the opening parentheses of a sub call or list, or opening square
484 bracket of an anonymous array, or opening curly brace of an anonymous
485 hash. With this option, the above list would become:
489 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
490 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
493 If the available line length (see B<-l=n> ) does not permit this much
494 space, perltidy will use less. For alternate placement of the
495 closing paren, see the next section.
497 This option has no effect on code BLOCKS, such as if/then/else blocks,
498 which always use whatever is specified with B<-i=n>. Also, the
499 existence of line breaks and/or block comments between the opening and
500 closing parens may cause perltidy to temporarily revert to its default
503 Note: The B<-lp> option may not be used together with the B<-t> tabs option.
504 It may, however, be used with the B<-et=n> tab method.
506 In addition, any parameter which significantly restricts the ability of
507 perltidy to choose newlines will conflict with B<-lp> and will cause
508 B<-lp> to be deactivated. These include B<-io>, B<-fnl>, B<-nanl>, and
509 B<-ndnl>. The reason is that the B<-lp> indentation style can require
510 the careful coordination of an arbitrary number of break points in
511 hierarchical lists, and these flags may prevent that.
513 =item B<-cti=n>, B<--closing-token-indentation>
515 The B<-cti=n> flag controls the indentation of a line beginning with
516 a C<)>, C<]>, or a non-block C<}>. Such a line receives:
518 -cti = 0 no extra indentation (default)
519 -cti = 1 extra indentation such that the closing token
520 aligns with its opening token.
521 -cti = 2 one extra indentation level if the line looks like:
523 -cti = 3 one extra indentation level always
525 The flags B<-cti=1> and B<-cti=2> work well with the B<-lp> flag (previous
528 # perltidy -lp -cti=1
530 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
531 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
534 # perltidy -lp -cti=2
536 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
537 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
540 These flags are merely hints to the formatter and they may not always be
541 followed. In particular, if -lp is not being used, the indentation for
542 B<cti=1> is constrained to be no more than one indentation level.
544 If desired, this control can be applied independently to each of the
545 closing container token types. In fact, B<-cti=n> is merely an
546 abbreviation for B<-cpi=n -csbi=n -cbi=n>, where:
547 B<-cpi> or B<--closing-paren-indentation> controls B<)>'s,
548 B<-csbi> or B<--closing-square-bracket-indentation> controls B<]>'s,
549 B<-cbi> or B<--closing-brace-indentation> controls non-block B<}>'s.
551 =item B<-icp>, B<--indent-closing-paren>
553 The B<-icp> flag is equivalent to
554 B<-cti=2>, described in the previous section. The B<-nicp> flag is
555 equivalent B<-cti=0>. They are included for backwards compatability.
557 =item B<-icb>, B<--indent-closing-brace>
559 The B<-icb> option gives one extra level of indentation to a brace which
560 terminates a code block . For example,
569 The default is not to do this, indicated by B<-nicb>.
571 =item B<-olq>, B<--outdent-long-quotes>
573 When B<-olq> is set, lines which is a quoted string longer than the
574 value B<maximum-line-length> will have their indentation removed to make
575 them more readable. This is the default. To prevent such out-denting,
576 use B<-nolq> or B<--nooutdent-long-lines>.
578 =item B<-oll>, B<--outdent-long-lines>
580 This command is equivalent to B<--outdent-long-quotes> and
581 B<--outdent-long-comments>, and it is included for compatibility with previous
582 versions of perltidy. The negation of this also works, B<-noll> or
583 B<--nooutdent-long-lines>, and is equivalent to setting B<-nolq> and B<-nolc>.
585 =item Outdenting Labels: B<-ola>, B<--outdent-labels>
587 This command will cause labels to be outdented by 2 spaces (or whatever B<-ci>
588 has been set to), if possible. This is the default. For example:
591 LOOP: while ( $i = <FOTOS> ) {
597 Use B<-nola> to not outdent labels.
599 =item Outdenting Keywords
603 =item B<-okw>, B<--outdent-keywords>
605 The command B<-okw> will will cause certain leading control keywords to
606 be outdented by 2 spaces (or whatever B<-ci> has been set to), if
607 possible. By default, these keywords are C<redo>, C<next>, C<last>,
608 C<goto>, and C<return>. The intention is to make these control keywords
609 easier to see. To change this list of keywords being outdented, see
612 For example, using C<perltidy -okw> on the previous example gives:
615 LOOP: while ( $i = <FOTOS> ) {
621 The default is not to do this.
623 =item Specifying Outdented Keywords: B<-okwl=string>, B<--outdent-keyword-list=string>
625 This command can be used to change the keywords which are outdented with
626 the B<-okw> command. The parameter B<string> is a required list of perl
627 keywords, which should be placed in quotes if there are more than one.
628 By itself, it does not cause any outdenting to occur, so the B<-okw>
629 command is still required.
631 For example, the commands C<-okwl="next last redo goto" -okw> will cause
632 those four keywords to be outdented. It is probably simplest to place
633 any B<-okwl> command in a F<.perltidyrc> file.
639 =head2 Whitespace Control
641 Whitespace refers to the blank space between variables, operators,
642 and other code tokens.
646 =item B<-fws>, B<--freeze-whitespace>
648 This flag causes your original whitespace to remain unchanged, and
649 causes the rest of the whitespace commands in this section, the
650 Code Indentation section, and
651 the Comment Control section to be ignored.
653 =item Tightness of curly braces, parentheses, and square brackets.
655 Here the term "tightness" will mean the closeness with which
656 pairs of enclosing tokens, such as parentheses, contain the quantities
657 within. A numerical value of 0, 1, or 2 defines the tightness, with
658 0 being least tight and 2 being most tight. Spaces within containers
659 are always symmetric, so if there is a space after a C<(> then there
660 will be a space before the corresponding C<)>.
662 The B<-pt=n> or B<--paren-tightness=n> parameter controls the space within
663 parens. The example below shows the effect of the three possible
666 if ( ( my $len_tab = length( $tabstr ) ) > 0 ) { # -pt=0
667 if ( ( my $len_tab = length($tabstr) ) > 0 ) { # -pt=1 (default)
668 if ((my $len_tab = length($tabstr)) > 0) { # -pt=2
670 When n is 0, there is always a space to the right of a '(' and to the left
671 of a ')'. For n=2 there is never a space. For n=1, the default, there
672 is a space unless the quantity within the parens is a single token, such
673 as an identifier or quoted string.
675 Likewise, the parameter B<-sbt=n> or B<--square-bracket-tightness=n>
676 controls the space within square brackets, as illustrated below.
678 $width = $col[ $j + $k ] - $col[ $j ]; # -sbt=0
679 $width = $col[ $j + $k ] - $col[$j]; # -sbt=1 (default)
680 $width = $col[$j + $k] - $col[$j]; # -sbt=2
682 Curly braces which do not contain code blocks are controlled by
683 the parameter B<-bt=n> or B<--brace-tightness=n>.
685 $obj->{ $parsed_sql->{ 'table' }[0] }; # -bt=0
686 $obj->{ $parsed_sql->{'table'}[0] }; # -bt=1 (default)
687 $obj->{$parsed_sql->{'table'}[0]}; # -bt=2
689 And finally, curly braces which contain blocks of code are controlled by the
690 parameter B<-bbt=n> or B<--block-brace-tightness=n> as illustrated in the
693 %bf = map { $_ => -M $_ } grep { /\.deb$/ } dirents '.'; # -bbt=0 (default)
694 %bf = map { $_ => -M $_ } grep {/\.deb$/} dirents '.'; # -bbt=1
695 %bf = map {$_ => -M $_} grep {/\.deb$/} dirents '.'; # -bbt=2
697 =item B<-sts>, B<--space-terminal-semicolon>
699 Some programmers prefer a space before all terminal semicolons. The
700 default is for no such space, and is indicated with B<-nsts> or
701 B<--nospace-terminal-semicolon>.
704 $i = 1; # -nsts (default)
706 =item B<-sfs>, B<--space-for-semicolon>
708 Semicolons within B<for> loops may sometimes be hard to see,
709 particularly when commas are also present. This option places spaces on
710 both sides of these special semicolons, and is the default. Use
711 B<-nsfs> or B<--nospace-for-semicolon> to deactivate it.
713 for ( @a = @$ap, $u = shift @a ; @a ; $u = $v ) { # -sfs (default)
714 for ( @a = @$ap, $u = shift @a; @a; $u = $v ) { # -nsfs
716 =item B<-asc>, B<--add-semicolons>
718 Setting B<-asc> allows perltidy to add any missing optional semicolon at the end
719 of a line which is followed by a closing curly brace on the next line. This
720 is the default, and may be deactivated with B<-nasc> or B<--noadd-semicolons>.
722 =item B<-dsm>, B<--delete-semicolons>
724 Setting B<-dsm> allows perltidy to delete extra semicolons which are
725 simply empty statements. This is the default, and may be deactivated
726 with B<-ndsm> or B<--nodelete-semicolons>. (Such semicolons are not
727 deleted, however, if they would promote a side comment to a block
730 =item B<-aws>, B<--add-whitespace>
732 Setting this option allows perltidy to add certain whitespace improve
733 code readability. This is the default. If you do not want any
734 whitespace added, but are willing to have some whitespace deleted, use
735 B<-naws>. (Use B<-fws> to leave whitespace completely unchanged).
737 =item B<-dws>, B<--delete-old-whitespace>
739 Setting this option allows perltidy to remove some old whitespace
740 between characters, if necessary. This is the default. If you
741 do not want any old whitespace removed, use B<-ndws> or
742 B<--nodelete-old-whitespace>.
744 =item Detailed whitespace controls around tokens
746 For those who want more detailed control over the whitespace around
747 tokens, there are four parameters which can directly modify the default
748 whitespace rules built into perltidy for any token. They are:
750 B<-wls=s> or B<--want-left-space=s>,
752 B<-nwls=s> or B<--nowant-left-space=s>,
754 B<-wrs=s> or B<--want-right-space=s>,
756 B<-nwrs=s> or B<--nowant-right-space=s>.
758 These parameters are each followed by a quoted string, B<s>, containing a
759 list of token types. No more than one of each of these parameters
760 should be specified, because repeating a command-line parameter
761 always overwrites the previous one before perltidy ever sees it.
763 To illustrate how these are used, suppose it is desired that there be no
764 space on either side of the token types B<= + - / *>. The following two
765 parameters would specify this desire:
767 -nwls="= + - / *" -nwrs="= + - / *"
769 (Note that the token types are in quotes, and that they are separated by
770 spaces). With these modified whitespace rules, the following line of math:
772 $root = -$b + sqrt( $b * $b - 4. * $a * $c ) / ( 2. * $a );
776 $root=-$b+sqrt( $b*$b-4.*$a*$c )/( 2.*$a );
778 These parameters should be considered to be hints to perltidy rather
779 than fixed rules, because perltidy must try to resolve conflicts that
780 arise between them and all of the other rules that it uses. One
781 conflict that can arise is if, between two tokens, the left token wants
782 a space and the right one doesn't. In this case, the token not wanting
783 a space takes priority.
785 It is necessary to have a list of all token types in order to create
786 this type of input. Such a list can be obtained by the command
787 B<--dump-token-types>. Also try the -D flag on a short snippet of code
788 and look at the .DEBUG file to see the tokenization.
790 B<WARNING> Be sure to put these tokens in quotes to avoid having them
791 misinterpreted by your command shell.
793 =item Space between specific keywords and opening paren
795 When an opening paren follows a Perl keyword, no space is introduced after the
796 keyword, unless it is (by default) one of these:
798 my local our and or eq ne if else elsif until unless
799 while for foreach return switch case given when
801 These defaults can be modified with two commands:
803 B<-sak=s> or B<--space-after-keyword=s> adds keywords.
805 B<-nsak=s> or B<--nospace-after-keyword=s> removes keywords.
807 where B<s> is a list of keywords (in quotes if necessary). For example,
809 my ( $a, $b, $c ) = @_; # default
810 my( $a, $b, $c ) = @_; # -nsak="my local our"
812 To put a space after all keywords, see the next item.
814 =item Space between all keywords and opening parens
816 When an opening paren follows a function or keyword, no space is introduced
817 after the keyword except for the keywords noted in the previous item. To
818 always put a space between a function or keyword and its opening paren,
821 B<-skp> or B<--space-keyword-paren>
823 You will probably also want to use the flag B<-sfp> (next item) too.
825 =item Space between all function names and opening parens
827 When an opening paren follows a function the default is not to introduce
828 a space. To cause a space to be introduced use:
830 B<-sfp> or B<--space-function-paren>
832 myfunc( $a, $b, $c ); # default
833 myfunc ( $a, $b, $c ); # -sfp
835 You will probably also want to use the flag B<-skp> (previous item) too.
837 =item Trimming whitespace around C<qw> quotes
839 B<-tqw> or B<--trim-qw> provide the default behavior of trimming
840 spaces around multi-line C<qw> quotes and indenting them appropriately.
842 B<-ntqw> or B<--notrim-qw> cause leading and trailing whitespace around
843 multi-line C<qw> quotes to be left unchanged. This option will not
844 normally be necessary, but was added for testing purposes, because in
845 some versions of perl, trimming C<qw> quotes changes the syntax tree.
849 =head2 Comment Controls
851 Perltidy has a number of ways to control the appearance of both block comments
852 and side comments. The term B<block comment> here refers to a full-line
853 comment, whereas B<side comment> will refer to a comment which appears on a
854 line to the right of some code.
858 =item B<-ibc>, B<--indent-block-comments>
860 Block comments normally look best when they are indented to the same
861 level as the code which follows them. This is the default behavior, but
862 you may use B<-nibc> to keep block comments left-justified. Here is an
865 # this comment is indented (-ibc, default)
866 if ($task) { yyy(); }
868 The alternative is B<-nibc>:
870 # this comment is not indented (-nibc)
871 if ($task) { yyy(); }
873 See also the next item, B<-isbc>, as well as B<-sbc>, for other ways to
874 have some indented and some outdented block comments.
876 =item B<-isbc>, B<--indent-spaced-block-comments>
878 If there is no leading space on the line, then the comment will not be
879 indented, and otherwise it may be.
881 If both B<-ibc> and B<-isbc> are set, then B<-isbc> takes priority.
883 =item B<-olc>, B<--outdent-long-comments>
885 When B<-olc> is set, lines which are full-line (block) comments longer
886 than the value B<maximum-line-length> will have their indentation
887 removed. This is the default; use B<-nolc> to prevent outdenting.
889 =item B<-msc=n>, B<--minimum-space-to-comment=n>
891 Side comments look best when lined up several spaces to the right of
892 code. Perltidy will try to keep comments at least n spaces to the
893 right. The default is n=4 spaces.
895 =item B<-hsc>, B<--hanging-side-comments>
897 By default, perltidy tries to identify and align "hanging side
898 comments", which are something like this:
900 my $IGNORE = 0; # This is a side comment
901 # This is a hanging side comment
904 A comment is considered to be a hanging side comment if (1) it immediately
905 follows a line with a side comment, or another hanging side comment, and
906 (2) there is some leading whitespace on the line.
907 To deactivate this feature, use B<-nhsc> or B<--nohanging-side-comments>.
908 If block comments are preceded by a blank line, or have no leading
909 whitespace, they will not be mistaken as hanging side comments.
911 =item Closing Side Comments
913 A closing side comment is a special comment which perltidy can
914 automatically create and place after the closing brace of a code block.
915 They can be useful for code maintenance and debugging. The command
916 B<-csc> (or B<--closing-side-comments>) adds or updates closing side
917 comments. For example, here is a small code snippet
920 if ( !defined( $_[0] ) ) {
921 print("Hello, World\n");
924 print( $_[0], "\n" );
928 And here is the result of processing with C<perltidy -csc>:
931 if ( !defined( $_[0] ) ) {
932 print("Hello, World\n");
935 print( $_[0], "\n" );
939 A closing side comment was added for C<sub message> in this case, but not
940 for the C<if> and C<else> blocks, because they were below the 6 line
941 cutoff limit for adding closing side comments. This limit may be
942 changed with the B<-csci> command, described below.
944 The command B<-dcsc> (or B<--delete-closing-side-comments>) reverses this
945 process and removes these comments.
947 Several commands are available to modify the behavior of these two basic
948 commands, B<-csc> and B<-dcsc>:
952 =item B<-csci=n>, or B<--closing-side-comment-interval=n>
954 where C<n> is the minimum number of lines that a block must have in
955 order for a closing side comment to be added. The default value is
956 C<n=6>. To illustrate:
958 # perltidy -csci=2 -csc
960 if ( !defined( $_[0] ) ) {
961 print("Hello, World\n");
962 } ## end if ( !defined( $_[0] ))
964 print( $_[0], "\n" );
965 } ## end else [ if ( !defined( $_[0] ))
968 Now the C<if> and C<else> blocks are commented. However, now this has
969 become very cluttered.
971 =item B<-cscp=string>, or B<--closing-side-comment-prefix=string>
973 where string is the prefix used before the name of the block type. The
974 default prefix, shown above, is C<## end>. This string will be added to
975 closing side comments, and it will also be used to recognize them in
976 order to update, delete, and format them. Any comment identified as a
977 closing side comment will be placed just a single space to the right of
980 =item B<-cscl=string>, or B<--closing-side-comment-list-string>
982 where C<string> is a list of block types to be tagged with closing side
983 comments. By default, all code block types preceded by a keyword or
984 label (such as C<if>, C<sub>, and so on) will be tagged. The B<-cscl>
985 command changes the default list to be any selected block types; see
986 L<Specifying Block Types>.
987 For example, the following command
988 requests that only C<sub>'s, labels, C<BEGIN>, and C<END> blocks be
989 affected by any B<-csc> or B<-dcsc> operation:
991 -cscl="sub : BEGIN END"
993 =item B<-csct=n>, or B<--closing-side-comment-maximum-text=n>
995 The text appended to certain block types, such as an C<if> block, is
996 whatever lies between the keyword introducing the block, such as C<if>,
997 and the opening brace. Since this might be too much text for a side
998 comment, there needs to be a limit, and that is the purpose of this
999 parameter. The default value is C<n=20>, meaning that no additional
1000 tokens will be appended to this text after its length reaches 20
1001 characters. Omitted text is indicated with C<...>. (Tokens, including
1002 sub names, are never truncated, however, so actual lengths may exceed
1003 this). To illustrate, in the above example, the appended text of the
1004 first block is C< ( !defined( $_[0] )...>. The existing limit of
1005 C<n=20> caused this text to be truncated, as indicated by the C<...>.
1007 =item B<-csce=n>, or B<--closing-side-comment-else-flag=n>
1009 The default, B<n=0>, places the text of the opening C<if> statement after any
1012 If B<n=2> is used, then each C<elsif> is also given the text of the opening
1013 C<if> statement. Also, an C<else> will include the text of a preceding
1014 C<elsif> statement. Note that this may result some long closing
1017 If B<n=1> is used, the results will be the same as B<n=2> whenever the
1018 resulting line length is less than the maximum allowed.
1020 =item B<-cscw>, or B<--closing-side-comment-warnings>
1022 This parameter is intended to help make the initial transition to the use of
1023 closing side comments.
1025 things to happen if a closing side comment replaces an existing, different
1026 closing side comment: first, an error message will be issued, and second, the
1027 original side comment will be placed alone on a new specially marked comment
1028 line for later attention.
1030 The intent is to avoid clobbering existing hand-written side comments
1031 which happen to match the pattern of closing side comments. This flag
1032 should only be needed on the first run with B<-csc>.
1036 B<Important Notes on Closing Side Comments:>
1042 Closing side comments are only placed on lines terminated with a closing
1043 brace. Certain closing styles, such as the use of cuddled elses
1044 (B<-ce>), preclude the generation of some closing side comments.
1048 Please note that adding or deleting of closing side comments takes
1049 place only through the commands B<-csc> or B<-dcsc>. The other commands,
1050 if used, merely modify the behavior of these two commands.
1054 It is recommended that the B<-cscw> flag be used along with B<-csc> on
1055 the first use of perltidy on a given file. This will prevent loss of
1056 any existing side comment data which happens to have the csc prefix.
1060 Once you use B<-csc>, you should continue to use it so that any
1061 closing side comments remain correct as code changes. Otherwise, these
1062 comments will become incorrect as the code is updated.
1066 If you edit the closing side comments generated by perltidy, you must also
1067 change the prefix to be different from the closing side comment prefix.
1068 Otherwise, your edits will be lost when you rerun perltidy with B<-csc>. For
1069 example, you could simply change C<## end> to be C<## End>, since the test is
1070 case sensitive. You may also want to use the B<-ssc> flag to keep these
1071 modified closing side comments spaced the same as actual closing side comments.
1075 Temporarily generating closing side comments is a useful technique for
1076 exploring and/or debugging a perl script, especially one written by someone
1077 else. You can always remove them with B<-dcsc>.
1081 =item Static Block Comments
1083 Static block comments are block comments with a special leading pattern,
1084 C<##> by default, which will be treated slightly differently from other
1085 block comments. They effectively behave as if they had glue along their
1086 left and top edges, because they stick to the left edge and previous line
1087 when there is no blank spaces in those places. This option is
1088 particularly useful for controlling how commented code is displayed.
1092 =item B<-sbc>, B<--static-block-comments>
1094 When B<-sbc> is used, a block comment with a special leading pattern, C<##> by
1095 default, will be treated specially.
1097 Comments so identified are treated as follows:
1103 If there is no leading space on the line, then the comment will not
1104 be indented, and otherwise it may be,
1108 no new blank line will be
1109 inserted before such a comment, and
1113 such a comment will never become
1114 a hanging side comment.
1118 For example, assuming C<@month_of_year> is
1121 @month_of_year = ( # -sbc (default)
1122 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1126 Without this convention, the above code would become
1128 @month_of_year = ( # -nsbc
1129 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1135 which is not as clear.
1136 The default is to use B<-sbc>. This may be deactivated with B<-nsbc>.
1138 =item B<-sbcp=string>, B<--static-block-comment-prefix=string>
1140 This parameter defines the prefix used to identify static block comments
1141 when the B<-sbc> parameter is set. The default prefix is C<##>,
1142 corresponding to C<-sbcp=##>. The prefix is actually part of a perl
1143 pattern used to match lines and it must either begin with C<#> or C<^#>.
1144 In the first case a prefix ^\s* will be added to match any leading
1145 whitespace, while in the second case the pattern will match only
1146 comments with no leading whitespace. For example, to
1147 identify all comments as static block comments, one would use C<-sbcp=#>.
1148 To identify all left-adjusted comments as static block comments, use C<-sbcp='^#'>.
1150 Please note that B<-sbcp> merely defines the pattern used to identify static
1151 block comments; it will not be used unless the switch B<-sbc> is set. Also,
1152 please be aware that since this string is used in a perl regular expression
1153 which identifies these comments, it must enable a valid regular expression to
1156 =item B<-osbc>, B<--outdent-static-block-comments>
1158 The command B<-osbc> will will cause static block comments to be outdented by 2
1159 spaces (or whatever B<-ci=n> has been set to), if possible.
1163 =item Static Side Comments
1165 Static side comments are side comments with a special leading pattern.
1166 This option can be useful for controlling how commented code is displayed
1167 when it is a side comment.
1171 =item B<-ssc>, B<--static-side-comments>
1173 When B<-ssc> is used, a side comment with a static leading pattern, which is
1174 C<##> by default, will be be spaced only a single space from previous
1175 character, and it will not be vertically aligned with other side comments.
1177 The default is B<-nssc>.
1179 =item B<-sscp=string>, B<--static-side-comment-prefix=string>
1181 This parameter defines the prefix used to identify static side comments
1182 when the B<-ssc> parameter is set. The default prefix is C<##>,
1183 corresponding to C<-sscp=##>.
1185 Please note that B<-sscp> merely defines the pattern used to identify
1186 static side comments; it will not be used unless the switch B<-ssc> is
1187 set. Also, note that this string is used in a perl regular expression
1188 which identifies these comments, so it must enable a valid regular
1189 expression to be formed.
1196 =head2 Skipping Selected Sections of Code
1198 Selected lines of code may be passed verbatim to the output without any
1199 formatting. This feature is enabled by default but can be disabled with
1200 the B<--noformat-skipping> or B<-nfs> flag. It should be used sparingly to
1201 avoid littering code with markers, but it might be helpful for working
1202 around occasional problems. For example it might be useful for keeping
1203 the indentation of old commented code unchanged, keeping indentation of
1204 long blocks of aligned comments unchanged, keeping certain list
1205 formatting unchanged, or working around a glitch in perltidy.
1209 =item B<-fs>, B<--format-skipping>
1211 This flag, which is enabled by default, causes any code between
1212 special beginning and ending comment markers to be passed to the
1213 output without formatting. The default beginning marker is #<<<
1214 and the default ending marker is #>>> but they
1215 may be changed (see next items below). Additional text may appear on
1216 these special comment lines provided that it is separated from the
1217 marker by at least one space. For example
1219 #<<< do not let perltidy touch this
1227 The comment markers may be placed at any location that a block comment may
1228 appear. If they do not appear to be working, use the -log flag and examine the
1229 F<.LOG> file. Use B<-nfs> to disable this feature.
1231 =item B<-fsb=string>, B<--format-skipping-begin=string>
1233 The B<-fsb=string> parameter may be used to change the beginning marker for
1234 format skipping. The default is equivalent to -fsb='#<<<'. The string that
1235 you enter must begin with a # and should be in quotes as necessary to get past
1236 the command shell of your system. It is actually the leading text of a pattern
1237 that is constructed by appending a '\s', so you must also include backslashes
1238 for characters to be taken literally rather than as patterns.
1240 Some examples show how example strings become patterns:
1242 -fsb='#\{\{\{' becomes /^#\{\{\{\s/ which matches #{{{ but not #{{{{
1243 -fsb='#\*\*' becomes /^#\*\*\s/ which matches #** but not #***
1244 -fsb='#\*{2,}' becomes /^#\*{2,}\s/ which matches #** and #*****
1246 =item B<-fse=string>, B<--format-skipping-end=string>
1248 The B<-fsb=string> is the corresponding parameter used to change the
1249 ending marker for format skipping. The default is equivalent to
1254 =head2 Line Break Control
1258 =item B<-fnl>, B<--freeze-newlines>
1260 If you do not want any changes to the line breaks in your script, set
1261 B<-fnl>, and they will remain fixed, and the rest of the commands in
1262 this section and sections
1263 L<Controlling List Formatting>,
1264 L<Retaining or Ignoring Existing Line Breaks>, and
1265 L<Blank Line Control> will be ignored. You may want to use B<-noll>
1268 =item B<-ce>, B<--cuddled-else>
1270 Enable the "cuddled else" style, in which C<else> and C<elsif> are
1271 follow immediately after the curly brace closing the previous block.
1272 The default is not to use cuddled elses, and is indicated with the flag
1273 B<-nce> or B<--nocuddled-else>. Here is a comparison of the
1285 else { # -nce (default)
1289 =item B<-bl>, B<--opening-brace-on-new-line>
1291 Use the flag B<-bl> to place the opening brace on a new line:
1293 if ( $input_file eq '-' ) # -bl
1295 important_function();
1298 This flag applies to all structural blocks, including sub's (unless
1299 the B<-sbl> flag is set -- see next item).
1301 The default style, B<-nbl>, places an opening brace on the same line as
1302 the keyword introducing it. For example,
1304 if ( $input_file eq '-' ) { # -nbl (default)
1306 =item B<-sbl>, B<--opening-sub-brace-on-new-line>
1308 The flag B<-sbl> can be used to override the value of B<-bl> for
1309 opening sub braces. For example,
1313 produces this result:
1317 if (!defined($_[0])) {
1318 print("Hello, World\n");
1325 This flag is negated with B<-nsbl>. If B<-sbl> is not specified,
1326 the value of B<-bl> is used.
1328 =item B<-bli>, B<--brace-left-and-indent>
1330 The flag B<-bli> is the same as B<-bl> but in addition it causes one
1331 unit of continuation indentation ( see B<-ci> ) to be placed before
1332 an opening and closing block braces.
1336 if ( $input_file eq '-' ) # -bli
1338 important_function();
1341 By default, this extra indentation occurs for blocks of type:
1342 B<if>, B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>,
1343 B<while>, B<until>, and also with a preceding label. The next item
1344 shows how to change this.
1346 =item B<-blil=s>, B<--brace-left-and-indent-list=s>
1348 Use this parameter to change the types of block braces for which the
1349 B<-bli> flag applies; see L<Specifying Block Types>. For example,
1350 B<-blil='if elsif else'> would apply it to only C<if/elsif/else> blocks.
1352 =item B<-bar>, B<--opening-brace-always-on-right>
1354 The default style, B<-nbl> places the opening code block brace on a new
1355 line if it does not fit on the same line as the opening keyword, like
1358 if ( $bigwasteofspace1 && $bigwasteofspace2
1359 || $bigwasteofspace3 && $bigwasteofspace4 )
1361 big_waste_of_time();
1364 To force the opening brace to always be on the right, use the B<-bar>
1365 flag. In this case, the above example becomes
1367 if ( $bigwasteofspace1 && $bigwasteofspace2
1368 || $bigwasteofspace3 && $bigwasteofspace4 ) {
1369 big_waste_of_time();
1372 A conflict occurs if both B<-bl> and B<-bar> are specified.
1374 =item B<-otr>, B<--opening-token-right> and related flags
1376 The B<-otr> flag is a hint that perltidy should not place a break between a
1377 comma and an opening token. For example:
1379 # default formatting
1380 push @{ $self->{$module}{$key} },
1382 accno => $ref->{accno},
1383 description => $ref->{description}
1387 push @{ $self->{$module}{$key} }, {
1388 accno => $ref->{accno},
1389 description => $ref->{description}
1392 The flag B<-otr> is actually a synonym for three other flags
1393 which can be used to control parens, hash braces, and square brackets
1394 separately if desired:
1396 -opr or --opening-paren-right
1397 -ohbr or --opening-hash-brace-right
1398 -osbr or --opening-square-bracket-right
1400 =item Vertical tightness of non-block curly braces, parentheses, and square brackets.
1402 These parameters control what shall be called vertical tightness. Here are the
1409 Opening tokens (except for block braces) are controlled by B<-vt=n>, or
1410 B<--vertical-tightness=n>, where
1412 -vt=0 always break a line after opening token (default).
1413 -vt=1 do not break unless this would produce more than one
1414 step in indentation in a line.
1415 -vt=2 never break a line after opening token
1419 You must also use the B<-lp> flag when you use the B<-vt> flag; the
1420 reason is explained below.
1424 Closing tokens (except for block braces) are controlled by B<-vtc=n>, or
1425 B<--vertical-tightness-closing=n>, where
1427 -vtc=0 always break a line before a closing token (default),
1428 -vtc=1 do not break before a closing token which is followed
1429 by a semicolon or another closing token, and is not in
1431 -vtc=2 never break before a closing token.
1433 The rules for B<-vtc=1> are designed to maintain a reasonable balance
1434 between tightness and readability in complex lists.
1438 Different controls may be applied to to different token types,
1439 and it is also possible to control block braces; see below.
1443 Finally, please note that these vertical tightness flags are merely
1444 hints to the formatter, and it cannot always follow them. Things which
1445 make it difficult or impossible include comments, blank lines, blocks of
1446 code within a list, and possibly the lack of the B<-lp> parameter.
1447 Also, these flags may be ignored for very small lists (2 or 3 lines in
1452 Here are some examples:
1454 # perltidy -lp -vt=0 -vtc=0
1462 # perltidy -lp -vt=1 -vtc=0
1463 %romanNumerals = ( one => 'I',
1469 # perltidy -lp -vt=1 -vtc=1
1470 %romanNumerals = ( one => 'I',
1475 The difference between B<-vt=1> and B<-vt=2> is shown here:
1477 # perltidy -lp -vt=1
1479 mysprintf( "(void)find_threadsv(%s);",
1480 cstring( $threadsv_names[ $op->targ ] )
1484 # perltidy -lp -vt=2
1485 $init->add( mysprintf( "(void)find_threadsv(%s);",
1486 cstring( $threadsv_names[ $op->targ ] )
1490 With B<-vt=1>, the line ending in C<add(> does not combine with the next
1491 line because the next line is not balanced. This can help with
1492 readability, but B<-vt=2> can be used to ignore this rule.
1494 The tightest, and least readable, code is produced with both C<-vt=2> and
1497 # perltidy -lp -vt=2 -vtc=2
1498 $init->add( mysprintf( "(void)find_threadsv(%s);",
1499 cstring( $threadsv_names[ $op->targ ] ) ) );
1501 Notice how the code in all of these examples collapses vertically as
1502 B<-vt> increases, but the indentation remains unchanged. This is
1503 because perltidy implements the B<-vt> parameter by first formatting as
1504 if B<-vt=0>, and then simply overwriting one output line on top of the
1505 next, if possible, to achieve the desired vertical tightness. The
1506 B<-lp> indentation style has been designed to allow this vertical
1507 collapse to occur, which is why it is required for the B<-vt> parameter.
1509 The B<-vt=n> and B<-vtc=n> parameters apply to each type of container
1510 token. If desired, vertical tightness controls can be applied
1511 independently to each of the closing container token types.
1513 The parameters for controlling parentheses are B<-pvt=n> or
1514 B<--paren-vertical-tightness=n>, and B<-pcvt=n> or
1515 B<--paren-vertical-tightness-closing=n>.
1517 Likewise, the parameters for square brackets are B<-sbvt=n> or
1518 B<--square-bracket-vertical-tightness=n>, and B<-sbcvt=n> or
1519 B<--square-bracket-vertical-tightness-closing=n>.
1521 Finally, the parameters for controlling non-code block braces are
1522 B<-bvt=n> or B<--brace-vertical-tightness=n>, and B<-bcvt=n> or
1523 B<--brace-vertical-tightness-closing=n>.
1525 In fact, the parameter B<-vt=n> is actually just an abbreviation for
1526 B<-pvt=n -bvt=n sbvt=n>, and likewise B<-vtc=n> is an abbreviation
1527 for B<-pvtc=n -bvtc=n sbvtc=n>.
1529 =item B<-bbvt=n> or B<--block-brace-vertical-tightness=n>
1531 The B<-bbvt=n> flag is just like the B<-vt=n> flag but applies
1532 to opening code block braces.
1534 -bbvt=0 break after opening block brace (default).
1535 -bbvt=1 do not break unless this would produce more than one
1536 step in indentation in a line.
1537 -bbvt=2 do not break after opening block brace.
1539 It is necessary to also use either B<-bl> or B<-bli> for this to work,
1540 because, as with other vertical tightness controls, it is implemented by
1541 simply overwriting a line ending with an opening block brace with the
1542 subsequent line. For example:
1544 # perltidy -bli -bbvt=0
1545 if ( open( FILE, "< $File" ) )
1547 while ( $File = <FILE> )
1555 # perltidy -bli -bbvt=1
1556 if ( open( FILE, "< $File" ) )
1557 { while ( $File = <FILE> )
1564 By default this applies to blocks associated with keywords B<if>,
1565 B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>, B<while>,
1566 B<until>, and also with a preceding label. This can be changed with
1567 the parameter B<-bbvtl=string>, or
1568 B<--block-brace-vertical-tightness-list=string>, where B<string> is a
1569 space-separated list of block types. For more information on the
1570 possible values of this string, see L<Specifying Block Types>
1572 For example, if we want to just apply this style to C<if>,
1573 C<elsif>, and C<else> blocks, we could use
1574 C<perltidy -bli -bbvt=1 -bbvtl='if elsif else'>.
1576 There is no vertical tightness control for closing block braces; with
1577 the exception of one-line blocks, they will normally remain on a
1580 =item B<-sot>, B<--stack-opening-token> and related flags
1582 The B<-sot> flag tells perltidy to "stack" opening tokens
1583 when possible to avoid lines with isolated opening tokens.
1588 $opt_c = Text::CSV_XS->new(
1597 $opt_c = Text::CSV_XS->new( {
1604 For detailed control of individual closing tokens the following
1605 controls can be used:
1607 -sop or --stack-opening-paren
1608 -sohb or --stack-opening-hash-brace
1609 -sosb or --stack-opening-square-bracket
1611 The flag B<-sot> is a synonym for B<-sop -sohb -sosb>.
1613 =item B<-sct>, B<--stack-closing-token> and related flags
1615 The B<-sct> flag tells perltidy to "stack" closing tokens
1616 when possible to avoid lines with isolated closing tokens.
1621 $opt_c = Text::CSV_XS->new(
1630 $opt_c = Text::CSV_XS->new(
1637 The B<-sct> flag is somewhat similar to the B<-vtc> flags, and in some
1638 cases it can give a similar result. The difference is that the B<-vtc>
1639 flags try to avoid lines with leading opening tokens by "hiding" them at
1640 the end of a previous line, whereas the B<-sct> flag merely tries to
1641 reduce the number of lines with isolated closing tokens by stacking them
1642 but does not try to hide them. For example:
1645 $opt_c = Text::CSV_XS->new(
1649 always_quote => 1, } );
1651 For detailed control of the stacking of individual closing tokens the
1652 following controls can be used:
1654 -scp or --stack-closing-paren
1655 -schb or --stack-closing-hash-brace
1656 -scsb or --stack-closing-square-bracket
1658 The flag B<-sct> is a synonym for B<-scp -schb -scsb>.
1660 =item B<-dnl>, B<--delete-old-newlines>
1662 By default, perltidy first deletes all old line break locations, and then it
1663 looks for good break points to match the desired line length. Use B<-ndnl>
1664 or B<--nodelete-old-newlines> to force perltidy to retain all old line break
1667 =item B<-anl>, B<--add-newlines>
1669 By default, perltidy will add line breaks when necessary to create
1670 continuations of long lines and to improve the script appearance. Use
1671 B<-nanl> or B<--noadd-newlines> to prevent any new line breaks.
1673 This flag does not prevent perltidy from eliminating existing line
1674 breaks; see B<--freeze-newlines> to completely prevent changes to line
1677 =item Controlling whether perltidy breaks before or after operators
1679 Two command line parameters provide some control over whether
1680 a line break should be before or after specific token types.
1682 B<-wba=s> or B<--want-break-after=s>, and
1684 B<-wbb=s> or B<--want-break-before=s>.
1686 These parameters are each followed by a quoted string, B<s>, containing
1687 a list of token types (separated only by spaces). No more than one of each
1688 of these parameters should be specified, because repeating a
1689 command-line parameter always overwrites the previous one before
1690 perltidy ever sees it.
1692 By default, perltidy breaks B<after> these token types:
1693 % + - * / x != == >= <= =~ !~ < > | & >= <
1694 = **= += *= &= <<= &&= -= /= |= >>= ||= .= %= ^= x=
1696 And perltidy breaks B<before> these token types by default:
1699 To illustrate, to cause a break after a concatenation operator, C<'.'>,
1700 rather than before it, the command line would be
1704 As another example, the following command would cause a break before
1705 math operators C<'+'>, C<'-'>, C<'/'>, and C<'*'>:
1709 These commands should work well for most of the token types that perltidy uses
1710 (use B<--dump-token-types> for a list). Also try the -D flag on a short
1711 snippet of code and look at the .DEBUG file to see the tokenization. However,
1712 for a few token types there may be conflicts with hardwired logic which cause
1713 unexpected results. One example is curly braces, which should be controlled
1714 with the parameter B<bl> provided for that purpose.
1716 B<WARNING> Be sure to put these tokens in quotes to avoid having them
1717 misinterpreted by your command shell.
1721 =head2 Controlling List Formatting
1723 Perltidy attempts to place comma-separated arrays of values in tables
1724 which look good. Its default algorithms usually work well, and they
1725 have been improving with each release, but several parameters are
1726 available to control list formatting.
1730 =item B<-boc>, B<--break-at-old-comma-breakpoints>
1732 This flag tells perltidy to try to break at all old commas. This is not
1733 the default. Normally, perltidy makes a best guess at list formatting,
1734 and seldom uses old comma breakpoints. Usually this works well,
1743 The default formatting will flatten this down to one line:
1745 # perltidy (default)
1746 my @list = ( 1, 1, 1, 1, 2, 1, 1, 3, 3, 1, 1, 4, 6, 4, 1, );
1748 which hides the structure. Using B<-boc>, plus additional flags
1749 to retain the original style, yields
1751 # perltidy -boc -lp -pt=2 -vt=1 -vtc=1
1758 A disadvantage of this flag is that all tables in the file
1759 must already be nicely formatted. For another possibility see
1760 the -fs flag in L<Skipping Selected Sections of Code>.
1762 =item B<-mft=n>, B<--maximum-fields-per-table=n>
1764 If the computed number of fields for any table exceeds B<n>, then it
1765 will be reduced to B<n>. The default value for B<n> is a large number,
1766 40. While this value should probably be left unchanged as a general
1767 rule, it might be used on a small section of code to force a list to
1768 have a particular number of fields per line, and then either the B<-boc>
1769 flag could be used to retain this formatting, or a single comment could
1770 be introduced somewhere to freeze the formatting in future applications
1783 =item B<-cab=n>, B<--comma-arrow-breakpoints=n>
1785 A comma which follows a comma arrow, '=>', requires special
1786 consideration. In a long list, it is common to break at all such
1787 commas. This parameter can be used to control how perltidy breaks at
1788 these commas. (However, it will have no effect if old comma breaks are
1789 being forced because B<-boc> is used). The possible values of B<n> are:
1791 n=0 break at all commas after =>
1792 n=1 stable: break at all commas after => unless this would break
1793 an existing one-line container (default)
1794 n=2 break at all commas after =>, but try to form the maximum
1795 maximum one-line container lengths
1796 n=3 do not treat commas after => specially at all
1798 For example, given the following single line, perltidy by default will
1799 not add any line breaks because it would break the existing one-line
1802 bless { B => $B, Root => $Root } => $package;
1804 Using B<-cab=0> will force a break after each comma-arrow item:
1812 If perltidy is subsequently run with this container broken, then by
1813 default it will break after each '=>' because the container is now
1814 broken. To reform a one-line container, the parameter B<-cab=2> would
1817 The flag B<-cab=3> can be used to prevent these commas from being
1818 treated specially. In this case, an item such as "01" => 31 is
1819 treated as a single item in a table. The number of fields in this table
1820 will be determined by the same rules that are used for any other table.
1825 "01" => 31, "02" => 29, "03" => 31, "04" => 30,
1826 "05" => 31, "06" => 30, "07" => 31, "08" => 31,
1827 "09" => 30, "10" => 31, "11" => 30, "12" => 31
1832 =head2 Retaining or Ignoring Existing Line Breaks
1834 Several additional parameters are available for controlling the extent
1835 to which line breaks in the input script influence the output script.
1836 In most cases, the default parameter values are set so that, if a choice
1837 is possible, the output style follows the input style. For example, if
1838 a short logical container is broken in the input script, then the
1839 default behavior is for it to remain broken in the output script.
1841 Most of the parameters in this section would only be required for a
1842 one-time conversion of a script from short container lengths to longer
1843 container lengths. The opposite effect, of converting long container
1844 lengths to shorter lengths, can be obtained by temporarily using a short
1845 maximum line length.
1849 =item B<-bol>, B<--break-at-old-logical-breakpoints>
1851 By default, if a logical expression is broken at a C<&&>, C<||>, C<and>,
1852 or C<or>, then the container will remain broken. Also, breaks
1853 at internal keywords C<if> and C<unless> will normally be retained.
1854 To prevent this, and thus form longer lines, use B<-nbol>.
1856 =item B<-bok>, B<--break-at-old-keyword-breakpoints>
1858 By default, perltidy will retain a breakpoint before keywords which may
1859 return lists, such as C<sort> and <map>. This allows chains of these
1860 operators to be displayed one per line. Use B<-nbok> to prevent
1861 retaining these breakpoints.
1863 =item B<-bot>, B<--break-at-old-ternary-breakpoints>
1865 By default, if a conditional (ternary) operator is broken at a C<:>,
1866 then it will remain broken. To prevent this, and thereby
1867 form longer lines, use B<-nbot>.
1869 =item B<-iob>, B<--ignore-old-breakpoints>
1871 Use this flag to tell perltidy to ignore existing line breaks to the
1872 maximum extent possible. This will tend to produce the longest possible
1873 containers, regardless of type, which do not exceed the line length
1878 =head2 Blank Line Control
1880 Blank lines can improve the readability of a script if they are carefully
1881 placed. Perltidy has several commands for controlling the insertion,
1882 retention, and removal of blank lines.
1886 =item B<-bbc>, B<--blanks-before-comments>
1888 A blank line will be introduced before a full-line comment. This is the
1889 default. Use B<-nbbc> or B<--noblanks-before-comments> to prevent
1890 such blank lines from being introduced.
1892 =item B<-bbs>, B<--blanks-before-subs>
1894 A blank line will be introduced before a B<sub> definition, unless it is a
1895 one-liner or preceded by a comment. A blank line will also be introduced
1896 before a B<package> statement and a B<BEGIN> and B<END> block. This is the
1897 default. The intention is to help display the structure of a program by
1898 setting off certain key sections of code. This is negated with B<-nbbs> or
1899 B<--noblanks-before-subs>.
1901 =item B<-bbb>, B<--blanks-before-blocks>
1903 A blank line will be introduced before blocks of coding delimited by
1904 B<for>, B<foreach>, B<while>, B<until>, and B<if>, B<unless>, in the following
1911 The block is not preceded by a comment.
1915 The block is not a one-line block.
1919 The number of consecutive non-blank lines at the current indentation depth is at least B<-lbl>
1924 This is the default. The intention of this option is to introduce
1925 some space within dense coding.
1926 This is negated with B<-nbbb> or B<--noblanks-before-blocks>.
1928 =item B<-lbl=n> B<--long-block-line-count=n>
1930 This controls how often perltidy is allowed to add blank lines before
1931 certain block types (see previous section). The default is 8. Entering
1932 a value of B<0> is equivalent to entering a very large number.
1934 =item B<-mbl=n> B<--maximum-consecutive-blank-lines=n>
1936 This parameter specifies the maximum number of consecutive blank lines
1937 in the output script. The default is n=1. If the input file has more
1938 than n consecutive blank lines, the number will be reduced to n.
1939 (This obviously does not apply to pod sections, here-documents, and quotes).
1941 =item B<-sob>, B<--swallow-optional-blank-lines>
1943 All blank lines not required by the above flags, B<-bbb>, B<-bbs>, and B<-bbc>,
1944 will be deleted. (But essential blank lines above pod documents will be
1945 retained). This is NOT the default.
1947 =item B<-nsob>, B<--noswallow-optional-blank-lines>
1949 Retain blank lines, including those which do not corresponding to flags
1950 B<-bbb>, B<-bbs>, and B<-bbc>. This is the default. The number of
1951 blanks retained is subject to the limit imposed by
1952 B<--maximum-consecutive-blank-lines>, however.
1958 A style refers to a convenient collection of existing parameters.
1962 =item B<-gnu>, B<--gnu-style>
1964 B<-gnu> gives an approximation to the GNU Coding Standards (which do
1965 not apply to perl) as they are sometimes implemented. At present, this
1966 style overrides the default style with the following parameters:
1968 -lp -bl -noll -pt=2 -bt=2 -sbt=2 -icp
1970 =item B<-pbp>, B<--perl-best-practices>
1972 B<-pbp> is an abbreviation for the parameters in the book B<Perl Best Practices>
1975 -l=78 -i=4 -ci=4 -st -se -vt=2 -cti=0 -pt=1 -bt=1 -sbt=1 -bbt=1 -nsfs -nolq
1976 -wbb="% + - * / x != == >= <= =~ !~ < > | & >= < =
1977 **= += *= &= <<= &&= -= /= |= >>= ||= .= %= ^= x="
1979 Note that the -st and -se flags make perltidy act as a filter on one file only.
1980 These can be overridden with -nst and -nse if necessary.
1984 =head2 Other Controls
1988 =item Deleting selected text
1990 Perltidy can selectively delete comments and/or pod documentation. The
1991 command B<-dac> or B<--delete-all-comments> will delete all comments
1992 B<and> all pod documentation, leaving just code and any leading system
1995 The command B<-dp> or B<--delete-pod> will remove all pod documentation
1998 Two commands which remove comments (but not pod) are: B<-dbc> or
1999 B<--delete-block-comments> and B<-dsc> or B<--delete-side-comments>.
2000 (Hanging side comments will be deleted with block comments here.)
2002 The negatives of these commands also work, and are the defaults. When
2003 block comments are deleted, any leading 'hash-bang' will be retained.
2004 Also, if the B<-x> flag is used, any system commands before a leading
2005 hash-bang will be retained (even if they are in the form of comments).
2007 =item Writing selected text to a file
2009 When perltidy writes a formatted text file, it has the ability to also
2010 send selected text to a file with a F<.TEE> extension. This text can
2011 include comments and pod documentation.
2013 The command B<-tac> or B<--tee-all-comments> will write all comments
2014 B<and> all pod documentation.
2016 The command B<-tp> or B<--tee-pod> will write all pod documentation (but
2019 The commands which write comments (but not pod) are: B<-tbc> or
2020 B<--tee-block-comments> and B<-tsc> or B<--tee-side-comments>.
2021 (Hanging side comments will be written with block comments here.)
2023 The negatives of these commands also work, and are the defaults.
2025 =item Using a F<.perltidyrc> command file
2027 If you use perltidy frequently, you probably won't be happy until you
2028 create a F<.perltidyrc> file to avoid typing commonly-used parameters.
2029 Perltidy will first look in your current directory for a command file
2030 named F<.perltidyrc>. If it does not find one, it will continue looking
2031 for one in other standard locations.
2033 These other locations are system-dependent, and may be displayed with
2034 the command C<perltidy -dpro>. Under Unix systems, it will look for a
2035 F<.perltidyrc> file in the home directory, and then for a system-wide
2036 file F</usr/local/etc/perltidyrc>, and then it will look for
2037 F</etc/perltidyrc>. Note that these last two system-wide files do not
2038 have a leading dot. Further system-dependent information will be found
2039 in the INSTALL file distributed with perltidy.
2041 This file is free format, and simply a list of parameters, just as they
2042 would be entered on a command line. Any number of lines may be used,
2043 with any number of parameters per line, although it may be easiest to
2044 read with one parameter per line. Blank lines are ignored, and text
2045 after a '#' is ignored to the end of a line.
2047 Here is an example of a F<.perltidyrc> file:
2049 # This is a simple of a .perltidyrc configuration file
2050 # This implements a highly spaced style
2051 -se # errors to standard error output
2052 -w # show all warnings
2053 -bl # braces on new lines
2054 -pt=0 # parens not tight at all
2055 -bt=0 # braces not tight
2056 -sbt=0 # square brackets not tight
2058 The parameters in the F<.perltidyrc> file are installed first, so any
2059 parameters given on the command line will have priority over them.
2061 To avoid confusion, perltidy ignores any command in the .perltidyrc
2062 file which would cause some kind of dump and an exit. These are:
2064 -h -v -ddf -dln -dop -dsn -dtt -dwls -dwrs -ss
2066 There are several options may be helpful in debugging a F<.perltidyrc>
2073 A very helpful command is B<--dump-profile> or B<-dpro>. It writes a
2074 list of all configuration filenames tested to standard output, and
2075 if a file is found, it dumps the content to standard output before
2076 exiting. So, to find out where perltidy looks for its configuration
2077 files, and which one if any it selects, just enter
2083 It may be simplest to develop and test configuration files with
2084 alternative names, and invoke them with B<-pro=filename> on the command
2085 line. Then rename the desired file to F<.perltidyrc> when finished.
2089 The parameters in the F<.perltidyrc> file can be switched off with
2090 the B<-npro> option.
2094 The commands B<--dump-options>, B<--dump-defaults>, B<--dump-long-names>,
2095 and B<--dump-short-names>, all described below, may all be helpful.
2099 =item Creating a new abbreviation
2101 A special notation is available for use in a F<.perltidyrc> file
2102 for creating an abbreviation for a group
2103 of options. This can be used to create a
2104 shorthand for one or more styles which are frequently, but not always,
2105 used. The notation is to group the options within curly braces which
2106 are preceded by the name of the alias (without leading dashes), like this:
2113 where B<newword> is the abbreviation, and B<opt1>, etc, are existing parameters
2114 I<or other abbreviations>. The main syntax requirement is that
2115 the new abbreviation must begin on a new line.
2116 Space before and after the curly braces is optional.
2118 specific example, the following line
2120 airy {-bl -pt=0 -bt=0 -sbt=0}
2122 could be placed in a F<.perltidyrc> file, and then invoked at will with
2124 perltidy -airy somefile.pl
2126 (Either C<-airy> or C<--airy> may be used).
2128 =item Skipping leading non-perl commands with B<-x> or B<--look-for-hash-bang>
2130 If your script has leading lines of system commands or other text which
2131 are not valid perl code, and which are separated from the start of the
2132 perl code by a "hash-bang" line, ( a line of the form C<#!...perl> ),
2133 you must use the B<-x> flag to tell perltidy not to parse and format any
2134 lines before the "hash-bang" line. This option also invokes perl with a
2135 -x flag when checking the syntax. This option was originally added to
2136 allow perltidy to parse interactive VMS scripts, but it should be used
2137 for any script which is normally invoked with C<perl -x>.
2139 =item Making a file unreadable
2141 The goal of perltidy is to improve the readability of files, but there
2142 are two commands which have the opposite effect, B<--mangle> and
2143 B<--extrude>. They are actually
2144 merely aliases for combinations of other parameters. Both of these
2145 strip all possible whitespace, but leave comments and pod documents,
2146 so that they are essentially reversible. The
2147 difference between these is that B<--mangle> puts the fewest possible
2148 line breaks in a script while B<--extrude> puts the maximum possible.
2149 Note that these options do not provided any meaningful obfuscation, because
2150 perltidy can be used to reformat the files. They were originally
2151 developed to help test the tokenization logic of perltidy, but they
2153 One use for B<--mangle> is the following:
2155 perltidy --mangle myfile.pl -st | perltidy -o myfile.pl.new
2157 This will form the maximum possible number of one-line blocks (see next
2158 section), and can sometimes help clean up a badly formatted script.
2160 A similar technique can be used with B<--extrude> instead of B<--mangle>
2161 to make the minimum number of one-line blocks.
2163 Another use for B<--mangle> is to combine it with B<-dac> to reduce
2164 the file size of a perl script.
2166 =item One-line blocks
2168 There are a few points to note regarding one-line blocks. A one-line
2169 block is something like this,
2171 if ($x > 0) { $y = 1 / $x }
2173 where the contents within the curly braces is short enough to fit
2176 With few exceptions, perltidy retains existing one-line blocks, if it
2177 is possible within the line-length constraint, but it does not attempt
2178 to form new ones. In other words, perltidy will try to follow the
2179 one-line block style of the input file.
2181 If an existing one-line block is longer than the maximum line length,
2182 however, it will be broken into multiple lines. When this happens, perltidy
2183 checks for and adds any optional terminating semicolon (unless the B<-nasc>
2184 option is used) if the block is a code block.
2186 The main exception is that perltidy will attempt to form new one-line
2187 blocks following the keywords C<map>, C<eval>, and C<sort>, because
2188 these code blocks are often small and most clearly displayed in a single
2191 One-line block rules can conflict with the cuddled-else option. When
2192 the cuddled-else option is used, perltidy retains existing one-line
2193 blocks, even if they do not obey cuddled-else formatting.
2195 Occasionally, when one-line blocks get broken because they exceed the
2196 available line length, the formatting will violate the requested brace style.
2197 If this happens, reformatting the script a second time should correct
2202 The following flags are available for debugging:
2204 B<--dump-defaults> or B<-ddf> will write the default option set to standard output and quit
2206 B<--dump-profile> or B<-dpro> will write the name of the current
2207 configuration file and its contents to standard output and quit.
2209 B<--dump-options> or B<-dop> will write current option set to standard
2212 B<--dump-long-names> or B<-dln> will write all command line long names (passed
2213 to Get_options) to standard output and quit.
2215 B<--dump-short-names> or B<-dsn> will write all command line short names
2216 to standard output and quit.
2218 B<--dump-token-types> or B<-dtt> will write a list of all token types
2219 to standard output and quit.
2221 B<--dump-want-left-space> or B<-dwls> will write the hash %want_left_space
2222 to standard output and quit. See the section on controlling whitespace
2225 B<--dump-want-right-space> or B<-dwrs> will write the hash %want_right_space
2226 to standard output and quit. See the section on controlling whitespace
2229 B<-DEBUG> will write a file with extension F<.DEBUG> for each input file
2230 showing the tokenization of all lines of code.
2232 =item Working with MakeMaker, AutoLoader and SelfLoader
2234 The first $VERSION line of a file which might be eval'd by MakeMaker
2235 is passed through unchanged except for indentation.
2236 Use B<--nopass-version-line>, or B<-npvl>, to deactivate this feature.
2238 If the AutoLoader module is used, perltidy will continue formatting
2239 code after seeing an __END__ line.
2240 Use B<--nolook-for-autoloader>, or B<-nlal>, to deactivate this feature.
2242 Likewise, if the SelfLoader module is used, perltidy will continue formatting
2243 code after seeing a __DATA__ line.
2244 Use B<--nolook-for-selfloader>, or B<-nlsl>, to deactivate this feature.
2246 =item Working around problems with older version of Perl
2248 Perltidy contains a number of rules which help avoid known subtleties
2249 and problems with older versions of perl, and these rules always
2250 take priority over whatever formatting flags have been set. For example,
2251 perltidy will usually avoid starting a new line with a bareword, because
2252 this might cause problems if C<use strict> is active.
2254 There is no way to override these rules.
2262 =item The B<-html> master switch
2264 The flag B<-html> causes perltidy to write an html file with extension
2265 F<.html>. So, for example, the following command
2267 perltidy -html somefile.pl
2269 will produce a syntax-colored html file named F<somefile.pl.html>
2270 which may be viewed with a browser.
2272 B<Please Note>: In this case, perltidy does not do any formatting to the
2273 input file, and it does not write a formatted file with extension
2274 F<.tdy>. This means that two perltidy runs are required to create a
2275 fully reformatted, html copy of a script.
2277 =item The B<-pre> flag for code snippets
2279 When the B<-pre> flag is given, only the pre-formatted section, within
2280 the <PRE> and </PRE> tags, will be output. This simplifies inclusion
2281 of the output in other files. The default is to output a complete
2284 =item The B<-nnn> flag for line numbering
2286 When the B<-nnn> flag is given, the output lines will be numbered.
2288 =item The B<-toc>, or B<--html-table-of-contents> flag
2290 By default, a table of contents to packages and subroutines will be
2291 written at the start of html output. Use B<-ntoc> to prevent this.
2292 This might be useful, for example, for a pod document which contains a
2293 number of unrelated code snippets. This flag only influences the code
2294 table of contents; it has no effect on any table of contents produced by
2295 pod2html (see next item).
2297 =item The B<-pod>, or B<--pod2html> flag
2299 There are two options for formatting pod documentation. The default is
2300 to pass the pod through the Pod::Html module (which forms the basis of
2301 the pod2html utility). Any code sections are formatted by perltidy, and
2302 the results then merged. Note: perltidy creates a temporary file when
2303 Pod::Html is used; see L<"FILES">. Also, Pod::Html creates temporary
2304 files for its cache.
2306 NOTE: Perltidy counts the number of C<=cut> lines, and either moves the
2307 pod text to the top of the html file if there is one C<=cut>, or leaves
2308 the pod text in its original order (interleaved with code) otherwise.
2310 Most of the flags accepted by pod2html may be included in the perltidy
2311 command line, and they will be passed to pod2html. In some cases,
2312 the flags have a prefix C<pod> to emphasize that they are for the
2313 pod2html, and this prefix will be removed before they are passed to
2314 pod2html. The flags which have the additional C<pod> prefix are:
2316 --[no]podheader --[no]podindex --[no]podrecurse --[no]podquiet
2317 --[no]podverbose --podflush
2319 The flags which are unchanged from their use in pod2html are:
2321 --backlink=s --cachedir=s --htmlroot=s --libpods=s --title=s
2322 --podpath=s --podroot=s
2324 where 's' is an appropriate character string. Not all of these flags are
2325 available in older versions of Pod::Html. See your Pod::Html documentation for
2328 The alternative, indicated with B<-npod>, is not to use Pod::Html, but
2329 rather to format pod text in italics (or whatever the stylesheet
2330 indicates), without special html markup. This is useful, for example,
2331 if pod is being used as an alternative way to write comments.
2333 =item The B<-frm>, or B<--frames> flag
2335 By default, a single html output file is produced. This can be changed
2336 with the B<-frm> option, which creates a frame holding a table of
2337 contents in the left panel and the source code in the right side. This
2338 simplifies code browsing. Assume, for example, that the input file is
2339 F<MyModule.pm>. Then, for default file extension choices, these three
2340 files will be created:
2342 MyModule.pm.html - the frame
2343 MyModule.pm.toc.html - the table of contents
2344 MyModule.pm.src.html - the formatted source code
2346 Obviously this file naming scheme requires that output be directed to a real
2347 file (as opposed to, say, standard output). If this is not the
2348 case, or if the file extension is unknown, the B<-frm> option will be
2351 =item The B<-text=s>, or B<--html-toc-extension> flag
2353 Use this flag to specify the extra file extension of the table of contents file
2354 when html frames are used. The default is "toc".
2355 See L<Specifying File Extensions>.
2357 =item The B<-sext=s>, or B<--html-src-extension> flag
2359 Use this flag to specify the extra file extension of the content file when html
2360 frames are used. The default is "src".
2361 See L<Specifying File Extensions>.
2363 =item The B<-hent>, or B<--html-entities> flag
2365 This flag controls the use of Html::Entities for html formatting. By
2366 default, the module Html::Entities is used to encode special symbols.
2367 This may not be the right thing for some browser/language
2368 combinations. Use --nohtml-entities or -nhent to prevent this.
2372 Style sheets make it very convenient to control and adjust the
2373 appearance of html pages. The default behavior is to write a page of
2374 html with an embedded style sheet.
2376 An alternative to an embedded style sheet is to create a page with a
2377 link to an external style sheet. This is indicated with the
2378 B<-css=filename>, where the external style sheet is F<filename>. The
2379 external style sheet F<filename> will be created if and only if it does
2380 not exist. This option is useful for controlling multiple pages from a
2383 To cause perltidy to write a style sheet to standard output and exit,
2384 use the B<-ss>, or B<--stylesheet>, flag. This is useful if the style
2385 sheet could not be written for some reason, such as if the B<-pre> flag
2386 was used. Thus, for example,
2388 perltidy -html -ss >mystyle.css
2390 will write a style sheet with the default properties to file
2393 The use of style sheets is encouraged, but a web page without a style
2394 sheets can be created with the flag B<-nss>. Use this option if you
2395 must to be sure that older browsers (roughly speaking, versions prior to
2396 4.0 of Netscape Navigator and Internet Explorer) can display the
2397 syntax-coloring of the html files.
2399 =item Controlling HTML properties
2401 Note: It is usually more convenient to accept the default properties
2402 and then edit the stylesheet which is produced. However, this section
2403 shows how to control the properties with flags to perltidy.
2405 Syntax colors may be changed from their default values by flags of the either
2406 the long form, B<-html-color-xxxxxx=n>, or more conveniently the short form,
2407 B<-hcx=n>, where B<xxxxxx> is one of the following words, and B<x> is the
2408 corresponding abbreviation:
2411 ---------- -------- --
2414 identifier identifier i
2415 bareword, function bareword w
2417 quite, pattern quote q
2418 here doc text here-doc-text h
2419 here doc target here-doc-target hh
2420 punctuation punctuation pu
2422 structural braces structure s
2423 semicolon semicolon sc
2427 sub definition name subroutine m
2428 pod text pod-text pd
2430 A default set of colors has been defined, but they may be changed by providing
2431 values to any of the following parameters, where B<n> is either a 6 digit
2432 hex RGB color value or an ascii name for a color, such as 'red'.
2434 To illustrate, the following command will produce an html
2435 file F<somefile.pl.html> with "aqua" keywords:
2437 perltidy -html -hck=00ffff somefile.pl
2439 and this should be equivalent for most browsers:
2441 perltidy -html -hck=aqua somefile.pl
2443 Perltidy merely writes any non-hex names that it sees in the html file.
2444 The following 16 color names are defined in the HTML 3.2 standard:
2463 Many more names are supported in specific browsers, but it is safest
2464 to use the hex codes for other colors. Helpful color tables can be
2465 located with an internet search for "HTML color tables".
2467 Besides color, two other character attributes may be set: bold, and italics.
2468 To set a token type to use bold, use the flag
2469 B<--html-bold-xxxxxx> or B<-hbx>, where B<xxxxxx> or B<x> are the long
2470 or short names from the above table. Conversely, to set a token type to
2471 NOT use bold, use B<--nohtml-bold-xxxxxx> or B<-nhbx>.
2473 Likewise, to set a token type to use an italic font, use the flag
2474 B<--html-italic-xxxxxx> or B<-hix>, where again B<xxxxxx> or B<x> are the
2475 long or short names from the above table. And to set a token type to
2476 NOT use italics, use B<--nohtml-italic-xxxxxx> or B<-nhix>.
2478 For example, to use bold braces and lime color, non-bold, italics keywords the
2479 following command would be used:
2481 perltidy -html -hbs -hck=00FF00 -nhbk -hik somefile.pl
2483 The background color can be specified with B<--html-color-background=n>,
2484 or B<-hcbg=n> for short, where n is a 6 character hex RGB value. The
2485 default color of text is the value given to B<punctuation>, which is
2488 Here are some notes and hints:
2490 1. If you find a preferred set of these parameters, you may want
2491 to create a F<.perltidyrc> file containing them. See the perltidy man
2492 page for an explanation.
2494 2. Rather than specifying values for these parameters, it is probably
2495 easier to accept the defaults and then edit a style sheet. The style
2496 sheet contains comments which should make this easy.
2498 3. The syntax-colored html files can be very large, so it may be best to
2499 split large files into smaller pieces to improve download times.
2503 =head1 SOME COMMON INPUT CONVENTIONS
2505 =head2 Specifying Block Types
2507 Several parameters which refer to code block types may be customized by also
2508 specifying an associated list of block types. The type of a block is the name
2509 of the keyword which introduces that block, such as B<if>, B<else>, or B<sub>.
2510 An exception is a labeled block, which has no keyword, and should be specified
2513 For example, the following parameter specifies C<sub>, labels, C<BEGIN>, and
2516 -cscl="sub : BEGIN END"
2518 (the meaning of the -cscl parameter is described above.) Note that
2519 quotes are required around the list of block types because of the
2522 =head2 Specifying File Extensions
2524 Several parameters allow default file extensions to be overridden. For
2525 example, a backup file extension may be specified with B<-bext=ext>,
2526 where B<ext> is some new extension. In order to provides the user some
2527 flexibility, the following convention is used in all cases to decide if
2528 a leading '.' should be used. If the extension C<ext> begins with
2529 C<A-Z>, C<a-z>, or C<0-9>, then it will be appended to the filename with
2530 an intermediate '.' (or perhaps an '_' on VMS systems). Otherwise, it
2531 will be appended directly.
2533 For example, suppose the file is F<somefile.pl>. For C<-bext=old>, a '.' is
2534 added to give F<somefile.pl.old>. For C<-bext=.old>, no additional '.' is
2535 added, so again the backup file is F<somefile.pl.old>. For C<-bext=~>, then no
2536 dot is added, and the backup file will be F<somefile.pl~> .
2538 =head1 SWITCHES WHICH MAY BE NEGATED
2540 The following list shows all short parameter names which allow a prefix
2541 'n' to produce the negated form:
2543 D anl asc aws b bbb bbc bbs bl bli boc bok bol bot ce
2544 csc dac dbc dcsc ddf dln dnl dop dp dpro dsc dsm dsn dtt dwls
2545 dwrs dws f fll frm fs hsc html ibc icb icp iob isbc lal log
2546 lp lsl ohbr okw ola oll opr opt osbr otr ple ple pod pvl q
2547 sbc sbl schb scp scsb sct se sfp sfs skp sob sohb sop sosb sot
2548 ssc st sts syn t tac tbc toc tp tqw tsc w x
2550 Equivalently, the prefix 'no' or 'no-' on the corresponding long names may be
2557 =item Parsing Limitations
2559 Perltidy should work properly on most perl scripts. It does a lot of
2560 self-checking, but still, it is possible that an error could be
2561 introduced and go undetected. Therefore, it is essential to make
2562 careful backups and to test reformatted scripts.
2564 The main current limitation is that perltidy does not scan modules
2565 included with 'use' statements. This makes it necessary to guess the
2566 context of any bare words introduced by such modules. Perltidy has good
2567 guessing algorithms, but they are not infallible. When it must guess,
2568 it leaves a message in the log file.
2570 If you encounter a bug, please report it.
2572 =item What perltidy does not parse and format
2574 Perltidy indents but does not reformat comments and C<qw> quotes.
2575 Perltidy does not in any way modify the contents of here documents or
2576 quoted text, even if they contain source code. (You could, however,
2577 reformat them separately). Perltidy does not format 'format' sections
2578 in any way. And, of course, it does not modify pod documents.
2586 =item Temporary files
2588 Under the -html option with the default --pod2html flag, a temporary file is
2589 required to pass text to Pod::Html. Unix systems will try to use the POSIX
2590 tmpnam() function. Otherwise the file F<perltidy.TMP> will be temporarily
2591 created in the current working directory.
2593 =item Special files when standard input is used
2595 When standard input is used, the log file, if saved, is F<perltidy.LOG>,
2596 and any errors are written to F<perltidy.ERR> unless the B<-se> flag is
2597 set. These are saved in the current working directory.
2599 =item Files overwritten
2601 The following file extensions are used by perltidy, and files with these
2602 extensions may be overwritten or deleted: F<.ERR>, F<.LOG>, F<.TEE>,
2603 and/or F<.tdy>, F<.html>, and F<.bak>, depending on the run type and
2606 =item Files extensions limitations
2608 Perltidy does not operate on files for which the run could produce a file with
2609 a duplicated file extension. These extensions include F<.LOG>, F<.ERR>,
2610 F<.TEE>, and perhaps F<.tdy> and F<.bak>, depending on the run type. The
2611 purpose of this rule is to prevent generating confusing filenames such as
2612 F<somefile.tdy.tdy.tdy>.
2618 perlstyle(1), Perl::Tidy(3)
2622 This man page documents perltidy version 20060719.
2626 Michael Cartmell supplied code for adaptation to VMS and helped with
2629 Yves Orton supplied code for adaptation to the various versions
2632 Axel Rose supplied a patch for MacPerl.
2634 Hugh S. Myers designed and implemented the initial Perl::Tidy module interface.
2636 Many others have supplied key ideas, suggestions, and bug reports;
2637 see the CHANGES file.
2642 email: perltidy at users.sourceforge.net
2643 http://perltidy.sourceforge.net
2647 Copyright (c) 2000-2006 by Steve Hancock
2651 This package is free software; you can redistribute it and/or modify it
2652 under the terms of the "GNU General Public License".
2654 Please refer to the file "COPYING" for details.
2658 This package is distributed in the hope that it will be useful,
2659 but WITHOUT ANY WARRANTY; without even the implied warranty of
2660 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
2662 See the "GNU General Public License" for more details.