6 my $arg_string = undef;
8 # give Macs a chance to provide command line parameters
10 $arg_string = MacPerl::Ask(
11 'Please enter @ARGV (-h for help)',
12 defined $ARGV[0] ? "\"$ARGV[0]\"" : ""
16 Perl::Tidy::perltidy( argv => $arg_string );
22 perltidy - a perl script indenter and reformatter
26 perltidy [ options ] file1 file2 file3 ...
27 (output goes to file1.tdy, file2.tdy, file3.tdy, ...)
28 perltidy [ options ] file1 -o outfile
29 perltidy [ options ] file1 -st >outfile
30 perltidy [ options ] <infile >outfile
34 Perltidy reads a perl script and writes an indented, reformatted script.
36 Many users will find enough information in L<"EXAMPLES"> to get
37 started. New users may benefit from the short tutorial
39 http://perltidy.sourceforge.net/tutorial.html
41 A convenient aid to systematically defining a set of style parameters
43 http://perltidy.sourceforge.net/stylekey.html
45 Perltidy can produce output on either of two modes, depending on the
46 existence of an B<-html> flag. Without this flag, the output is passed
47 through a formatter. The default formatting tries to follow the
48 recommendations in perlstyle(1), but it can be controlled in detail with
49 numerous input parameters, which are described in L<"FORMATTING
52 When the B<-html> flag is given, the output is passed through an HTML
53 formatter which is described in L<"HTML OPTIONS">.
59 This will produce a file F<somefile.pl.tdy> containing the script reformatted
60 using the default options, which approximate the style suggested in
61 perlstyle(1). The source file F<somefile.pl> is unchanged.
65 Execute perltidy on all F<.pl> files in the current directory with the
66 default options. The output will be in files with an appended F<.tdy>
67 extension. For any file with an error, there will be a file with extension
70 perltidy -b file1.pl file2.pl
72 Modify F<file1.pl> and F<file2.pl> in place, and backup the originals to
73 F<file1.pl.bak> and F<file2.pl.bak>. If F<file1.pl.bak> and/or F<file2.pl.bak>
74 already exist, they will be overwritten.
76 perltidy -b -bext='/' file1.pl file2.pl
78 Same as the previous example except that the backup files F<file1.pl.bak> and F<file2.pl.bak> will be deleted if there are no errors.
80 perltidy -gnu somefile.pl
82 Execute perltidy on file F<somefile.pl> with a style which approximates the
83 GNU Coding Standards for C programs. The output will be F<somefile.pl.tdy>.
85 perltidy -i=3 somefile.pl
87 Execute perltidy on file F<somefile.pl>, with 3 columns for each level of
88 indentation (B<-i=3>) instead of the default 4 columns. There will not be any
89 tabs in the reformatted script, except for any which already exist in comments,
90 pod documents, quotes, and here documents. Output will be F<somefile.pl.tdy>.
92 perltidy -i=3 -et=8 somefile.pl
94 Same as the previous example, except that leading whitespace will
95 be entabbed with one tab character per 8 spaces.
97 perltidy -ce -l=72 somefile.pl
99 Execute perltidy on file F<somefile.pl> with all defaults except use "cuddled
100 elses" (B<-ce>) and a maximum line length of 72 columns (B<-l=72>) instead of
101 the default 80 columns.
103 perltidy -g somefile.pl
105 Execute perltidy on file F<somefile.pl> and save a log file F<somefile.pl.LOG>
106 which shows the nesting of braces, parentheses, and square brackets at
107 the start of every line.
109 perltidy -html somefile.pl
111 This will produce a file F<somefile.pl.html> containing the script with
112 html markup. The output file will contain an embedded style sheet in
113 the <HEAD> section which may be edited to change the appearance.
115 perltidy -html -css=mystyle.css somefile.pl
117 This will produce a file F<somefile.pl.html> containing the script with
118 html markup. This output file will contain a link to a separate style
119 sheet file F<mystyle.css>. If the file F<mystyle.css> does not exist,
120 it will be created. If it exists, it will not be overwritten.
122 perltidy -html -pre somefile.pl
124 Write an html snippet with only the PRE section to F<somefile.pl.html>.
125 This is useful when code snippets are being formatted for inclusion in a
126 larger web page. No style sheet will be written in this case.
128 perltidy -html -ss >mystyle.css
130 Write a style sheet to F<mystyle.css> and exit.
132 perltidy -html -frm mymodule.pm
134 Write html with a frame holding a table of contents and the source code. The
135 output files will be F<mymodule.pm.html> (the frame), F<mymodule.pm.toc.html>
136 (the table of contents), and F<mymodule.pm.src.html> (the source code).
138 =head1 OPTIONS - OVERVIEW
140 The entire command line is scanned for options, and they are processed
141 before any files are processed. As a result, it does not matter
142 whether flags are before or after any filenames. However, the relative
143 order of parameters is important, with later parameters overriding the
144 values of earlier parameters.
146 For each parameter, there is a long name and a short name. The short
147 names are convenient for keyboard input, while the long names are
148 self-documenting and therefore useful in scripts. It is customary to
149 use two leading dashes for long names, but one may be used.
151 Most parameters which serve as on/off flags can be negated with a
152 leading "n" (for the short name) or a leading "no" or "no-" (for the
153 long name). For example, the flag to outdent long quotes is B<-olq>
154 or B<--outdent-long-quotes>. The flag to skip this is B<-nolq>
155 or B<--nooutdent-long-quotes> or B<--no-outdent-long-quotes>.
157 Options may not be bundled together. In other words, options B<-q> and
158 B<-g> may NOT be entered as B<-qg>.
160 Option names may be terminated early as long as they are uniquely identified.
161 For example, instead of B<--dump-token-types>, it would be sufficient to enter
162 B<--dump-tok>, or even B<--dump-t>, to uniquely identify this command.
166 The following parameters concern the files which are read and written.
170 =item B<-h>, B<--help>
172 Show summary of usage and exit.
174 =item B<-o>=filename, B<--outfile>=filename
176 Name of the output file (only if a single input file is being
177 processed). If no output file is specified, and output is not
178 redirected to the standard output, the output will go to F<filename.tdy>.
180 =item B<-st>, B<--standard-output>
182 Perltidy must be able to operate on an arbitrarily large number of files
183 in a single run, with each output being directed to a different output
184 file. Obviously this would conflict with outputting to the single
185 standard output device, so a special flag, B<-st>, is required to
186 request outputting to the standard output. For example,
188 perltidy somefile.pl -st >somefile.new.pl
190 This option may only be used if there is just a single input file.
191 The default is B<-nst> or B<--nostandard-output>.
193 =item B<-se>, B<--standard-error-output>
195 If perltidy detects an error when processing file F<somefile.pl>, its
196 default behavior is to write error messages to file F<somefile.pl.ERR>.
197 Use B<-se> to cause all error messages to be sent to the standard error
198 output stream instead. This directive may be negated with B<-nse>.
199 Thus, you may place B<-se> in a F<.perltidyrc> and override it when
200 desired with B<-nse> on the command line.
202 =item B<-oext>=ext, B<--output-file-extension>=ext
204 Change the extension of the output file to be F<ext> instead of the
205 default F<tdy> (or F<html> in case the -B<-html> option is used).
206 See L<Specifying File Extensions>.
208 =item B<-opath>=path, B<--output-path>=path
210 When perltidy creates a filename for an output file, by default it merely
211 appends an extension to the path and basename of the input file. This
212 parameter causes the path to be changed to F<path> instead.
214 The path should end in a valid path separator character, but perltidy will try
215 to add one if it is missing.
219 perltidy somefile.pl -opath=/tmp/
221 will produce F</tmp/somefile.pl.tdy>. Otherwise, F<somefile.pl.tdy> will
222 appear in whatever directory contains F<somefile.pl>.
224 If the path contains spaces, it should be placed in quotes.
226 This parameter will be ignored if output is being directed to standard output,
227 or if it is being specified explicitly with the B<-o=s> parameter.
229 =item B<-b>, B<--backup-and-modify-in-place>
231 Modify the input file or files in-place and save the original with the
232 extension F<.bak>. Any existing F<.bak> file will be deleted. See next
233 item for changing the default backup extension, and for eliminating the
234 backup file altogether.
236 A B<-b> flag will be ignored if input is from standard input or goes to
237 standard output, or if the B<-html> flag is set.
239 In particular, if you want to use both the B<-b> flag and the B<-pbp>
240 (--perl-best-practices) flag, then you must put a B<-nst> flag after the
241 B<-pbp> flag because it contains a B<-st> flag as one of its components,
242 which means that output will go to the standard output stream.
244 =item B<-bext>=ext, B<--backup-file-extension>=ext
246 This parameter serves two purposes: (1) to change the extension of the backup
247 file to be something other than the default F<.bak>, and (2) to indicate
248 that no backup file should be saved.
250 To change the default extension to something other than F<.bak> see
251 L<Specifying File Extensions>.
253 A backup file of the source is always written, but you can request that it
254 be deleted at the end of processing if there were no errors. This is risky
255 unless the source code is being maintained with a source code control
258 To indicate that the backup should be deleted include one forward slash,
259 B</>, in the extension. If any text remains after the slash is removed
260 it will be used to define the backup file extension (which is always
261 created and only deleted if there were no errors).
263 Here are some examples:
265 Parameter Extension Backup File Treatment
266 <-bext=bak> F<.bak> Keep (same as the default behavior)
267 <-bext='/'> F<.bak> Delete if no errors
268 <-bext='/backup'> F<.backup> Delete if no errors
269 <-bext='original/'> F<.original> Delete if no errors
271 =item B<-w>, B<--warning-output>
273 Setting B<-w> causes any non-critical warning
274 messages to be reported as errors. These include messages
275 about possible pod problems, possibly bad starting indentation level,
276 and cautions about indirect object usage. The default, B<-nw> or
277 B<--nowarning-output>, is not to include these warnings.
279 =item B<-q>, B<--quiet>
281 Deactivate error messages and syntax checking (for running under
284 For example, if you use a vi-style editor, such as vim, you may execute
285 perltidy as a filter from within the editor using something like
289 where C<n1,n2> represents the selected text. Without the B<-q> flag,
290 any error message may mess up your screen, so be prepared to use your
293 =item B<-log>, B<--logfile>
295 Save the F<.LOG> file, which has many useful diagnostics. Perltidy always
296 creates a F<.LOG> file, but by default it is deleted unless a program bug is
297 suspected. Setting the B<-log> flag forces the log file to be saved.
299 =item B<-g=n>, B<--logfile-gap=n>
301 Set maximum interval between input code lines in the logfile. This purpose of
302 this flag is to assist in debugging nesting errors. The value of C<n> is
303 optional. If you set the flag B<-g> without the value of C<n>, it will be
304 taken to be 1, meaning that every line will be written to the log file. This
305 can be helpful if you are looking for a brace, paren, or bracket nesting error.
307 Setting B<-g> also causes the logfile to be saved, so it is not necessary to
308 also include B<-log>.
310 If no B<-g> flag is given, a value of 50 will be used, meaning that at least
311 every 50th line will be recorded in the logfile. This helps prevent
312 excessively long log files.
314 Setting a negative value of C<n> is the same as not setting B<-g> at all.
316 =item B<-npro> B<--noprofile>
318 Ignore any F<.perltidyrc> command file. Normally, perltidy looks first in
319 your current directory for a F<.perltidyrc> file of parameters. (The format
320 is described below). If it finds one, it applies those options to the
321 initial default values, and then it applies any that have been defined
322 on the command line. If no F<.perltidyrc> file is found, it looks for one
323 in your home directory.
325 If you set the B<-npro> flag, perltidy will not look for this file.
327 =item B<-pro=filename> or B<--profile=filename>
329 To simplify testing and switching .perltidyrc files, this command may be
330 used to specify a configuration file which will override the default
331 name of .perltidyrc. There must not be a space on either side of the
332 '=' sign. For example, the line
334 perltidy -pro=testcfg
336 would cause file F<testcfg> to be used instead of the
337 default F<.perltidyrc>.
339 A pathname begins with three dots, e.g. ".../.perltidyrc", indicates that
340 the file should be searched for starting in the current directory and
341 working upwards. This makes it easier to have multiple projects each with
342 their own .perltidyrc in their root directories.
344 =item B<-opt>, B<--show-options>
346 Write a list of all options used to the F<.LOG> file.
347 Please see B<--dump-options> for a simpler way to do this.
349 =item B<-f>, B<--force-read-binary>
351 Force perltidy to process binary files. To avoid producing excessive
352 error messages, perltidy skips files identified by the system as non-text.
353 However, valid perl scripts containing binary data may sometimes be identified
354 as non-text, and this flag forces perltidy to process them.
358 =head1 FORMATTING OPTIONS
366 This flag disables all formatting and causes the input to be copied unchanged
367 to the output except for possible changes in line ending characters and any
368 pre- and post-filters. This can be useful in conjunction with a hierarchical
369 set of F<.perltidyrc> files to avoid unwanted code tidying. See also
370 L<Skipping Selected Sections of Code> for a way to avoid tidying specific
373 =item B<-i=n>, B<--indent-columns=n>
375 Use n columns per indentation level (default n=4).
377 =item B<-l=n>, B<--maximum-line-length=n>
379 The default maximum line length is n=80 characters. Perltidy will try
380 to find line break points to keep lines below this length. However, long
381 quotes and side comments may cause lines to exceed this length.
382 Setting B<-l=0> is equivalent to setting B<-l=(a large number)>.
384 =item B<-vmll>, B<--variable-maximum-line-length>
386 A problem arises using a fixed maximum line length with very deeply nested code
387 and data structures because eventually the amount of leading whitespace used
388 for indicating indentation takes up most or all of the available line width,
389 leaving little or no space for the actual code or data. One solution is to use
390 a vary long line length. Another solution is to use the B<-vmll> flag, which
391 basically tells perltidy to ignore leading whitespace when measuring the line
394 To be precise, when the B<-vmll> parameter is set, the maximum line length of a
395 line of code will be M+L*I, where
397 M is the value of --maximum-line-length=M (-l=M), default 80,
398 I is the value of --indent-columns=I (-i=I), default 4,
399 L is the indentation level of the line of code
401 When this flag is set, the choice of breakpoints for a block of code should be
402 essentially independent of its nesting depth. However, the absolute line
403 lengths, including leading whitespace, can still be arbitrarily large. This
404 problem can be avoided by including the next parameter.
406 The default is not to do this (B<-nvmll>).
408 =item B<-wc=n>, B<--whitespace-cycle=n>
410 This flag also addresses problems with very deeply nested code and data
411 structures. When the nesting depth exceeds the value B<n> the leading
412 whitespace will be reduced and start at a depth of 1 again. The result is that
413 blocks of code will shift back to the left rather than moving arbitrarily far
414 to the right. This occurs cyclically to any depth.
416 For example if one level of indentation equals 4 spaces (B<-i=4>, the default),
417 and one uses B<-wc=15>, then if the leading whitespace on a line exceeds about
418 4*15=60 spaces it will be reduced back to 4*1=4 spaces and continue increasing
419 from there. If the whitespace never exceeds this limit the formatting remains
422 The combination of B<-vmll> and B<-wc=n> provides a solution to the problem of
423 displaying arbitrarily deep data structures and code in a finite window,
424 although B<-wc=n> may of course be used without B<-vmll>.
426 The default is not to use this, which can also be indicated using B<-wc=0>.
430 Using tab characters will almost certainly lead to future portability
431 and maintenance problems, so the default and recommendation is not to
432 use them. For those who prefer tabs, however, there are two different
435 Except for possibly introducing tab indentation characters, as outlined
436 below, perltidy does not introduce any tab characters into your file,
437 and it removes any tabs from the code (unless requested not to do so
438 with B<-fws>). If you have any tabs in your comments, quotes, or
439 here-documents, they will remain.
443 =item B<-et=n>, B<--entab-leading-whitespace>
445 This flag causes each B<n> initial space characters to be replaced by
446 one tab character. Note that the integer B<n> is completely independent
447 of the integer specified for indentation parameter, B<-i=n>.
449 =item B<-t>, B<--tabs>
451 This flag causes one leading tab character to be inserted for each level
452 of indentation. Certain other features are incompatible with this
453 option, and if these options are also given, then a warning message will
454 be issued and this flag will be unset. One example is the B<-lp>
457 =item B<-dt=n>, B<--default-tabsize=n>
459 If the first line of code passed to perltidy contains leading tabs but no
460 tab scheme is specified for the output stream then perltidy must guess how many
461 spaces correspond to each leading tab. This number of spaces B<n>
462 corresponding to each leading tab of the input stream may be specified with
463 B<-dt=n>. The default is B<n=8>.
465 This flag has no effect if a tab scheme is specified for the output stream,
466 because then the input stream is assumed to use the same tab scheme and
467 indentation spaces as for the output stream (any other assumption would lead to
472 =item B<-syn>, B<--check-syntax>
474 This flag causes perltidy to run C<perl -c -T> to check syntax of input
475 and output. (To change the flags passed to perl, see the next
476 item, B<-pscf>). The results are written to the F<.LOG> file, which
477 will be saved if an error is detected in the output script. The output
478 script is not checked if the input script has a syntax error. Perltidy
479 does its own checking, but this option employs perl to get a "second
482 If perl reports errors in the input file, they will not be reported in
483 the error output unless the B<--warning-output> flag is given.
485 The default is B<NOT> to do this type of syntax checking (although
486 perltidy will still do as much self-checking as possible). The reason
487 is that it causes all code in BEGIN blocks to be executed, for all
488 modules being used, and this opens the door to security issues and
489 infinite loops when running perltidy.
491 =item B<-pscf=s>, B<-perl-syntax-check-flags=s>
493 When perl is invoked to check syntax, the normal flags are C<-c -T>. In
494 addition, if the B<-x> flag is given to perltidy, then perl will also be
495 passed a B<-x> flag. It should not normally be necessary to change
496 these flags, but it can be done with the B<-pscf=s> flag. For example,
497 if the taint flag, C<-T>, is not wanted, the flag could be set to be just
500 Perltidy will pass your string to perl with the exception that it will
501 add a B<-c> and B<-x> if appropriate. The F<.LOG> file will show
502 exactly what flags were passed to perl.
504 =item B<-io>, B<--indent-only>
506 This flag is used to deactivate all formatting and line break changes
507 within non-blank lines of code.
508 When it is in effect, the only change to the script will be
509 to the indentation and blank lines.
510 And any flags controlling whitespace and newlines will be ignored. You
511 might want to use this if you are perfectly happy with your whitespace
512 and line breaks, and merely want perltidy to handle the indentation.
513 (This also speeds up perltidy by well over a factor of two, so it might be
514 useful when perltidy is merely being used to help find a brace error in
517 Setting this flag is equivalent to setting B<--freeze-newlines> and
518 B<--freeze-whitespace>.
520 If you also want to keep your existing blank lines exactly
521 as they are, you can add B<--freeze-blank-lines>.
523 =item B<-ole=s>, B<--output-line-ending=s>
525 where s=C<win>, C<dos>, C<unix>, or C<mac>. This flag tells perltidy
526 to output line endings for a specific system. Normally,
527 perltidy writes files with the line separator character of the host
528 system. The C<win> and C<dos> flags have an identical result.
530 =item B<-ple>, B<--preserve-line-endings>
532 This flag tells perltidy to write its output files with the same line
533 endings as the input file, if possible. It should work for
534 B<dos>, B<unix>, and B<mac> line endings. It will only work if perltidy
535 input comes from a filename (rather than stdin, for example). If
536 perltidy has trouble determining the input file line ending, it will
537 revert to the default behavior of using the line ending of the host system.
539 =item B<-it=n>, B<--iterations=n>
541 This flag causes perltidy to do B<n> complete iterations. The reason for this
542 flag is that code beautification is an iterative process and in some
543 cases the output from perltidy can be different if it is applied a second time.
544 For most purposes the default of B<n=1> should be satisfactory. However B<n=2>
545 can be useful when a major style change is being made, or when code is being
546 beautified on check-in to a source code control system. It has been found to
547 be extremely rare for the output to change after 2 iterations. If a value
548 B<n> is greater than 2 is input then a convergence test will be used to stop
549 the iterations as soon as possible, almost always after 2 iterations. See
550 the next item for a simplified iteration control.
552 This flag has no effect when perltidy is used to generate html.
554 =item B<-conv>, B<--converge>
556 This flag is equivalent to B<-it=4> and is included to simplify iteration
557 control. For all practical purposes one either does or does not want to be
558 sure that the output is converged, and there is no penalty to using a large
559 iteration limit since perltidy will check for convergence and stop iterating as
560 soon as possible. The default is B<-nconv> (no convergence check). Using
561 B<-conv> will approximately double run time since normally one extra iteration
562 is required to verify convergence.
566 =head2 Code Indentation Control
570 =item B<-ci=n>, B<--continuation-indentation=n>
572 Continuation indentation is extra indentation spaces applied when
573 a long line is broken. The default is n=2, illustrated here:
576 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
578 The same example, with n=0, is a little harder to read:
581 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
583 The value given to B<-ci> is also used by some commands when a small
584 space is required. Examples are commands for outdenting labels,
585 B<-ola>, and control keywords, B<-okw>.
587 When default values are not used, it is suggested that the value B<n>
588 given with B<-ci=n> be no more than about one-half of the number of
589 spaces assigned to a full indentation level on the B<-i=n> command.
591 =item B<-sil=n> B<--starting-indentation-level=n>
593 By default, perltidy examines the input file and tries to determine the
594 starting indentation level. While it is often zero, it may not be
595 zero for a code snippet being sent from an editing session.
597 To guess the starting indentation level perltidy simply assumes that
598 indentation scheme used to create the code snippet is the same as is being used
599 for the current perltidy process. This is the only sensible guess that can be
600 made. It should be correct if this is true, but otherwise it probably won't.
601 For example, if the input script was written with -i=2 and the current peltidy
602 flags have -i=4, the wrong initial indentation will be guessed for a code
603 snippet which has non-zero initial indentation. Likewise, if an entabbing
604 scheme is used in the input script and not in the current process then the
605 guessed indentation will be wrong.
607 If the default method does not work correctly, or you want to change the
608 starting level, use B<-sil=n>, to force the starting level to be n.
610 =item List indentation using B<-lp>, B<--line-up-parentheses>
612 By default, perltidy indents lists with 4 spaces, or whatever value
613 is specified with B<-i=n>. Here is a small list formatted in this way:
617 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
618 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
621 Use the B<-lp> flag to add extra indentation to cause the data to begin
622 past the opening parentheses of a sub call or list, or opening square
623 bracket of an anonymous array, or opening curly brace of an anonymous
624 hash. With this option, the above list would become:
628 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
629 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
632 If the available line length (see B<-l=n> ) does not permit this much
633 space, perltidy will use less. For alternate placement of the
634 closing paren, see the next section.
636 This option has no effect on code BLOCKS, such as if/then/else blocks,
637 which always use whatever is specified with B<-i=n>. Also, the
638 existence of line breaks and/or block comments between the opening and
639 closing parens may cause perltidy to temporarily revert to its default
642 Note: The B<-lp> option may not be used together with the B<-t> tabs option.
643 It may, however, be used with the B<-et=n> tab method.
645 In addition, any parameter which significantly restricts the ability of
646 perltidy to choose newlines will conflict with B<-lp> and will cause
647 B<-lp> to be deactivated. These include B<-io>, B<-fnl>, B<-nanl>, and
648 B<-ndnl>. The reason is that the B<-lp> indentation style can require
649 the careful coordination of an arbitrary number of break points in
650 hierarchical lists, and these flags may prevent that.
652 =item B<-cti=n>, B<--closing-token-indentation>
654 The B<-cti=n> flag controls the indentation of a line beginning with
655 a C<)>, C<]>, or a non-block C<}>. Such a line receives:
657 -cti = 0 no extra indentation (default)
658 -cti = 1 extra indentation such that the closing token
659 aligns with its opening token.
660 -cti = 2 one extra indentation level if the line looks like:
662 -cti = 3 one extra indentation level always
664 The flags B<-cti=1> and B<-cti=2> work well with the B<-lp> flag (previous
667 # perltidy -lp -cti=1
669 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
670 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
673 # perltidy -lp -cti=2
675 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
676 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
679 These flags are merely hints to the formatter and they may not always be
680 followed. In particular, if -lp is not being used, the indentation for
681 B<cti=1> is constrained to be no more than one indentation level.
683 If desired, this control can be applied independently to each of the
684 closing container token types. In fact, B<-cti=n> is merely an
685 abbreviation for B<-cpi=n -csbi=n -cbi=n>, where:
686 B<-cpi> or B<--closing-paren-indentation> controls B<)>'s,
687 B<-csbi> or B<--closing-square-bracket-indentation> controls B<]>'s,
688 B<-cbi> or B<--closing-brace-indentation> controls non-block B<}>'s.
690 =item B<-icp>, B<--indent-closing-paren>
692 The B<-icp> flag is equivalent to
693 B<-cti=2>, described in the previous section. The B<-nicp> flag is
694 equivalent B<-cti=0>. They are included for backwards compatibility.
696 =item B<-icb>, B<--indent-closing-brace>
698 The B<-icb> option gives one extra level of indentation to a brace which
699 terminates a code block . For example,
708 The default is not to do this, indicated by B<-nicb>.
710 =item B<-olq>, B<--outdent-long-quotes>
712 When B<-olq> is set, lines which is a quoted string longer than the
713 value B<maximum-line-length> will have their indentation removed to make
714 them more readable. This is the default. To prevent such out-denting,
715 use B<-nolq> or B<--nooutdent-long-lines>.
717 =item B<-oll>, B<--outdent-long-lines>
719 This command is equivalent to B<--outdent-long-quotes> and
720 B<--outdent-long-comments>, and it is included for compatibility with previous
721 versions of perltidy. The negation of this also works, B<-noll> or
722 B<--nooutdent-long-lines>, and is equivalent to setting B<-nolq> and B<-nolc>.
724 =item Outdenting Labels: B<-ola>, B<--outdent-labels>
726 This command will cause labels to be outdented by 2 spaces (or whatever B<-ci>
727 has been set to), if possible. This is the default. For example:
730 LOOP: while ( $i = <FOTOS> ) {
736 Use B<-nola> to not outdent labels.
738 =item Outdenting Keywords
742 =item B<-okw>, B<--outdent-keywords>
744 The command B<-okw> will cause certain leading control keywords to
745 be outdented by 2 spaces (or whatever B<-ci> has been set to), if
746 possible. By default, these keywords are C<redo>, C<next>, C<last>,
747 C<goto>, and C<return>. The intention is to make these control keywords
748 easier to see. To change this list of keywords being outdented, see
751 For example, using C<perltidy -okw> on the previous example gives:
754 LOOP: while ( $i = <FOTOS> ) {
760 The default is not to do this.
762 =item Specifying Outdented Keywords: B<-okwl=string>, B<--outdent-keyword-list=string>
764 This command can be used to change the keywords which are outdented with
765 the B<-okw> command. The parameter B<string> is a required list of perl
766 keywords, which should be placed in quotes if there are more than one.
767 By itself, it does not cause any outdenting to occur, so the B<-okw>
768 command is still required.
770 For example, the commands C<-okwl="next last redo goto" -okw> will cause
771 those four keywords to be outdented. It is probably simplest to place
772 any B<-okwl> command in a F<.perltidyrc> file.
778 =head2 Whitespace Control
780 Whitespace refers to the blank space between variables, operators,
781 and other code tokens.
785 =item B<-fws>, B<--freeze-whitespace>
787 This flag causes your original whitespace to remain unchanged, and
788 causes the rest of the whitespace commands in this section, the
789 Code Indentation section, and
790 the Comment Control section to be ignored.
792 =item Tightness of curly braces, parentheses, and square brackets.
794 Here the term "tightness" will mean the closeness with which
795 pairs of enclosing tokens, such as parentheses, contain the quantities
796 within. A numerical value of 0, 1, or 2 defines the tightness, with
797 0 being least tight and 2 being most tight. Spaces within containers
798 are always symmetric, so if there is a space after a C<(> then there
799 will be a space before the corresponding C<)>.
801 The B<-pt=n> or B<--paren-tightness=n> parameter controls the space within
802 parens. The example below shows the effect of the three possible
805 if ( ( my $len_tab = length( $tabstr ) ) > 0 ) { # -pt=0
806 if ( ( my $len_tab = length($tabstr) ) > 0 ) { # -pt=1 (default)
807 if ((my $len_tab = length($tabstr)) > 0) { # -pt=2
809 When n is 0, there is always a space to the right of a '(' and to the left
810 of a ')'. For n=2 there is never a space. For n=1, the default, there
811 is a space unless the quantity within the parens is a single token, such
812 as an identifier or quoted string.
814 Likewise, the parameter B<-sbt=n> or B<--square-bracket-tightness=n>
815 controls the space within square brackets, as illustrated below.
817 $width = $col[ $j + $k ] - $col[ $j ]; # -sbt=0
818 $width = $col[ $j + $k ] - $col[$j]; # -sbt=1 (default)
819 $width = $col[$j + $k] - $col[$j]; # -sbt=2
821 Curly braces which do not contain code blocks are controlled by
822 the parameter B<-bt=n> or B<--brace-tightness=n>.
824 $obj->{ $parsed_sql->{ 'table' }[0] }; # -bt=0
825 $obj->{ $parsed_sql->{'table'}[0] }; # -bt=1 (default)
826 $obj->{$parsed_sql->{'table'}[0]}; # -bt=2
828 And finally, curly braces which contain blocks of code are controlled by the
829 parameter B<-bbt=n> or B<--block-brace-tightness=n> as illustrated in the
832 %bf = map { $_ => -M $_ } grep { /\.deb$/ } dirents '.'; # -bbt=0 (default)
833 %bf = map { $_ => -M $_ } grep {/\.deb$/} dirents '.'; # -bbt=1
834 %bf = map {$_ => -M $_} grep {/\.deb$/} dirents '.'; # -bbt=2
836 To simplify input in the case that all of the tightness flags have the same
837 value <n>, the parameter <-act=n> or B<--all-containers-tightness=n> is an
838 abbreviation for the combination <-pt=n -sbt=n -bt=n -bbt=n>.
841 =item B<-tso>, B<--tight-secret-operators>
843 The flag B<-tso> causes certain perl token sequences (secret operators)
844 which might be considered to be a single operator to be formatted "tightly"
845 (without spaces). The operators currently modified by this flag are:
847 0+ +0 ()x!! ~~<> ,=> =( )=
849 For example the sequence B<0 +>, which converts a string to a number,
850 would be formatted without a space: B<0+> when the B<-tso> flag is set. This
851 flag is off by default.
853 =item B<-sts>, B<--space-terminal-semicolon>
855 Some programmers prefer a space before all terminal semicolons. The
856 default is for no such space, and is indicated with B<-nsts> or
857 B<--nospace-terminal-semicolon>.
860 $i = 1; # -nsts (default)
862 =item B<-sfs>, B<--space-for-semicolon>
864 Semicolons within B<for> loops may sometimes be hard to see,
865 particularly when commas are also present. This option places spaces on
866 both sides of these special semicolons, and is the default. Use
867 B<-nsfs> or B<--nospace-for-semicolon> to deactivate it.
869 for ( @a = @$ap, $u = shift @a ; @a ; $u = $v ) { # -sfs (default)
870 for ( @a = @$ap, $u = shift @a; @a; $u = $v ) { # -nsfs
872 =item B<-asc>, B<--add-semicolons>
874 Setting B<-asc> allows perltidy to add any missing optional semicolon at the end
875 of a line which is followed by a closing curly brace on the next line. This
876 is the default, and may be deactivated with B<-nasc> or B<--noadd-semicolons>.
878 =item B<-dsm>, B<--delete-semicolons>
880 Setting B<-dsm> allows perltidy to delete extra semicolons which are
881 simply empty statements. This is the default, and may be deactivated
882 with B<-ndsm> or B<--nodelete-semicolons>. (Such semicolons are not
883 deleted, however, if they would promote a side comment to a block
886 =item B<-aws>, B<--add-whitespace>
888 Setting this option allows perltidy to add certain whitespace improve
889 code readability. This is the default. If you do not want any
890 whitespace added, but are willing to have some whitespace deleted, use
891 B<-naws>. (Use B<-fws> to leave whitespace completely unchanged).
893 =item B<-dws>, B<--delete-old-whitespace>
895 Setting this option allows perltidy to remove some old whitespace
896 between characters, if necessary. This is the default. If you
897 do not want any old whitespace removed, use B<-ndws> or
898 B<--nodelete-old-whitespace>.
900 =item Detailed whitespace controls around tokens
902 For those who want more detailed control over the whitespace around
903 tokens, there are four parameters which can directly modify the default
904 whitespace rules built into perltidy for any token. They are:
906 B<-wls=s> or B<--want-left-space=s>,
908 B<-nwls=s> or B<--nowant-left-space=s>,
910 B<-wrs=s> or B<--want-right-space=s>,
912 B<-nwrs=s> or B<--nowant-right-space=s>.
914 These parameters are each followed by a quoted string, B<s>, containing a
915 list of token types. No more than one of each of these parameters
916 should be specified, because repeating a command-line parameter
917 always overwrites the previous one before perltidy ever sees it.
919 To illustrate how these are used, suppose it is desired that there be no
920 space on either side of the token types B<= + - / *>. The following two
921 parameters would specify this desire:
923 -nwls="= + - / *" -nwrs="= + - / *"
925 (Note that the token types are in quotes, and that they are separated by
926 spaces). With these modified whitespace rules, the following line of math:
928 $root = -$b + sqrt( $b * $b - 4. * $a * $c ) / ( 2. * $a );
932 $root=-$b+sqrt( $b*$b-4.*$a*$c )/( 2.*$a );
934 These parameters should be considered to be hints to perltidy rather
935 than fixed rules, because perltidy must try to resolve conflicts that
936 arise between them and all of the other rules that it uses. One
937 conflict that can arise is if, between two tokens, the left token wants
938 a space and the right one doesn't. In this case, the token not wanting
939 a space takes priority.
941 It is necessary to have a list of all token types in order to create
942 this type of input. Such a list can be obtained by the command
943 B<--dump-token-types>. Also try the B<-D> flag on a short snippet of code
944 and look at the .DEBUG file to see the tokenization.
946 B<WARNING> Be sure to put these tokens in quotes to avoid having them
947 misinterpreted by your command shell.
949 =item Space between specific keywords and opening paren
951 When an opening paren follows a Perl keyword, no space is introduced after the
952 keyword, unless it is (by default) one of these:
954 my local our and or eq ne if else elsif until unless
955 while for foreach return switch case given when
957 These defaults can be modified with two commands:
959 B<-sak=s> or B<--space-after-keyword=s> adds keywords.
961 B<-nsak=s> or B<--nospace-after-keyword=s> removes keywords.
963 where B<s> is a list of keywords (in quotes if necessary). For example,
965 my ( $a, $b, $c ) = @_; # default
966 my( $a, $b, $c ) = @_; # -nsak="my local our"
968 The abbreviation B<-nsak='*'> is equivalent to including all of the
969 keywords in the above list.
971 When both B<-nsak=s> and B<-sak=s> commands are included, the B<-nsak=s>
972 command is executed first. For example, to have space after only the
973 keywords (my, local, our) you could use B<-nsak="*" -sak="my local our">.
975 To put a space after all keywords, see the next item.
977 =item Space between all keywords and opening parens
979 When an opening paren follows a function or keyword, no space is introduced
980 after the keyword except for the keywords noted in the previous item. To
981 always put a space between a function or keyword and its opening paren,
984 B<-skp> or B<--space-keyword-paren>
986 You will probably also want to use the flag B<-sfp> (next item) too.
988 =item Space between all function names and opening parens
990 When an opening paren follows a function the default is not to introduce
991 a space. To cause a space to be introduced use:
993 B<-sfp> or B<--space-function-paren>
995 myfunc( $a, $b, $c ); # default
996 myfunc ( $a, $b, $c ); # -sfp
998 You will probably also want to use the flag B<-skp> (previous item) too.
1000 =item Trimming whitespace around C<qw> quotes
1002 B<-tqw> or B<--trim-qw> provide the default behavior of trimming
1003 spaces around multi-line C<qw> quotes and indenting them appropriately.
1005 B<-ntqw> or B<--notrim-qw> cause leading and trailing whitespace around
1006 multi-line C<qw> quotes to be left unchanged. This option will not
1007 normally be necessary, but was added for testing purposes, because in
1008 some versions of perl, trimming C<qw> quotes changes the syntax tree.
1010 =item Trimming trailing whitespace from lines of POD
1012 B<-trp> or B<--trim-pod> will remove trailing whitespace from lines of POD.
1013 The default is not to do this.
1017 =head2 Comment Controls
1019 Perltidy has a number of ways to control the appearance of both block comments
1020 and side comments. The term B<block comment> here refers to a full-line
1021 comment, whereas B<side comment> will refer to a comment which appears on a
1022 line to the right of some code.
1026 =item B<-ibc>, B<--indent-block-comments>
1028 Block comments normally look best when they are indented to the same
1029 level as the code which follows them. This is the default behavior, but
1030 you may use B<-nibc> to keep block comments left-justified. Here is an
1033 # this comment is indented (-ibc, default)
1034 if ($task) { yyy(); }
1036 The alternative is B<-nibc>:
1038 # this comment is not indented (-nibc)
1039 if ($task) { yyy(); }
1041 See also the next item, B<-isbc>, as well as B<-sbc>, for other ways to
1042 have some indented and some outdented block comments.
1044 =item B<-isbc>, B<--indent-spaced-block-comments>
1046 If there is no leading space on the line, then the comment will not be
1047 indented, and otherwise it may be.
1049 If both B<-ibc> and B<-isbc> are set, then B<-isbc> takes priority.
1051 =item B<-olc>, B<--outdent-long-comments>
1053 When B<-olc> is set, lines which are full-line (block) comments longer
1054 than the value B<maximum-line-length> will have their indentation
1055 removed. This is the default; use B<-nolc> to prevent outdenting.
1057 =item B<-msc=n>, B<--minimum-space-to-comment=n>
1059 Side comments look best when lined up several spaces to the right of
1060 code. Perltidy will try to keep comments at least n spaces to the
1061 right. The default is n=4 spaces.
1063 =item B<-fpsc=n>, B<--fixed-position-side-comment=n>
1065 This parameter tells perltidy to line up side comments in column number B<n>
1066 whenever possible. The default, n=0, will not do this.
1068 =item B<-iscl>, B<--ignore-side-comment-lengths>
1070 This parameter causes perltidy to ignore the length of side comments when
1071 setting line breaks. The default, B<-niscl>, is to include the length of
1072 side comments when breaking lines to stay within the length prescribed
1073 by the B<-l=n> maximum line length parameter. For example, the following
1074 long single line would remain intact with -l=80 and -iscl:
1076 perltidy -l=80 -iscl
1077 $vmsfile =~ s/;[\d\-]*$//; # Clip off version number; we can use a newer version as well
1079 whereas without the -iscl flag the line will be broken:
1082 $vmsfile =~ s/;[\d\-]*$//
1083 ; # Clip off version number; we can use a newer version as well
1086 =item B<-hsc>, B<--hanging-side-comments>
1088 By default, perltidy tries to identify and align "hanging side
1089 comments", which are something like this:
1091 my $IGNORE = 0; # This is a side comment
1092 # This is a hanging side comment
1095 A comment is considered to be a hanging side comment if (1) it immediately
1096 follows a line with a side comment, or another hanging side comment, and
1097 (2) there is some leading whitespace on the line.
1098 To deactivate this feature, use B<-nhsc> or B<--nohanging-side-comments>.
1099 If block comments are preceded by a blank line, or have no leading
1100 whitespace, they will not be mistaken as hanging side comments.
1102 =item Closing Side Comments
1104 A closing side comment is a special comment which perltidy can
1105 automatically create and place after the closing brace of a code block.
1106 They can be useful for code maintenance and debugging. The command
1107 B<-csc> (or B<--closing-side-comments>) adds or updates closing side
1108 comments. For example, here is a small code snippet
1111 if ( !defined( $_[0] ) ) {
1112 print("Hello, World\n");
1115 print( $_[0], "\n" );
1119 And here is the result of processing with C<perltidy -csc>:
1122 if ( !defined( $_[0] ) ) {
1123 print("Hello, World\n");
1126 print( $_[0], "\n" );
1128 } ## end sub message
1130 A closing side comment was added for C<sub message> in this case, but not
1131 for the C<if> and C<else> blocks, because they were below the 6 line
1132 cutoff limit for adding closing side comments. This limit may be
1133 changed with the B<-csci> command, described below.
1135 The command B<-dcsc> (or B<--delete-closing-side-comments>) reverses this
1136 process and removes these comments.
1138 Several commands are available to modify the behavior of these two basic
1139 commands, B<-csc> and B<-dcsc>:
1143 =item B<-csci=n>, or B<--closing-side-comment-interval=n>
1145 where C<n> is the minimum number of lines that a block must have in
1146 order for a closing side comment to be added. The default value is
1147 C<n=6>. To illustrate:
1149 # perltidy -csci=2 -csc
1151 if ( !defined( $_[0] ) ) {
1152 print("Hello, World\n");
1153 } ## end if ( !defined( $_[0] ))
1155 print( $_[0], "\n" );
1156 } ## end else [ if ( !defined( $_[0] ))
1157 } ## end sub message
1159 Now the C<if> and C<else> blocks are commented. However, now this has
1160 become very cluttered.
1162 =item B<-cscp=string>, or B<--closing-side-comment-prefix=string>
1164 where string is the prefix used before the name of the block type. The
1165 default prefix, shown above, is C<## end>. This string will be added to
1166 closing side comments, and it will also be used to recognize them in
1167 order to update, delete, and format them. Any comment identified as a
1168 closing side comment will be placed just a single space to the right of
1171 =item B<-cscl=string>, or B<--closing-side-comment-list-string>
1173 where C<string> is a list of block types to be tagged with closing side
1174 comments. By default, all code block types preceded by a keyword or
1175 label (such as C<if>, C<sub>, and so on) will be tagged. The B<-cscl>
1176 command changes the default list to be any selected block types; see
1177 L<Specifying Block Types>.
1178 For example, the following command
1179 requests that only C<sub>'s, labels, C<BEGIN>, and C<END> blocks be
1180 affected by any B<-csc> or B<-dcsc> operation:
1182 -cscl="sub : BEGIN END"
1184 =item B<-csct=n>, or B<--closing-side-comment-maximum-text=n>
1186 The text appended to certain block types, such as an C<if> block, is
1187 whatever lies between the keyword introducing the block, such as C<if>,
1188 and the opening brace. Since this might be too much text for a side
1189 comment, there needs to be a limit, and that is the purpose of this
1190 parameter. The default value is C<n=20>, meaning that no additional
1191 tokens will be appended to this text after its length reaches 20
1192 characters. Omitted text is indicated with C<...>. (Tokens, including
1193 sub names, are never truncated, however, so actual lengths may exceed
1194 this). To illustrate, in the above example, the appended text of the
1195 first block is C< ( !defined( $_[0] )...>. The existing limit of
1196 C<n=20> caused this text to be truncated, as indicated by the C<...>. See
1197 the next flag for additional control of the abbreviated text.
1199 =item B<-cscb>, or B<--closing-side-comments-balanced>
1201 As discussed in the previous item, when the
1202 closing-side-comment-maximum-text limit is exceeded the comment text must
1203 be truncated. Older versions of perltidy terminated with three dots, and this
1204 can still be achieved with -ncscb:
1206 perltidy -csc -ncscb
1207 } ## end foreach my $foo (sort { $b cmp $a ...
1209 However this causes a problem with editors which cannot recognize
1210 comments or are not configured to do so because they cannot "bounce" around in
1211 the text correctly. The B<-cscb> flag has been added to
1212 help them by appending appropriate balancing structure:
1215 } ## end foreach my $foo (sort { $b cmp $a ... })
1217 The default is B<-cscb>.
1219 =item B<-csce=n>, or B<--closing-side-comment-else-flag=n>
1221 The default, B<n=0>, places the text of the opening C<if> statement after any
1224 If B<n=2> is used, then each C<elsif> is also given the text of the opening
1225 C<if> statement. Also, an C<else> will include the text of a preceding
1226 C<elsif> statement. Note that this may result some long closing
1229 If B<n=1> is used, the results will be the same as B<n=2> whenever the
1230 resulting line length is less than the maximum allowed.
1231 =item B<-cscb>, or B<--closing-side-comments-balanced>
1233 When using closing-side-comments, and the closing-side-comment-maximum-text
1234 limit is exceeded, then the comment text must be abbreviated.
1235 It is terminated with three dots if the B<-cscb> flag is negated:
1237 perltidy -csc -ncscb
1238 } ## end foreach my $foo (sort { $b cmp $a ...
1240 This causes a problem with older editors which do not recognize comments
1241 because they cannot "bounce" around in the text correctly. The B<-cscb>
1242 flag tries to help them by appending appropriate terminal balancing structures:
1245 } ## end foreach my $foo (sort { $b cmp $a ... })
1247 The default is B<-cscb>.
1250 =item B<-cscw>, or B<--closing-side-comment-warnings>
1252 This parameter is intended to help make the initial transition to the use of
1253 closing side comments.
1255 things to happen if a closing side comment replaces an existing, different
1256 closing side comment: first, an error message will be issued, and second, the
1257 original side comment will be placed alone on a new specially marked comment
1258 line for later attention.
1260 The intent is to avoid clobbering existing hand-written side comments
1261 which happen to match the pattern of closing side comments. This flag
1262 should only be needed on the first run with B<-csc>.
1266 B<Important Notes on Closing Side Comments:>
1272 Closing side comments are only placed on lines terminated with a closing
1273 brace. Certain closing styles, such as the use of cuddled elses
1274 (B<-ce>), preclude the generation of some closing side comments.
1278 Please note that adding or deleting of closing side comments takes
1279 place only through the commands B<-csc> or B<-dcsc>. The other commands,
1280 if used, merely modify the behavior of these two commands.
1284 It is recommended that the B<-cscw> flag be used along with B<-csc> on
1285 the first use of perltidy on a given file. This will prevent loss of
1286 any existing side comment data which happens to have the csc prefix.
1290 Once you use B<-csc>, you should continue to use it so that any
1291 closing side comments remain correct as code changes. Otherwise, these
1292 comments will become incorrect as the code is updated.
1296 If you edit the closing side comments generated by perltidy, you must also
1297 change the prefix to be different from the closing side comment prefix.
1298 Otherwise, your edits will be lost when you rerun perltidy with B<-csc>. For
1299 example, you could simply change C<## end> to be C<## End>, since the test is
1300 case sensitive. You may also want to use the B<-ssc> flag to keep these
1301 modified closing side comments spaced the same as actual closing side comments.
1305 Temporarily generating closing side comments is a useful technique for
1306 exploring and/or debugging a perl script, especially one written by someone
1307 else. You can always remove them with B<-dcsc>.
1311 =item Static Block Comments
1313 Static block comments are block comments with a special leading pattern,
1314 C<##> by default, which will be treated slightly differently from other
1315 block comments. They effectively behave as if they had glue along their
1316 left and top edges, because they stick to the left edge and previous line
1317 when there is no blank spaces in those places. This option is
1318 particularly useful for controlling how commented code is displayed.
1322 =item B<-sbc>, B<--static-block-comments>
1324 When B<-sbc> is used, a block comment with a special leading pattern, C<##> by
1325 default, will be treated specially.
1327 Comments so identified are treated as follows:
1333 If there is no leading space on the line, then the comment will not
1334 be indented, and otherwise it may be,
1338 no new blank line will be
1339 inserted before such a comment, and
1343 such a comment will never become
1344 a hanging side comment.
1348 For example, assuming C<@month_of_year> is
1351 @month_of_year = ( # -sbc (default)
1352 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1356 Without this convention, the above code would become
1358 @month_of_year = ( # -nsbc
1359 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1365 which is not as clear.
1366 The default is to use B<-sbc>. This may be deactivated with B<-nsbc>.
1368 =item B<-sbcp=string>, B<--static-block-comment-prefix=string>
1370 This parameter defines the prefix used to identify static block comments
1371 when the B<-sbc> parameter is set. The default prefix is C<##>,
1372 corresponding to C<-sbcp=##>. The prefix is actually part of a perl
1373 pattern used to match lines and it must either begin with C<#> or C<^#>.
1374 In the first case a prefix ^\s* will be added to match any leading
1375 whitespace, while in the second case the pattern will match only
1376 comments with no leading whitespace. For example, to
1377 identify all comments as static block comments, one would use C<-sbcp=#>.
1378 To identify all left-adjusted comments as static block comments, use C<-sbcp='^#'>.
1380 Please note that B<-sbcp> merely defines the pattern used to identify static
1381 block comments; it will not be used unless the switch B<-sbc> is set. Also,
1382 please be aware that since this string is used in a perl regular expression
1383 which identifies these comments, it must enable a valid regular expression to
1386 A pattern which can be useful is:
1390 This pattern requires a static block comment to have at least one character
1391 which is neither a # nor a space. It allows a line containing only '#'
1392 characters to be rejected as a static block comment. Such lines are often used
1393 at the start and end of header information in subroutines and should not be
1394 separated from the intervening comments, which typically begin with just a
1397 =item B<-osbc>, B<--outdent-static-block-comments>
1399 The command B<-osbc> will cause static block comments to be outdented by 2
1400 spaces (or whatever B<-ci=n> has been set to), if possible.
1404 =item Static Side Comments
1406 Static side comments are side comments with a special leading pattern.
1407 This option can be useful for controlling how commented code is displayed
1408 when it is a side comment.
1412 =item B<-ssc>, B<--static-side-comments>
1414 When B<-ssc> is used, a side comment with a static leading pattern, which is
1415 C<##> by default, will be spaced only a single space from previous
1416 character, and it will not be vertically aligned with other side comments.
1418 The default is B<-nssc>.
1420 =item B<-sscp=string>, B<--static-side-comment-prefix=string>
1422 This parameter defines the prefix used to identify static side comments
1423 when the B<-ssc> parameter is set. The default prefix is C<##>,
1424 corresponding to C<-sscp=##>.
1426 Please note that B<-sscp> merely defines the pattern used to identify
1427 static side comments; it will not be used unless the switch B<-ssc> is
1428 set. Also, note that this string is used in a perl regular expression
1429 which identifies these comments, so it must enable a valid regular
1430 expression to be formed.
1437 =head2 Skipping Selected Sections of Code
1439 Selected lines of code may be passed verbatim to the output without any
1440 formatting. This feature is enabled by default but can be disabled with
1441 the B<--noformat-skipping> or B<-nfs> flag. It should be used sparingly to
1442 avoid littering code with markers, but it might be helpful for working
1443 around occasional problems. For example it might be useful for keeping
1444 the indentation of old commented code unchanged, keeping indentation of
1445 long blocks of aligned comments unchanged, keeping certain list
1446 formatting unchanged, or working around a glitch in perltidy.
1450 =item B<-fs>, B<--format-skipping>
1452 This flag, which is enabled by default, causes any code between
1453 special beginning and ending comment markers to be passed to the
1454 output without formatting. The default beginning marker is #<<<
1455 and the default ending marker is #>>> but they
1456 may be changed (see next items below). Additional text may appear on
1457 these special comment lines provided that it is separated from the
1458 marker by at least one space. For example
1460 #<<< do not let perltidy touch this
1468 The comment markers may be placed at any location that a block comment may
1469 appear. If they do not appear to be working, use the -log flag and examine the
1470 F<.LOG> file. Use B<-nfs> to disable this feature.
1472 =item B<-fsb=string>, B<--format-skipping-begin=string>
1474 The B<-fsb=string> parameter may be used to change the beginning marker for
1475 format skipping. The default is equivalent to -fsb='#<<<'. The string that
1476 you enter must begin with a # and should be in quotes as necessary to get past
1477 the command shell of your system. It is actually the leading text of a pattern
1478 that is constructed by appending a '\s', so you must also include backslashes
1479 for characters to be taken literally rather than as patterns.
1481 Some examples show how example strings become patterns:
1483 -fsb='#\{\{\{' becomes /^#\{\{\{\s/ which matches #{{{ but not #{{{{
1484 -fsb='#\*\*' becomes /^#\*\*\s/ which matches #** but not #***
1485 -fsb='#\*{2,}' becomes /^#\*{2,}\s/ which matches #** and #*****
1487 =item B<-fse=string>, B<--format-skipping-end=string>
1489 The B<-fsb=string> is the corresponding parameter used to change the
1490 ending marker for format skipping. The default is equivalent to
1495 =head2 Line Break Control
1497 The parameters in this section control breaks after
1498 non-blank lines of code. Blank lines are controlled
1499 separately by parameters in the section L<Blank Line
1504 =item B<-fnl>, B<--freeze-newlines>
1506 If you do not want any changes to the line breaks within
1507 lines of code in your script, set
1508 B<-fnl>, and they will remain fixed, and the rest of the commands in
1509 this section and sections
1510 L<Controlling List Formatting>,
1511 L<Retaining or Ignoring Existing Line Breaks>.
1512 You may want to use B<-noll> with this.
1514 Note: If you also want to keep your blank lines exactly
1515 as they are, you can use the B<-fbl> flag which is described
1516 in the section L<Blank Line Control>.
1518 =item B<-ce>, B<--cuddled-else>
1520 Enable the "cuddled else" style, in which C<else> and C<elsif> are
1521 follow immediately after the curly brace closing the previous block.
1522 The default is not to use cuddled elses, and is indicated with the flag
1523 B<-nce> or B<--nocuddled-else>. Here is a comparison of the
1535 else { # -nce (default)
1539 =item B<-bl>, B<--opening-brace-on-new-line>
1541 Use the flag B<-bl> to place the opening brace on a new line:
1543 if ( $input_file eq '-' ) # -bl
1545 important_function();
1548 This flag applies to all structural blocks, including named sub's (unless
1549 the B<-sbl> flag is set -- see next item).
1551 The default style, B<-nbl>, places an opening brace on the same line as
1552 the keyword introducing it. For example,
1554 if ( $input_file eq '-' ) { # -nbl (default)
1556 =item B<-sbl>, B<--opening-sub-brace-on-new-line>
1558 The flag B<-sbl> can be used to override the value of B<-bl> for
1559 the opening braces of named sub's. For example,
1563 produces this result:
1567 if (!defined($_[0])) {
1568 print("Hello, World\n");
1575 This flag is negated with B<-nsbl>. If B<-sbl> is not specified,
1576 the value of B<-bl> is used.
1578 =item B<-asbl>, B<--opening-anonymous-sub-brace-on-new-line>
1580 The flag B<-asbl> is like the B<-sbl> flag except that it applies
1581 to anonymous sub's instead of named subs. For example
1585 produces this result:
1589 if ( !defined( $_[0] ) ) {
1590 print("Hello, World\n");
1593 print( $_[0], "\n" );
1597 This flag is negated with B<-nasbl>, and the default is B<-nasbl>.
1599 =item B<-bli>, B<--brace-left-and-indent>
1601 The flag B<-bli> is the same as B<-bl> but in addition it causes one
1602 unit of continuation indentation ( see B<-ci> ) to be placed before
1603 an opening and closing block braces.
1607 if ( $input_file eq '-' ) # -bli
1609 important_function();
1612 By default, this extra indentation occurs for blocks of type:
1613 B<if>, B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>,
1614 B<while>, B<until>, and also with a preceding label. The next item
1615 shows how to change this.
1617 =item B<-blil=s>, B<--brace-left-and-indent-list=s>
1619 Use this parameter to change the types of block braces for which the
1620 B<-bli> flag applies; see L<Specifying Block Types>. For example,
1621 B<-blil='if elsif else'> would apply it to only C<if/elsif/else> blocks.
1623 =item B<-bar>, B<--opening-brace-always-on-right>
1625 The default style, B<-nbl> places the opening code block brace on a new
1626 line if it does not fit on the same line as the opening keyword, like
1629 if ( $bigwasteofspace1 && $bigwasteofspace2
1630 || $bigwasteofspace3 && $bigwasteofspace4 )
1632 big_waste_of_time();
1635 To force the opening brace to always be on the right, use the B<-bar>
1636 flag. In this case, the above example becomes
1638 if ( $bigwasteofspace1 && $bigwasteofspace2
1639 || $bigwasteofspace3 && $bigwasteofspace4 ) {
1640 big_waste_of_time();
1643 A conflict occurs if both B<-bl> and B<-bar> are specified.
1645 =item B<-otr>, B<--opening-token-right> and related flags
1647 The B<-otr> flag is a hint that perltidy should not place a break between a
1648 comma and an opening token. For example:
1650 # default formatting
1651 push @{ $self->{$module}{$key} },
1653 accno => $ref->{accno},
1654 description => $ref->{description}
1658 push @{ $self->{$module}{$key} }, {
1659 accno => $ref->{accno},
1660 description => $ref->{description}
1663 The flag B<-otr> is actually an abbreviation for three other flags
1664 which can be used to control parens, hash braces, and square brackets
1665 separately if desired:
1667 -opr or --opening-paren-right
1668 -ohbr or --opening-hash-brace-right
1669 -osbr or --opening-square-bracket-right
1671 =item Vertical tightness of non-block curly braces, parentheses, and square brackets.
1673 These parameters control what shall be called vertical tightness. Here are the
1680 Opening tokens (except for block braces) are controlled by B<-vt=n>, or
1681 B<--vertical-tightness=n>, where
1683 -vt=0 always break a line after opening token (default).
1684 -vt=1 do not break unless this would produce more than one
1685 step in indentation in a line.
1686 -vt=2 never break a line after opening token
1690 You must also use the B<-lp> flag when you use the B<-vt> flag; the
1691 reason is explained below.
1695 Closing tokens (except for block braces) are controlled by B<-vtc=n>, or
1696 B<--vertical-tightness-closing=n>, where
1698 -vtc=0 always break a line before a closing token (default),
1699 -vtc=1 do not break before a closing token which is followed
1700 by a semicolon or another closing token, and is not in
1702 -vtc=2 never break before a closing token.
1704 The rules for B<-vtc=1> are designed to maintain a reasonable balance
1705 between tightness and readability in complex lists.
1709 Different controls may be applied to different token types,
1710 and it is also possible to control block braces; see below.
1714 Finally, please note that these vertical tightness flags are merely
1715 hints to the formatter, and it cannot always follow them. Things which
1716 make it difficult or impossible include comments, blank lines, blocks of
1717 code within a list, and possibly the lack of the B<-lp> parameter.
1718 Also, these flags may be ignored for very small lists (2 or 3 lines in
1723 Here are some examples:
1725 # perltidy -lp -vt=0 -vtc=0
1733 # perltidy -lp -vt=1 -vtc=0
1734 %romanNumerals = ( one => 'I',
1740 # perltidy -lp -vt=1 -vtc=1
1741 %romanNumerals = ( one => 'I',
1746 The difference between B<-vt=1> and B<-vt=2> is shown here:
1748 # perltidy -lp -vt=1
1750 mysprintf( "(void)find_threadsv(%s);",
1751 cstring( $threadsv_names[ $op->targ ] )
1755 # perltidy -lp -vt=2
1756 $init->add( mysprintf( "(void)find_threadsv(%s);",
1757 cstring( $threadsv_names[ $op->targ ] )
1761 With B<-vt=1>, the line ending in C<add(> does not combine with the next
1762 line because the next line is not balanced. This can help with
1763 readability, but B<-vt=2> can be used to ignore this rule.
1765 The tightest, and least readable, code is produced with both C<-vt=2> and
1768 # perltidy -lp -vt=2 -vtc=2
1769 $init->add( mysprintf( "(void)find_threadsv(%s);",
1770 cstring( $threadsv_names[ $op->targ ] ) ) );
1772 Notice how the code in all of these examples collapses vertically as
1773 B<-vt> increases, but the indentation remains unchanged. This is
1774 because perltidy implements the B<-vt> parameter by first formatting as
1775 if B<-vt=0>, and then simply overwriting one output line on top of the
1776 next, if possible, to achieve the desired vertical tightness. The
1777 B<-lp> indentation style has been designed to allow this vertical
1778 collapse to occur, which is why it is required for the B<-vt> parameter.
1780 The B<-vt=n> and B<-vtc=n> parameters apply to each type of container
1781 token. If desired, vertical tightness controls can be applied
1782 independently to each of the closing container token types.
1784 The parameters for controlling parentheses are B<-pvt=n> or
1785 B<--paren-vertical-tightness=n>, and B<-pcvt=n> or
1786 B<--paren-vertical-tightness-closing=n>.
1788 Likewise, the parameters for square brackets are B<-sbvt=n> or
1789 B<--square-bracket-vertical-tightness=n>, and B<-sbcvt=n> or
1790 B<--square-bracket-vertical-tightness-closing=n>.
1792 Finally, the parameters for controlling non-code block braces are
1793 B<-bvt=n> or B<--brace-vertical-tightness=n>, and B<-bcvt=n> or
1794 B<--brace-vertical-tightness-closing=n>.
1796 In fact, the parameter B<-vt=n> is actually just an abbreviation for
1797 B<-pvt=n -bvt=n sbvt=n>, and likewise B<-vtc=n> is an abbreviation
1798 for B<-pvtc=n -bvtc=n sbvtc=n>.
1800 =item B<-bbvt=n> or B<--block-brace-vertical-tightness=n>
1802 The B<-bbvt=n> flag is just like the B<-vt=n> flag but applies
1803 to opening code block braces.
1805 -bbvt=0 break after opening block brace (default).
1806 -bbvt=1 do not break unless this would produce more than one
1807 step in indentation in a line.
1808 -bbvt=2 do not break after opening block brace.
1810 It is necessary to also use either B<-bl> or B<-bli> for this to work,
1811 because, as with other vertical tightness controls, it is implemented by
1812 simply overwriting a line ending with an opening block brace with the
1813 subsequent line. For example:
1815 # perltidy -bli -bbvt=0
1816 if ( open( FILE, "< $File" ) )
1818 while ( $File = <FILE> )
1826 # perltidy -bli -bbvt=1
1827 if ( open( FILE, "< $File" ) )
1828 { while ( $File = <FILE> )
1835 By default this applies to blocks associated with keywords B<if>,
1836 B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>, B<while>,
1837 B<until>, and also with a preceding label. This can be changed with
1838 the parameter B<-bbvtl=string>, or
1839 B<--block-brace-vertical-tightness-list=string>, where B<string> is a
1840 space-separated list of block types. For more information on the
1841 possible values of this string, see L<Specifying Block Types>
1843 For example, if we want to just apply this style to C<if>,
1844 C<elsif>, and C<else> blocks, we could use
1845 C<perltidy -bli -bbvt=1 -bbvtl='if elsif else'>.
1847 There is no vertical tightness control for closing block braces; with
1848 one exception they will be placed on separate lines.
1849 The exception is that a cascade of closing block braces may
1850 be stacked on a single line. See B<-scbb>.
1852 =item B<-sot>, B<--stack-opening-tokens> and related flags
1854 The B<-sot> flag tells perltidy to "stack" opening tokens
1855 when possible to avoid lines with isolated opening tokens.
1860 $opt_c = Text::CSV_XS->new(
1869 $opt_c = Text::CSV_XS->new( {
1876 For detailed control of individual closing tokens the following
1877 controls can be used:
1879 -sop or --stack-opening-paren
1880 -sohb or --stack-opening-hash-brace
1881 -sosb or --stack-opening-square-bracket
1882 -sobb or --stack-opening-block-brace
1884 The flag B<-sot> is an abbreviation for B<-sop -sohb -sosb>.
1886 The flag B<-sobb> is a abbreviation for B<-bbvt=2 -bbvtl='*'>. This
1887 will case a cascade of opening block braces to appear on a single line,
1888 although this an uncommon occurrence except in test scripts.
1890 =item B<-sct>, B<--stack-closing-tokens> and related flags
1892 The B<-sct> flag tells perltidy to "stack" closing tokens
1893 when possible to avoid lines with isolated closing tokens.
1898 $opt_c = Text::CSV_XS->new(
1907 $opt_c = Text::CSV_XS->new(
1914 The B<-sct> flag is somewhat similar to the B<-vtc> flags, and in some
1915 cases it can give a similar result. The difference is that the B<-vtc>
1916 flags try to avoid lines with leading opening tokens by "hiding" them at
1917 the end of a previous line, whereas the B<-sct> flag merely tries to
1918 reduce the number of lines with isolated closing tokens by stacking them
1919 but does not try to hide them. For example:
1922 $opt_c = Text::CSV_XS->new(
1926 always_quote => 1, } );
1928 For detailed control of the stacking of individual closing tokens the
1929 following controls can be used:
1931 -scp or --stack-closing-paren
1932 -schb or --stack-closing-hash-brace
1933 -scsb or --stack-closing-square-bracket
1934 -scbb or --stack-closing-block-brace
1936 The flag B<-sct> is an abbreviation for stacking the non-block closing
1937 tokens, B<-scp -schb -scsb>.
1939 Stacking of closing block braces, B<-scbb>, causes a cascade of isolated
1940 closing block braces to be combined into a single line as in the following
1948 push( @lines, "$w1 $w2 $w3 $w4\n" );
1951 To simplify input even further for the case in which both opening and closing
1952 non-block containers are stacked, the flag B<-sac> or B<--stack-all-containers>
1953 is an abbreviation for B<-sot -sot>.
1955 =item B<-dnl>, B<--delete-old-newlines>
1957 By default, perltidy first deletes all old line break locations, and then it
1958 looks for good break points to match the desired line length. Use B<-ndnl>
1959 or B<--nodelete-old-newlines> to force perltidy to retain all old line break
1962 =item B<-anl>, B<--add-newlines>
1964 By default, perltidy will add line breaks when necessary to create
1965 continuations of long lines and to improve the script appearance. Use
1966 B<-nanl> or B<--noadd-newlines> to prevent any new line breaks.
1968 This flag does not prevent perltidy from eliminating existing line
1969 breaks; see B<--freeze-newlines> to completely prevent changes to line
1972 =item Controlling whether perltidy breaks before or after operators
1974 Four command line parameters provide some control over whether
1975 a line break should be before or after specific token types.
1976 Two parameters give detailed control:
1978 B<-wba=s> or B<--want-break-after=s>, and
1980 B<-wbb=s> or B<--want-break-before=s>.
1982 These parameters are each followed by a quoted string, B<s>, containing
1983 a list of token types (separated only by spaces). No more than one of each
1984 of these parameters should be specified, because repeating a
1985 command-line parameter always overwrites the previous one before
1986 perltidy ever sees it.
1988 By default, perltidy breaks B<after> these token types:
1989 % + - * / x != == >= <= =~ !~ < > | &
1990 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
1992 And perltidy breaks B<before> these token types by default:
1995 To illustrate, to cause a break after a concatenation operator, C<'.'>,
1996 rather than before it, the command line would be
2000 As another example, the following command would cause a break before
2001 math operators C<'+'>, C<'-'>, C<'/'>, and C<'*'>:
2005 These commands should work well for most of the token types that perltidy uses
2006 (use B<--dump-token-types> for a list). Also try the B<-D> flag on a short
2007 snippet of code and look at the .DEBUG file to see the tokenization. However,
2008 for a few token types there may be conflicts with hardwired logic which cause
2009 unexpected results. One example is curly braces, which should be controlled
2010 with the parameter B<bl> provided for that purpose.
2012 B<WARNING> Be sure to put these tokens in quotes to avoid having them
2013 misinterpreted by your command shell.
2015 Two additional parameters are available which, though they provide no further
2016 capability, can simplify input are:
2018 B<-baao> or B<--break-after-all-operators>,
2020 B<-bbao> or B<--break-before-all-operators>.
2022 The -baao sets the default to be to break after all of the following operators:
2024 % + - * / x != == >= <= =~ !~ < > | &
2025 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
2026 . : ? && || and or err xor
2028 and the B<-bbao> flag sets the default to break before all of these operators.
2029 These can be used to define an initial break preference which can be fine-tuned
2030 with the B<-wba> and B<-wbb> flags. For example, to break before all operators
2031 except an B<=> one could use --bbao -wba='=' rather than listing every
2032 single perl operator except B<=> on a -wbb flag.
2036 =head2 Controlling List Formatting
2038 Perltidy attempts to place comma-separated arrays of values in tables
2039 which look good. Its default algorithms usually work well, and they
2040 have been improving with each release, but several parameters are
2041 available to control list formatting.
2045 =item B<-boc>, B<--break-at-old-comma-breakpoints>
2047 This flag tells perltidy to try to break at all old commas. This is not
2048 the default. Normally, perltidy makes a best guess at list formatting,
2049 and seldom uses old comma breakpoints. Usually this works well,
2058 The default formatting will flatten this down to one line:
2060 # perltidy (default)
2061 my @list = ( 1, 1, 1, 1, 2, 1, 1, 3, 3, 1, 1, 4, 6, 4, 1, );
2063 which hides the structure. Using B<-boc>, plus additional flags
2064 to retain the original style, yields
2066 # perltidy -boc -lp -pt=2 -vt=1 -vtc=1
2073 A disadvantage of this flag is that all tables in the file
2074 must already be nicely formatted. For another possibility see
2075 the -fs flag in L<Skipping Selected Sections of Code>.
2077 =item B<-mft=n>, B<--maximum-fields-per-table=n>
2079 If the computed number of fields for any table exceeds B<n>, then it
2080 will be reduced to B<n>. The default value for B<n> is a large number,
2081 40. While this value should probably be left unchanged as a general
2082 rule, it might be used on a small section of code to force a list to
2083 have a particular number of fields per line, and then either the B<-boc>
2084 flag could be used to retain this formatting, or a single comment could
2085 be introduced somewhere to freeze the formatting in future applications
2098 =item B<-cab=n>, B<--comma-arrow-breakpoints=n>
2100 A comma which follows a comma arrow, '=>', is given special
2101 consideration. In a long list, it is common to break at all such
2102 commas. This parameter can be used to control how perltidy breaks at
2103 these commas. (However, it will have no effect if old comma breaks are
2104 being forced because B<-boc> is used). The possible values of B<n> are:
2106 n=0 break at all commas after =>
2107 n=1 stable: break at all commas after => if container is open,
2108 EXCEPT FOR one-line containers
2109 n=2 break at all commas after =>, BUT try to form the maximum
2110 maximum one-line container lengths
2111 n=3 do not treat commas after => specially at all
2112 n=4 break everything: like n=0 but ALSO break a short container with
2113 a => not followed by a comma when -vt=0 is used
2114 n=5 stable: like n=1 but ALSO break at open one-line containers when
2115 -vt=0 is used (default)
2117 For example, given the following single line, perltidy by default will
2118 not add any line breaks because it would break the existing one-line
2121 bless { B => $B, Root => $Root } => $package;
2123 Using B<-cab=0> will force a break after each comma-arrow item:
2131 If perltidy is subsequently run with this container broken, then by
2132 default it will break after each '=>' because the container is now
2133 broken. To reform a one-line container, the parameter B<-cab=2> could
2136 The flag B<-cab=3> can be used to prevent these commas from being
2137 treated specially. In this case, an item such as "01" => 31 is
2138 treated as a single item in a table. The number of fields in this table
2139 will be determined by the same rules that are used for any other table.
2144 "01" => 31, "02" => 29, "03" => 31, "04" => 30,
2145 "05" => 31, "06" => 30, "07" => 31, "08" => 31,
2146 "09" => 30, "10" => 31, "11" => 30, "12" => 31
2151 =head2 Retaining or Ignoring Existing Line Breaks
2153 Several additional parameters are available for controlling the extent
2154 to which line breaks in the input script influence the output script.
2155 In most cases, the default parameter values are set so that, if a choice
2156 is possible, the output style follows the input style. For example, if
2157 a short logical container is broken in the input script, then the
2158 default behavior is for it to remain broken in the output script.
2160 Most of the parameters in this section would only be required for a
2161 one-time conversion of a script from short container lengths to longer
2162 container lengths. The opposite effect, of converting long container
2163 lengths to shorter lengths, can be obtained by temporarily using a short
2164 maximum line length.
2168 =item B<-bol>, B<--break-at-old-logical-breakpoints>
2170 By default, if a logical expression is broken at a C<&&>, C<||>, C<and>,
2171 or C<or>, then the container will remain broken. Also, breaks
2172 at internal keywords C<if> and C<unless> will normally be retained.
2173 To prevent this, and thus form longer lines, use B<-nbol>.
2175 =item B<-bok>, B<--break-at-old-keyword-breakpoints>
2177 By default, perltidy will retain a breakpoint before keywords which may
2178 return lists, such as C<sort> and <map>. This allows chains of these
2179 operators to be displayed one per line. Use B<-nbok> to prevent
2180 retaining these breakpoints.
2182 =item B<-bot>, B<--break-at-old-ternary-breakpoints>
2184 By default, if a conditional (ternary) operator is broken at a C<:>,
2185 then it will remain broken. To prevent this, and thereby
2186 form longer lines, use B<-nbot>.
2188 =item B<-boa>, B<--break-at-old-attribute-breakpoints>
2190 By default, if an attribute list is broken at a C<:> in the source file, then
2191 it will remain broken. For example, given the following code, the line breaks
2192 at the ':'s will be retained:
2197 : Get('Name' => 'foo') : Set('Name');
2199 If the attributes are on a single line in the source code then they will remain
2200 on a single line if possible.
2202 To prevent this, and thereby always form longer lines, use B<-nboa>.
2204 =item B<-iob>, B<--ignore-old-breakpoints>
2206 Use this flag to tell perltidy to ignore existing line breaks to the
2207 maximum extent possible. This will tend to produce the longest possible
2208 containers, regardless of type, which do not exceed the line length
2211 =item B<-kis>, B<--keep-interior-semicolons>
2213 Use the B<-kis> flag to prevent breaking at a semicolon if
2214 there was no break there in the input file. Normally
2215 perltidy places a newline after each semicolon which
2216 terminates a statement unless several statements are
2217 contained within a one-line brace block. To illustrate,
2218 consider the following input lines:
2220 dbmclose(%verb_delim); undef %verb_delim;
2221 dbmclose(%expanded); undef %expanded;
2223 The default is to break after each statement, giving
2225 dbmclose(%verb_delim);
2227 dbmclose(%expanded);
2230 With B<perltidy -kis> the multiple statements are retained:
2232 dbmclose(%verb_delim); undef %verb_delim;
2233 dbmclose(%expanded); undef %expanded;
2235 The statements are still subject to the specified value
2236 of B<maximum-line-length> and will be broken if this
2237 maximum is exceeded.
2241 =head2 Blank Line Control
2243 Blank lines can improve the readability of a script if they are carefully
2244 placed. Perltidy has several commands for controlling the insertion,
2245 retention, and removal of blank lines.
2249 =item B<-fbl>, B<--freeze-blank-lines>
2251 Set B<-fbl> if you want to the blank lines in your script to
2252 remain exactly as they are. The rest of the parameters in
2253 this section may then be ignored. (Note: setting the B<-fbl> flag
2254 is equivalent to setting B<-mbl=0> and B<-kbl=2>).
2256 =item B<-bbc>, B<--blanks-before-comments>
2258 A blank line will be introduced before a full-line comment. This is the
2259 default. Use B<-nbbc> or B<--noblanks-before-comments> to prevent
2260 such blank lines from being introduced.
2262 =item B<-blbs=n>, B<--blank-lines-before-subs=n>
2264 The parameter B<-blbs=n> requests that least B<n> blank lines precede a sub
2265 definition which does not follow a comment and which is more than one-line
2266 long. The default is <-blbs=1>. B<BEGIN> and B<END> blocks are included.
2268 The requested number of blanks statement will be inserted regardless of the
2269 value of B<--maximum-consecutive-blank-lines=n> (B<-mbl=n>) with the exception
2270 that if B<-mbl=0> then no blanks will be output.
2272 This parameter interacts with the value B<k> of the parameter B<--maximum-consecutive-blank-lines=k> (B<-mbl=k>) as follows:
2274 1. If B<-mbl=0> then no blanks will be output. This allows all blanks to be suppressed with a single parameter. Otherwise,
2276 2. If the number of old blank lines in the script is less than B<n> then
2277 additional blanks will be inserted to make the total B<n> regardless of the
2280 3. If the number of old blank lines in the script equals or exceeds B<n> then
2281 this parameter has no effect, however the total will not exceed
2282 value specified on the B<-mbl=k> flag.
2285 =item B<-blbp=n>, B<--blank-lines-before-packages=n>
2287 The parameter B<-blbp=n> requests that least B<n> blank lines precede a package
2288 which does not follow a comment. The default is <-blbp=1>.
2290 This parameter interacts with the value B<k> of the parameter
2291 B<--maximum-consecutive-blank-lines=k> (B<-mbl=k>) in the same way as described
2292 for the previous item B<-blbs=n>.
2295 =item B<-bbs>, B<--blanks-before-subs>
2297 For compatibility with previous versions, B<-bbs> or B<--blanks-before-subs>
2298 is equivalent to F<-blbp=1> and F<-blbs=1>.
2300 Likewise, B<-nbbs> or B<--noblanks-before-subs>
2301 is equivalent to F<-blbp=0> and F<-blbs=0>.
2303 =item B<-bbb>, B<--blanks-before-blocks>
2305 A blank line will be introduced before blocks of coding delimited by
2306 B<for>, B<foreach>, B<while>, B<until>, and B<if>, B<unless>, in the following
2313 The block is not preceded by a comment.
2317 The block is not a one-line block.
2321 The number of consecutive non-blank lines at the current indentation depth is at least B<-lbl>
2326 This is the default. The intention of this option is to introduce
2327 some space within dense coding.
2328 This is negated with B<-nbbb> or B<--noblanks-before-blocks>.
2330 =item B<-lbl=n> B<--long-block-line-count=n>
2332 This controls how often perltidy is allowed to add blank lines before
2333 certain block types (see previous section). The default is 8. Entering
2334 a value of B<0> is equivalent to entering a very large number.
2336 =item B<-mbl=n> B<--maximum-consecutive-blank-lines=n>
2338 This parameter specifies the maximum number of consecutive blank lines which
2339 will be output within code sections of a script. The default is n=1. If the
2340 input file has more than n consecutive blank lines, the number will be reduced
2341 to n except as noted above for the B<-blbp> and B<-blbs> parameters. If B<n=0>
2342 then no blank lines will be output (unless all old blank lines are retained
2343 with the B<-kbl=2> flag of the next section).
2345 This flag obviously does not apply to pod sections,
2346 here-documents, and quotes.
2348 =item B<-kbl=n>, B<--keep-old-blank-lines=n>
2350 The B<-kbl=n> flag gives you control over how your existing blank lines are
2353 The possible values of B<n> are:
2355 n=0 ignore all old blank lines
2356 n=1 stable: keep old blanks, but limited by the value of the B<-mbl=n> flag
2357 n=2 keep all old blank lines, regardless of the value of the B<-mbl=n> flag
2359 The default is B<n=1>.
2361 =item B<-sob>, B<--swallow-optional-blank-lines>
2363 This is equivalent to B<kbl=0> and is included for compatibility with
2366 =item B<-nsob>, B<--noswallow-optional-blank-lines>
2368 This is equivalent to B<kbl=1> and is included for compatibility with
2375 A style refers to a convenient collection of existing parameters.
2379 =item B<-gnu>, B<--gnu-style>
2381 B<-gnu> gives an approximation to the GNU Coding Standards (which do
2382 not apply to perl) as they are sometimes implemented. At present, this
2383 style overrides the default style with the following parameters:
2385 -lp -bl -noll -pt=2 -bt=2 -sbt=2 -icp
2387 =item B<-pbp>, B<--perl-best-practices>
2389 B<-pbp> is an abbreviation for the parameters in the book B<Perl Best Practices>
2392 -l=78 -i=4 -ci=4 -st -se -vt=2 -cti=0 -pt=1 -bt=1 -sbt=1 -bbt=1 -nsfs -nolq
2393 -wbb="% + - * / x != == >= <= =~ !~ < > | & =
2394 **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x="
2396 Please note that this parameter set includes -st and -se flags, which make
2397 perltidy act as a filter on one file only. These can be overridden by placing
2398 B<-nst> and/or B<-nse> after the -pbp parameter.
2400 Also note that the value of continuation indentation, -ci=4, is equal to the
2401 value of the full indentation, -i=4. In some complex statements perltidy will
2402 produce nicer results with -ci=2. This can be implemented by including -ci=2
2403 after the -pbp parameter. For example,
2408 : $type eq 'item' ? "the $section entry"
2409 : "the section on $section"
2413 ? ( $section ? ' in ' : '' ) . "the $page$page_ext manpage"
2414 : ' elsewhere in this document'
2417 # perltidy -pbp -ci=2
2420 : $type eq 'item' ? "the $section entry"
2421 : "the section on $section"
2425 ? ( $section ? ' in ' : '' ) . "the $page$page_ext manpage"
2426 : ' elsewhere in this document'
2431 =head2 Other Controls
2435 =item Deleting selected text
2437 Perltidy can selectively delete comments and/or pod documentation. The
2438 command B<-dac> or B<--delete-all-comments> will delete all comments
2439 B<and> all pod documentation, leaving just code and any leading system
2442 The command B<-dp> or B<--delete-pod> will remove all pod documentation
2445 Two commands which remove comments (but not pod) are: B<-dbc> or
2446 B<--delete-block-comments> and B<-dsc> or B<--delete-side-comments>.
2447 (Hanging side comments will be deleted with block comments here.)
2449 The negatives of these commands also work, and are the defaults. When
2450 block comments are deleted, any leading 'hash-bang' will be retained.
2451 Also, if the B<-x> flag is used, any system commands before a leading
2452 hash-bang will be retained (even if they are in the form of comments).
2454 =item Writing selected text to a file
2456 When perltidy writes a formatted text file, it has the ability to also
2457 send selected text to a file with a F<.TEE> extension. This text can
2458 include comments and pod documentation.
2460 The command B<-tac> or B<--tee-all-comments> will write all comments
2461 B<and> all pod documentation.
2463 The command B<-tp> or B<--tee-pod> will write all pod documentation (but
2466 The commands which write comments (but not pod) are: B<-tbc> or
2467 B<--tee-block-comments> and B<-tsc> or B<--tee-side-comments>.
2468 (Hanging side comments will be written with block comments here.)
2470 The negatives of these commands also work, and are the defaults.
2472 =item Using a F<.perltidyrc> command file
2474 If you use perltidy frequently, you probably won't be happy until you
2475 create a F<.perltidyrc> file to avoid typing commonly-used parameters.
2476 Perltidy will first look in your current directory for a command file
2477 named F<.perltidyrc>. If it does not find one, it will continue looking
2478 for one in other standard locations.
2480 These other locations are system-dependent, and may be displayed with
2481 the command C<perltidy -dpro>. Under Unix systems, it will first look
2482 for an environment variable B<PERLTIDY>. Then it will look for a
2483 F<.perltidyrc> file in the home directory, and then for a system-wide
2484 file F</usr/local/etc/perltidyrc>, and then it will look for
2485 F</etc/perltidyrc>. Note that these last two system-wide files do not
2486 have a leading dot. Further system-dependent information will be found
2487 in the INSTALL file distributed with perltidy.
2489 Under Windows, perltidy will also search for a configuration file named perltidy.ini since Windows does not allow files with a leading period (.).
2490 Use C<perltidy -dpro> to see the possible locations for your system.
2491 An example might be F<C:\Documents and Settings\All Users\perltidy.ini>.
2493 Another option is the use of the PERLTIDY environment variable.
2494 The method for setting environment variables depends upon the version of
2495 Windows that you are using. Instructions for Windows 95 and later versions can
2498 http://www.netmanage.com/000/20021101_005_tcm21-6336.pdf
2500 Under Windows NT / 2000 / XP the PERLTIDY environment variable can be placed in
2501 either the user section or the system section. The later makes the
2502 configuration file common to all users on the machine. Be sure to enter the
2503 full path of the configuration file in the value of the environment variable.
2504 Ex. PERLTIDY=C:\Documents and Settings\perltidy.ini
2506 The configuration file is free format, and simply a list of parameters, just as
2507 they would be entered on a command line. Any number of lines may be used, with
2508 any number of parameters per line, although it may be easiest to read with one
2509 parameter per line. Comment text begins with a #, and there must
2510 also be a space before the # for side comments. It is a good idea to
2511 put complex parameters in either single or double quotes.
2513 Here is an example of a F<.perltidyrc> file:
2515 # This is a simple of a .perltidyrc configuration file
2516 # This implements a highly spaced style
2517 -se # errors to standard error output
2518 -w # show all warnings
2519 -bl # braces on new lines
2520 -pt=0 # parens not tight at all
2521 -bt=0 # braces not tight
2522 -sbt=0 # square brackets not tight
2524 The parameters in the F<.perltidyrc> file are installed first, so any
2525 parameters given on the command line will have priority over them.
2527 To avoid confusion, perltidy ignores any command in the .perltidyrc
2528 file which would cause some kind of dump and an exit. These are:
2530 -h -v -ddf -dln -dop -dsn -dtt -dwls -dwrs -ss
2532 There are several options may be helpful in debugging a F<.perltidyrc>
2539 A very helpful command is B<--dump-profile> or B<-dpro>. It writes a
2540 list of all configuration filenames tested to standard output, and
2541 if a file is found, it dumps the content to standard output before
2542 exiting. So, to find out where perltidy looks for its configuration
2543 files, and which one if any it selects, just enter
2549 It may be simplest to develop and test configuration files with
2550 alternative names, and invoke them with B<-pro=filename> on the command
2551 line. Then rename the desired file to F<.perltidyrc> when finished.
2555 The parameters in the F<.perltidyrc> file can be switched off with
2556 the B<-npro> option.
2560 The commands B<--dump-options>, B<--dump-defaults>, B<--dump-long-names>,
2561 and B<--dump-short-names>, all described below, may all be helpful.
2565 =item Creating a new abbreviation
2567 A special notation is available for use in a F<.perltidyrc> file
2568 for creating an abbreviation for a group
2569 of options. This can be used to create a
2570 shorthand for one or more styles which are frequently, but not always,
2571 used. The notation is to group the options within curly braces which
2572 are preceded by the name of the alias (without leading dashes), like this:
2579 where B<newword> is the abbreviation, and B<opt1>, etc, are existing parameters
2580 I<or other abbreviations>. The main syntax requirement is that
2581 the new abbreviation must begin on a new line.
2582 Space before and after the curly braces is optional.
2584 specific example, the following line
2586 airy {-bl -pt=0 -bt=0 -sbt=0}
2588 could be placed in a F<.perltidyrc> file, and then invoked at will with
2590 perltidy -airy somefile.pl
2592 (Either C<-airy> or C<--airy> may be used).
2594 =item Skipping leading non-perl commands with B<-x> or B<--look-for-hash-bang>
2596 If your script has leading lines of system commands or other text which
2597 are not valid perl code, and which are separated from the start of the
2598 perl code by a "hash-bang" line, ( a line of the form C<#!...perl> ),
2599 you must use the B<-x> flag to tell perltidy not to parse and format any
2600 lines before the "hash-bang" line. This option also invokes perl with a
2601 -x flag when checking the syntax. This option was originally added to
2602 allow perltidy to parse interactive VMS scripts, but it should be used
2603 for any script which is normally invoked with C<perl -x>.
2605 =item Making a file unreadable
2607 The goal of perltidy is to improve the readability of files, but there
2608 are two commands which have the opposite effect, B<--mangle> and
2609 B<--extrude>. They are actually
2610 merely aliases for combinations of other parameters. Both of these
2611 strip all possible whitespace, but leave comments and pod documents,
2612 so that they are essentially reversible. The
2613 difference between these is that B<--mangle> puts the fewest possible
2614 line breaks in a script while B<--extrude> puts the maximum possible.
2615 Note that these options do not provided any meaningful obfuscation, because
2616 perltidy can be used to reformat the files. They were originally
2617 developed to help test the tokenization logic of perltidy, but they
2619 One use for B<--mangle> is the following:
2621 perltidy --mangle myfile.pl -st | perltidy -o myfile.pl.new
2623 This will form the maximum possible number of one-line blocks (see next
2624 section), and can sometimes help clean up a badly formatted script.
2626 A similar technique can be used with B<--extrude> instead of B<--mangle>
2627 to make the minimum number of one-line blocks.
2629 Another use for B<--mangle> is to combine it with B<-dac> to reduce
2630 the file size of a perl script.
2632 =item One-line blocks
2634 There are a few points to note regarding one-line blocks. A one-line
2635 block is something like this,
2637 if ($x > 0) { $y = 1 / $x }
2639 where the contents within the curly braces is short enough to fit
2642 With few exceptions, perltidy retains existing one-line blocks, if it
2643 is possible within the line-length constraint, but it does not attempt
2644 to form new ones. In other words, perltidy will try to follow the
2645 one-line block style of the input file.
2647 If an existing one-line block is longer than the maximum line length,
2648 however, it will be broken into multiple lines. When this happens, perltidy
2649 checks for and adds any optional terminating semicolon (unless the B<-nasc>
2650 option is used) if the block is a code block.
2652 The main exception is that perltidy will attempt to form new one-line
2653 blocks following the keywords C<map>, C<eval>, and C<sort>, because
2654 these code blocks are often small and most clearly displayed in a single
2657 One-line block rules can conflict with the cuddled-else option. When
2658 the cuddled-else option is used, perltidy retains existing one-line
2659 blocks, even if they do not obey cuddled-else formatting.
2661 Occasionally, when one-line blocks get broken because they exceed the
2662 available line length, the formatting will violate the requested brace style.
2663 If this happens, reformatting the script a second time should correct
2668 The following flags are available for debugging:
2670 B<--dump-defaults> or B<-ddf> will write the default option set to standard output and quit
2672 B<--dump-profile> or B<-dpro> will write the name of the current
2673 configuration file and its contents to standard output and quit.
2675 B<--dump-options> or B<-dop> will write current option set to standard
2678 B<--dump-long-names> or B<-dln> will write all command line long names (passed
2679 to Get_options) to standard output and quit.
2681 B<--dump-short-names> or B<-dsn> will write all command line short names
2682 to standard output and quit.
2684 B<--dump-token-types> or B<-dtt> will write a list of all token types
2685 to standard output and quit.
2687 B<--dump-want-left-space> or B<-dwls> will write the hash %want_left_space
2688 to standard output and quit. See the section on controlling whitespace
2691 B<--dump-want-right-space> or B<-dwrs> will write the hash %want_right_space
2692 to standard output and quit. See the section on controlling whitespace
2695 B<--no-memoize> or B<-nmem> will turn of memoizing.
2696 Memoization can reduce run time when running perltidy repeatedly in a
2697 single process. It is on by default but can be deactivated for
2698 testing with B<-nmem>.
2700 B<-DEBUG> will write a file with extension F<.DEBUG> for each input file
2701 showing the tokenization of all lines of code.
2703 =item Working with MakeMaker, AutoLoader and SelfLoader
2705 The first $VERSION line of a file which might be eval'd by MakeMaker
2706 is passed through unchanged except for indentation.
2707 Use B<--nopass-version-line>, or B<-npvl>, to deactivate this feature.
2709 If the AutoLoader module is used, perltidy will continue formatting
2710 code after seeing an __END__ line.
2711 Use B<--nolook-for-autoloader>, or B<-nlal>, to deactivate this feature.
2713 Likewise, if the SelfLoader module is used, perltidy will continue formatting
2714 code after seeing a __DATA__ line.
2715 Use B<--nolook-for-selfloader>, or B<-nlsl>, to deactivate this feature.
2717 =item Working around problems with older version of Perl
2719 Perltidy contains a number of rules which help avoid known subtleties
2720 and problems with older versions of perl, and these rules always
2721 take priority over whatever formatting flags have been set. For example,
2722 perltidy will usually avoid starting a new line with a bareword, because
2723 this might cause problems if C<use strict> is active.
2725 There is no way to override these rules.
2733 =item The B<-html> master switch
2735 The flag B<-html> causes perltidy to write an html file with extension
2736 F<.html>. So, for example, the following command
2738 perltidy -html somefile.pl
2740 will produce a syntax-colored html file named F<somefile.pl.html>
2741 which may be viewed with a browser.
2743 B<Please Note>: In this case, perltidy does not do any formatting to the
2744 input file, and it does not write a formatted file with extension
2745 F<.tdy>. This means that two perltidy runs are required to create a
2746 fully reformatted, html copy of a script.
2748 =item The B<-pre> flag for code snippets
2750 When the B<-pre> flag is given, only the pre-formatted section, within
2751 the <PRE> and </PRE> tags, will be output. This simplifies inclusion
2752 of the output in other files. The default is to output a complete
2755 =item The B<-nnn> flag for line numbering
2757 When the B<-nnn> flag is given, the output lines will be numbered.
2759 =item The B<-toc>, or B<--html-table-of-contents> flag
2761 By default, a table of contents to packages and subroutines will be
2762 written at the start of html output. Use B<-ntoc> to prevent this.
2763 This might be useful, for example, for a pod document which contains a
2764 number of unrelated code snippets. This flag only influences the code
2765 table of contents; it has no effect on any table of contents produced by
2766 pod2html (see next item).
2768 =item The B<-pod>, or B<--pod2html> flag
2770 There are two options for formatting pod documentation. The default is
2771 to pass the pod through the Pod::Html module (which forms the basis of
2772 the pod2html utility). Any code sections are formatted by perltidy, and
2773 the results then merged. Note: perltidy creates a temporary file when
2774 Pod::Html is used; see L<"FILES">. Also, Pod::Html creates temporary
2775 files for its cache.
2777 NOTE: Perltidy counts the number of C<=cut> lines, and either moves the
2778 pod text to the top of the html file if there is one C<=cut>, or leaves
2779 the pod text in its original order (interleaved with code) otherwise.
2781 Most of the flags accepted by pod2html may be included in the perltidy
2782 command line, and they will be passed to pod2html. In some cases,
2783 the flags have a prefix C<pod> to emphasize that they are for the
2784 pod2html, and this prefix will be removed before they are passed to
2785 pod2html. The flags which have the additional C<pod> prefix are:
2787 --[no]podheader --[no]podindex --[no]podrecurse --[no]podquiet
2788 --[no]podverbose --podflush
2790 The flags which are unchanged from their use in pod2html are:
2792 --backlink=s --cachedir=s --htmlroot=s --libpods=s --title=s
2793 --podpath=s --podroot=s
2795 where 's' is an appropriate character string. Not all of these flags are
2796 available in older versions of Pod::Html. See your Pod::Html documentation for
2799 The alternative, indicated with B<-npod>, is not to use Pod::Html, but
2800 rather to format pod text in italics (or whatever the stylesheet
2801 indicates), without special html markup. This is useful, for example,
2802 if pod is being used as an alternative way to write comments.
2804 =item The B<-frm>, or B<--frames> flag
2806 By default, a single html output file is produced. This can be changed
2807 with the B<-frm> option, which creates a frame holding a table of
2808 contents in the left panel and the source code in the right side. This
2809 simplifies code browsing. Assume, for example, that the input file is
2810 F<MyModule.pm>. Then, for default file extension choices, these three
2811 files will be created:
2813 MyModule.pm.html - the frame
2814 MyModule.pm.toc.html - the table of contents
2815 MyModule.pm.src.html - the formatted source code
2817 Obviously this file naming scheme requires that output be directed to a real
2818 file (as opposed to, say, standard output). If this is not the
2819 case, or if the file extension is unknown, the B<-frm> option will be
2822 =item The B<-text=s>, or B<--html-toc-extension> flag
2824 Use this flag to specify the extra file extension of the table of contents file
2825 when html frames are used. The default is "toc".
2826 See L<Specifying File Extensions>.
2828 =item The B<-sext=s>, or B<--html-src-extension> flag
2830 Use this flag to specify the extra file extension of the content file when html
2831 frames are used. The default is "src".
2832 See L<Specifying File Extensions>.
2834 =item The B<-hent>, or B<--html-entities> flag
2836 This flag controls the use of Html::Entities for html formatting. By
2837 default, the module Html::Entities is used to encode special symbols.
2838 This may not be the right thing for some browser/language
2839 combinations. Use --nohtml-entities or -nhent to prevent this.
2843 Style sheets make it very convenient to control and adjust the
2844 appearance of html pages. The default behavior is to write a page of
2845 html with an embedded style sheet.
2847 An alternative to an embedded style sheet is to create a page with a
2848 link to an external style sheet. This is indicated with the
2849 B<-css=filename>, where the external style sheet is F<filename>. The
2850 external style sheet F<filename> will be created if and only if it does
2851 not exist. This option is useful for controlling multiple pages from a
2854 To cause perltidy to write a style sheet to standard output and exit,
2855 use the B<-ss>, or B<--stylesheet>, flag. This is useful if the style
2856 sheet could not be written for some reason, such as if the B<-pre> flag
2857 was used. Thus, for example,
2859 perltidy -html -ss >mystyle.css
2861 will write a style sheet with the default properties to file
2864 The use of style sheets is encouraged, but a web page without a style
2865 sheets can be created with the flag B<-nss>. Use this option if you
2866 must to be sure that older browsers (roughly speaking, versions prior to
2867 4.0 of Netscape Navigator and Internet Explorer) can display the
2868 syntax-coloring of the html files.
2870 =item Controlling HTML properties
2872 Note: It is usually more convenient to accept the default properties
2873 and then edit the stylesheet which is produced. However, this section
2874 shows how to control the properties with flags to perltidy.
2876 Syntax colors may be changed from their default values by flags of the either
2877 the long form, B<-html-color-xxxxxx=n>, or more conveniently the short form,
2878 B<-hcx=n>, where B<xxxxxx> is one of the following words, and B<x> is the
2879 corresponding abbreviation:
2882 ---------- -------- --
2885 identifier identifier i
2886 bareword, function bareword w
2888 quite, pattern quote q
2889 here doc text here-doc-text h
2890 here doc target here-doc-target hh
2891 punctuation punctuation pu
2893 structural braces structure s
2894 semicolon semicolon sc
2898 sub definition name subroutine m
2899 pod text pod-text pd
2901 A default set of colors has been defined, but they may be changed by providing
2902 values to any of the following parameters, where B<n> is either a 6 digit
2903 hex RGB color value or an ascii name for a color, such as 'red'.
2905 To illustrate, the following command will produce an html
2906 file F<somefile.pl.html> with "aqua" keywords:
2908 perltidy -html -hck=00ffff somefile.pl
2910 and this should be equivalent for most browsers:
2912 perltidy -html -hck=aqua somefile.pl
2914 Perltidy merely writes any non-hex names that it sees in the html file.
2915 The following 16 color names are defined in the HTML 3.2 standard:
2934 Many more names are supported in specific browsers, but it is safest
2935 to use the hex codes for other colors. Helpful color tables can be
2936 located with an internet search for "HTML color tables".
2938 Besides color, two other character attributes may be set: bold, and italics.
2939 To set a token type to use bold, use the flag
2940 B<--html-bold-xxxxxx> or B<-hbx>, where B<xxxxxx> or B<x> are the long
2941 or short names from the above table. Conversely, to set a token type to
2942 NOT use bold, use B<--nohtml-bold-xxxxxx> or B<-nhbx>.
2944 Likewise, to set a token type to use an italic font, use the flag
2945 B<--html-italic-xxxxxx> or B<-hix>, where again B<xxxxxx> or B<x> are the
2946 long or short names from the above table. And to set a token type to
2947 NOT use italics, use B<--nohtml-italic-xxxxxx> or B<-nhix>.
2949 For example, to use bold braces and lime color, non-bold, italics keywords the
2950 following command would be used:
2952 perltidy -html -hbs -hck=00FF00 -nhbk -hik somefile.pl
2954 The background color can be specified with B<--html-color-background=n>,
2955 or B<-hcbg=n> for short, where n is a 6 character hex RGB value. The
2956 default color of text is the value given to B<punctuation>, which is
2959 Here are some notes and hints:
2961 1. If you find a preferred set of these parameters, you may want
2962 to create a F<.perltidyrc> file containing them. See the perltidy man
2963 page for an explanation.
2965 2. Rather than specifying values for these parameters, it is probably
2966 easier to accept the defaults and then edit a style sheet. The style
2967 sheet contains comments which should make this easy.
2969 3. The syntax-colored html files can be very large, so it may be best to
2970 split large files into smaller pieces to improve download times.
2974 =head1 SOME COMMON INPUT CONVENTIONS
2976 =head2 Specifying Block Types
2978 Several parameters which refer to code block types may be customized by also
2979 specifying an associated list of block types. The type of a block is the name
2980 of the keyword which introduces that block, such as B<if>, B<else>, or B<sub>.
2981 An exception is a labeled block, which has no keyword, and should be specified
2982 with just a colon. To specify all blocks use B<'*'>.
2984 For example, the following parameter specifies C<sub>, labels, C<BEGIN>, and
2987 -cscl="sub : BEGIN END"
2989 (the meaning of the -cscl parameter is described above.) Note that
2990 quotes are required around the list of block types because of the
2991 spaces. For another example, the following list specifies all block types
2992 for vertical tightness:
2996 =head2 Specifying File Extensions
2998 Several parameters allow default file extensions to be overridden. For
2999 example, a backup file extension may be specified with B<-bext=ext>,
3000 where B<ext> is some new extension. In order to provides the user some
3001 flexibility, the following convention is used in all cases to decide if
3002 a leading '.' should be used. If the extension C<ext> begins with
3003 C<A-Z>, C<a-z>, or C<0-9>, then it will be appended to the filename with
3004 an intermediate '.' (or perhaps an '_' on VMS systems). Otherwise, it
3005 will be appended directly.
3007 For example, suppose the file is F<somefile.pl>. For C<-bext=old>, a '.' is
3008 added to give F<somefile.pl.old>. For C<-bext=.old>, no additional '.' is
3009 added, so again the backup file is F<somefile.pl.old>. For C<-bext=~>, then no
3010 dot is added, and the backup file will be F<somefile.pl~> .
3012 =head1 SWITCHES WHICH MAY BE NEGATED
3014 The following list shows all short parameter names which allow a prefix
3015 'n' to produce the negated form:
3017 D anl asc aws b bbb bbc bbs bl bli boc bok bol bot ce
3018 csc dac dbc dcsc ddf dln dnl dop dp dpro dsc dsm dsn dtt dwls
3019 dwrs dws f fll frm fs hsc html ibc icb icp iob isbc lal log
3020 lp lsl ohbr okw ola oll opr opt osbr otr ple pod pvl q
3021 sbc sbl schb scp scsb sct se sfp sfs skp sob sohb sop sosb sot
3022 ssc st sts syn t tac tbc toc tp tqw tsc w x bar kis
3024 Equivalently, the prefix 'no' or 'no-' on the corresponding long names may be
3031 =item Parsing Limitations
3033 Perltidy should work properly on most perl scripts. It does a lot of
3034 self-checking, but still, it is possible that an error could be
3035 introduced and go undetected. Therefore, it is essential to make
3036 careful backups and to test reformatted scripts.
3038 The main current limitation is that perltidy does not scan modules
3039 included with 'use' statements. This makes it necessary to guess the
3040 context of any bare words introduced by such modules. Perltidy has good
3041 guessing algorithms, but they are not infallible. When it must guess,
3042 it leaves a message in the log file.
3044 If you encounter a bug, please report it.
3046 =item What perltidy does not parse and format
3048 Perltidy indents but does not reformat comments and C<qw> quotes.
3049 Perltidy does not in any way modify the contents of here documents or
3050 quoted text, even if they contain source code. (You could, however,
3051 reformat them separately). Perltidy does not format 'format' sections
3052 in any way. And, of course, it does not modify pod documents.
3060 =item Temporary files
3062 Under the -html option with the default --pod2html flag, a temporary file is
3063 required to pass text to Pod::Html. Unix systems will try to use the POSIX
3064 tmpnam() function. Otherwise the file F<perltidy.TMP> will be temporarily
3065 created in the current working directory.
3067 =item Special files when standard input is used
3069 When standard input is used, the log file, if saved, is F<perltidy.LOG>,
3070 and any errors are written to F<perltidy.ERR> unless the B<-se> flag is
3071 set. These are saved in the current working directory.
3073 =item Files overwritten
3075 The following file extensions are used by perltidy, and files with these
3076 extensions may be overwritten or deleted: F<.ERR>, F<.LOG>, F<.TEE>,
3077 and/or F<.tdy>, F<.html>, and F<.bak>, depending on the run type and
3080 =item Files extensions limitations
3082 Perltidy does not operate on files for which the run could produce a file with
3083 a duplicated file extension. These extensions include F<.LOG>, F<.ERR>,
3084 F<.TEE>, and perhaps F<.tdy> and F<.bak>, depending on the run type. The
3085 purpose of this rule is to prevent generating confusing filenames such as
3086 F<somefile.tdy.tdy.tdy>.
3092 perlstyle(1), Perl::Tidy(3)
3096 This man page documents perltidy version 20140328.
3100 Michael Cartmell supplied code for adaptation to VMS and helped with
3103 Yves Orton supplied code for adaptation to the various versions
3106 Axel Rose supplied a patch for MacPerl.
3108 Hugh S. Myers designed and implemented the initial Perl::Tidy module interface.
3110 Many others have supplied key ideas, suggestions, and bug reports;
3111 see the CHANGES file.
3116 email: perltidy at users.sourceforge.net
3117 http://perltidy.sourceforge.net
3121 Copyright (c) 2000-2012 by Steve Hancock
3125 This package is free software; you can redistribute it and/or modify it
3126 under the terms of the "GNU General Public License".
3128 Please refer to the file "COPYING" for details.
3132 This package is distributed in the hope that it will be useful,
3133 but WITHOUT ANY WARRANTY; without even the implied warranty of
3134 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
3136 See the "GNU General Public License" for more details.