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 (see B<-st>), the output will go to
179 F<filename.tdy>. [Note: - does not redirect to standard output. Use
182 =item B<-st>, B<--standard-output>
184 Perltidy must be able to operate on an arbitrarily large number of files
185 in a single run, with each output being directed to a different output
186 file. Obviously this would conflict with outputting to the single
187 standard output device, so a special flag, B<-st>, is required to
188 request outputting to the standard output. For example,
190 perltidy somefile.pl -st >somefile.new.pl
192 This option may only be used if there is just a single input file.
193 The default is B<-nst> or B<--nostandard-output>.
195 =item B<-se>, B<--standard-error-output>
197 If perltidy detects an error when processing file F<somefile.pl>, its
198 default behavior is to write error messages to file F<somefile.pl.ERR>.
199 Use B<-se> to cause all error messages to be sent to the standard error
200 output stream instead. This directive may be negated with B<-nse>.
201 Thus, you may place B<-se> in a F<.perltidyrc> and override it when
202 desired with B<-nse> on the command line.
204 =item B<-oext>=ext, B<--output-file-extension>=ext
206 Change the extension of the output file to be F<ext> instead of the
207 default F<tdy> (or F<html> in case the -B<-html> option is used).
208 See L<Specifying File Extensions>.
210 =item B<-opath>=path, B<--output-path>=path
212 When perltidy creates a filename for an output file, by default it merely
213 appends an extension to the path and basename of the input file. This
214 parameter causes the path to be changed to F<path> instead.
216 The path should end in a valid path separator character, but perltidy will try
217 to add one if it is missing.
221 perltidy somefile.pl -opath=/tmp/
223 will produce F</tmp/somefile.pl.tdy>. Otherwise, F<somefile.pl.tdy> will
224 appear in whatever directory contains F<somefile.pl>.
226 If the path contains spaces, it should be placed in quotes.
228 This parameter will be ignored if output is being directed to standard output,
229 or if it is being specified explicitly with the B<-o=s> parameter.
231 =item B<-b>, B<--backup-and-modify-in-place>
233 Modify the input file or files in-place and save the original with the
234 extension F<.bak>. Any existing F<.bak> file will be deleted. See next
235 item for changing the default backup extension, and for eliminating the
236 backup file altogether.
238 A B<-b> flag will be ignored if input is from standard input or goes to
239 standard output, or if the B<-html> flag is set.
241 In particular, if you want to use both the B<-b> flag and the B<-pbp>
242 (--perl-best-practices) flag, then you must put a B<-nst> flag after the
243 B<-pbp> flag because it contains a B<-st> flag as one of its components,
244 which means that output will go to the standard output stream.
246 =item B<-bext>=ext, B<--backup-file-extension>=ext
248 This parameter serves two purposes: (1) to change the extension of the backup
249 file to be something other than the default F<.bak>, and (2) to indicate
250 that no backup file should be saved.
252 To change the default extension to something other than F<.bak> see
253 L<Specifying File Extensions>.
255 A backup file of the source is always written, but you can request that it
256 be deleted at the end of processing if there were no errors. This is risky
257 unless the source code is being maintained with a source code control
260 To indicate that the backup should be deleted include one forward slash,
261 B</>, in the extension. If any text remains after the slash is removed
262 it will be used to define the backup file extension (which is always
263 created and only deleted if there were no errors).
265 Here are some examples:
267 Parameter Extension Backup File Treatment
268 <-bext=bak> F<.bak> Keep (same as the default behavior)
269 <-bext='/'> F<.bak> Delete if no errors
270 <-bext='/backup'> F<.backup> Delete if no errors
271 <-bext='original/'> F<.original> Delete if no errors
273 =item B<-w>, B<--warning-output>
275 Setting B<-w> causes any non-critical warning
276 messages to be reported as errors. These include messages
277 about possible pod problems, possibly bad starting indentation level,
278 and cautions about indirect object usage. The default, B<-nw> or
279 B<--nowarning-output>, is not to include these warnings.
281 =item B<-q>, B<--quiet>
283 Deactivate error messages and syntax checking (for running under
286 For example, if you use a vi-style editor, such as vim, you may execute
287 perltidy as a filter from within the editor using something like
291 where C<n1,n2> represents the selected text. Without the B<-q> flag,
292 any error message may mess up your screen, so be prepared to use your
295 =item B<-log>, B<--logfile>
297 Save the F<.LOG> file, which has many useful diagnostics. Perltidy always
298 creates a F<.LOG> file, but by default it is deleted unless a program bug is
299 suspected. Setting the B<-log> flag forces the log file to be saved.
301 =item B<-g=n>, B<--logfile-gap=n>
303 Set maximum interval between input code lines in the logfile. This purpose of
304 this flag is to assist in debugging nesting errors. The value of C<n> is
305 optional. If you set the flag B<-g> without the value of C<n>, it will be
306 taken to be 1, meaning that every line will be written to the log file. This
307 can be helpful if you are looking for a brace, paren, or bracket nesting error.
309 Setting B<-g> also causes the logfile to be saved, so it is not necessary to
310 also include B<-log>.
312 If no B<-g> flag is given, a value of 50 will be used, meaning that at least
313 every 50th line will be recorded in the logfile. This helps prevent
314 excessively long log files.
316 Setting a negative value of C<n> is the same as not setting B<-g> at all.
318 =item B<-npro> B<--noprofile>
320 Ignore any F<.perltidyrc> command file. Normally, perltidy looks first in
321 your current directory for a F<.perltidyrc> file of parameters. (The format
322 is described below). If it finds one, it applies those options to the
323 initial default values, and then it applies any that have been defined
324 on the command line. If no F<.perltidyrc> file is found, it looks for one
325 in your home directory.
327 If you set the B<-npro> flag, perltidy will not look for this file.
329 =item B<-pro=filename> or B<--profile=filename>
331 To simplify testing and switching .perltidyrc files, this command may be
332 used to specify a configuration file which will override the default
333 name of .perltidyrc. There must not be a space on either side of the
334 '=' sign. For example, the line
336 perltidy -pro=testcfg
338 would cause file F<testcfg> to be used instead of the
339 default F<.perltidyrc>.
341 A pathname begins with three dots, e.g. ".../.perltidyrc", indicates that
342 the file should be searched for starting in the current directory and
343 working upwards. This makes it easier to have multiple projects each with
344 their own .perltidyrc in their root directories.
346 =item B<-opt>, B<--show-options>
348 Write a list of all options used to the F<.LOG> file.
349 Please see B<--dump-options> for a simpler way to do this.
351 =item B<-f>, B<--force-read-binary>
353 Force perltidy to process binary files. To avoid producing excessive
354 error messages, perltidy skips files identified by the system as non-text.
355 However, valid perl scripts containing binary data may sometimes be identified
356 as non-text, and this flag forces perltidy to process them.
360 =head1 FORMATTING OPTIONS
368 This flag disables all formatting and causes the input to be copied unchanged
369 to the output except for possible changes in line ending characters and any
370 pre- and post-filters. This can be useful in conjunction with a hierarchical
371 set of F<.perltidyrc> files to avoid unwanted code tidying. See also
372 L<Skipping Selected Sections of Code> for a way to avoid tidying specific
375 =item B<-i=n>, B<--indent-columns=n>
377 Use n columns per indentation level (default n=4).
379 =item B<-l=n>, B<--maximum-line-length=n>
381 The default maximum line length is n=80 characters. Perltidy will try
382 to find line break points to keep lines below this length. However, long
383 quotes and side comments may cause lines to exceed this length.
384 Setting B<-l=0> is equivalent to setting B<-l=(a large number)>.
386 =item B<-vmll>, B<--variable-maximum-line-length>
388 A problem arises using a fixed maximum line length with very deeply nested code
389 and data structures because eventually the amount of leading whitespace used
390 for indicating indentation takes up most or all of the available line width,
391 leaving little or no space for the actual code or data. One solution is to use
392 a vary long line length. Another solution is to use the B<-vmll> flag, which
393 basically tells perltidy to ignore leading whitespace when measuring the line
396 To be precise, when the B<-vmll> parameter is set, the maximum line length of a
397 line of code will be M+L*I, where
399 M is the value of --maximum-line-length=M (-l=M), default 80,
400 I is the value of --indent-columns=I (-i=I), default 4,
401 L is the indentation level of the line of code
403 When this flag is set, the choice of breakpoints for a block of code should be
404 essentially independent of its nesting depth. However, the absolute line
405 lengths, including leading whitespace, can still be arbitrarily large. This
406 problem can be avoided by including the next parameter.
408 The default is not to do this (B<-nvmll>).
410 =item B<-wc=n>, B<--whitespace-cycle=n>
412 This flag also addresses problems with very deeply nested code and data
413 structures. When the nesting depth exceeds the value B<n> the leading
414 whitespace will be reduced and start at a depth of 1 again. The result is that
415 blocks of code will shift back to the left rather than moving arbitrarily far
416 to the right. This occurs cyclically to any depth.
418 For example if one level of indentation equals 4 spaces (B<-i=4>, the default),
419 and one uses B<-wc=15>, then if the leading whitespace on a line exceeds about
420 4*15=60 spaces it will be reduced back to 4*1=4 spaces and continue increasing
421 from there. If the whitespace never exceeds this limit the formatting remains
424 The combination of B<-vmll> and B<-wc=n> provides a solution to the problem of
425 displaying arbitrarily deep data structures and code in a finite window,
426 although B<-wc=n> may of course be used without B<-vmll>.
428 The default is not to use this, which can also be indicated using B<-wc=0>.
432 Using tab characters will almost certainly lead to future portability
433 and maintenance problems, so the default and recommendation is not to
434 use them. For those who prefer tabs, however, there are two different
437 Except for possibly introducing tab indentation characters, as outlined
438 below, perltidy does not introduce any tab characters into your file,
439 and it removes any tabs from the code (unless requested not to do so
440 with B<-fws>). If you have any tabs in your comments, quotes, or
441 here-documents, they will remain.
445 =item B<-et=n>, B<--entab-leading-whitespace>
447 This flag causes each B<n> initial space characters to be replaced by
448 one tab character. Note that the integer B<n> is completely independent
449 of the integer specified for indentation parameter, B<-i=n>.
451 =item B<-t>, B<--tabs>
453 This flag causes one leading tab character to be inserted for each level
454 of indentation. Certain other features are incompatible with this
455 option, and if these options are also given, then a warning message will
456 be issued and this flag will be unset. One example is the B<-lp>
459 =item B<-dt=n>, B<--default-tabsize=n>
461 If the first line of code passed to perltidy contains leading tabs but no
462 tab scheme is specified for the output stream then perltidy must guess how many
463 spaces correspond to each leading tab. This number of spaces B<n>
464 corresponding to each leading tab of the input stream may be specified with
465 B<-dt=n>. The default is B<n=8>.
467 This flag has no effect if a tab scheme is specified for the output stream,
468 because then the input stream is assumed to use the same tab scheme and
469 indentation spaces as for the output stream (any other assumption would lead to
474 =item B<-syn>, B<--check-syntax>
476 This flag causes perltidy to run C<perl -c -T> to check syntax of input
477 and output. (To change the flags passed to perl, see the next
478 item, B<-pscf>). The results are written to the F<.LOG> file, which
479 will be saved if an error is detected in the output script. The output
480 script is not checked if the input script has a syntax error. Perltidy
481 does its own checking, but this option employs perl to get a "second
484 If perl reports errors in the input file, they will not be reported in
485 the error output unless the B<--warning-output> flag is given.
487 The default is B<NOT> to do this type of syntax checking (although
488 perltidy will still do as much self-checking as possible). The reason
489 is that it causes all code in BEGIN blocks to be executed, for all
490 modules being used, and this opens the door to security issues and
491 infinite loops when running perltidy.
493 =item B<-pscf=s>, B<-perl-syntax-check-flags=s>
495 When perl is invoked to check syntax, the normal flags are C<-c -T>. In
496 addition, if the B<-x> flag is given to perltidy, then perl will also be
497 passed a B<-x> flag. It should not normally be necessary to change
498 these flags, but it can be done with the B<-pscf=s> flag. For example,
499 if the taint flag, C<-T>, is not wanted, the flag could be set to be just
502 Perltidy will pass your string to perl with the exception that it will
503 add a B<-c> and B<-x> if appropriate. The F<.LOG> file will show
504 exactly what flags were passed to perl.
506 =item B<-io>, B<--indent-only>
508 This flag is used to deactivate all formatting and line break changes
509 within non-blank lines of code.
510 When it is in effect, the only change to the script will be
511 to the indentation and blank lines.
512 And any flags controlling whitespace and newlines will be ignored. You
513 might want to use this if you are perfectly happy with your whitespace
514 and line breaks, and merely want perltidy to handle the indentation.
515 (This also speeds up perltidy by well over a factor of two, so it might be
516 useful when perltidy is merely being used to help find a brace error in
519 Setting this flag is equivalent to setting B<--freeze-newlines> and
520 B<--freeze-whitespace>.
522 If you also want to keep your existing blank lines exactly
523 as they are, you can add B<--freeze-blank-lines>.
525 =item B<-ole=s>, B<--output-line-ending=s>
527 where s=C<win>, C<dos>, C<unix>, or C<mac>. This flag tells perltidy
528 to output line endings for a specific system. Normally,
529 perltidy writes files with the line separator character of the host
530 system. The C<win> and C<dos> flags have an identical result.
532 =item B<-ple>, B<--preserve-line-endings>
534 This flag tells perltidy to write its output files with the same line
535 endings as the input file, if possible. It should work for
536 B<dos>, B<unix>, and B<mac> line endings. It will only work if perltidy
537 input comes from a filename (rather than stdin, for example). If
538 perltidy has trouble determining the input file line ending, it will
539 revert to the default behavior of using the line ending of the host system.
541 =item B<-it=n>, B<--iterations=n>
543 This flag causes perltidy to do B<n> complete iterations. The reason for this
544 flag is that code beautification is an iterative process and in some
545 cases the output from perltidy can be different if it is applied a second time.
546 For most purposes the default of B<n=1> should be satisfactory. However B<n=2>
547 can be useful when a major style change is being made, or when code is being
548 beautified on check-in to a source code control system. It has been found to
549 be extremely rare for the output to change after 2 iterations. If a value
550 B<n> is greater than 2 is input then a convergence test will be used to stop
551 the iterations as soon as possible, almost always after 2 iterations. See
552 the next item for a simplified iteration control.
554 This flag has no effect when perltidy is used to generate html.
556 =item B<-conv>, B<--converge>
558 This flag is equivalent to B<-it=4> and is included to simplify iteration
559 control. For all practical purposes one either does or does not want to be
560 sure that the output is converged, and there is no penalty to using a large
561 iteration limit since perltidy will check for convergence and stop iterating as
562 soon as possible. The default is B<-nconv> (no convergence check). Using
563 B<-conv> will approximately double run time since normally one extra iteration
564 is required to verify convergence.
568 =head2 Code Indentation Control
572 =item B<-ci=n>, B<--continuation-indentation=n>
574 Continuation indentation is extra indentation spaces applied when
575 a long line is broken. The default is n=2, illustrated here:
578 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
580 The same example, with n=0, is a little harder to read:
583 ( $max_index_to_go >= 0 ) ? $levels_to_go[0] : $last_output_level;
585 The value given to B<-ci> is also used by some commands when a small
586 space is required. Examples are commands for outdenting labels,
587 B<-ola>, and control keywords, B<-okw>.
589 When default values are not used, it is suggested that the value B<n>
590 given with B<-ci=n> be no more than about one-half of the number of
591 spaces assigned to a full indentation level on the B<-i=n> command.
593 =item B<-sil=n> B<--starting-indentation-level=n>
595 By default, perltidy examines the input file and tries to determine the
596 starting indentation level. While it is often zero, it may not be
597 zero for a code snippet being sent from an editing session.
599 To guess the starting indentation level perltidy simply assumes that
600 indentation scheme used to create the code snippet is the same as is being used
601 for the current perltidy process. This is the only sensible guess that can be
602 made. It should be correct if this is true, but otherwise it probably won't.
603 For example, if the input script was written with -i=2 and the current peltidy
604 flags have -i=4, the wrong initial indentation will be guessed for a code
605 snippet which has non-zero initial indentation. Likewise, if an entabbing
606 scheme is used in the input script and not in the current process then the
607 guessed indentation will be wrong.
609 If the default method does not work correctly, or you want to change the
610 starting level, use B<-sil=n>, to force the starting level to be n.
612 =item List indentation using B<-lp>, B<--line-up-parentheses>
614 By default, perltidy indents lists with 4 spaces, or whatever value
615 is specified with B<-i=n>. Here is a small list formatted in this way:
619 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
620 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
623 Use the B<-lp> flag to add extra indentation to cause the data to begin
624 past the opening parentheses of a sub call or list, or opening square
625 bracket of an anonymous array, or opening curly brace of an anonymous
626 hash. With this option, the above list would become:
630 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
631 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
634 If the available line length (see B<-l=n> ) does not permit this much
635 space, perltidy will use less. For alternate placement of the
636 closing paren, see the next section.
638 This option has no effect on code BLOCKS, such as if/then/else blocks,
639 which always use whatever is specified with B<-i=n>. Also, the
640 existence of line breaks and/or block comments between the opening and
641 closing parens may cause perltidy to temporarily revert to its default
644 Note: The B<-lp> option may not be used together with the B<-t> tabs option.
645 It may, however, be used with the B<-et=n> tab method.
647 In addition, any parameter which significantly restricts the ability of
648 perltidy to choose newlines will conflict with B<-lp> and will cause
649 B<-lp> to be deactivated. These include B<-io>, B<-fnl>, B<-nanl>, and
650 B<-ndnl>. The reason is that the B<-lp> indentation style can require
651 the careful coordination of an arbitrary number of break points in
652 hierarchical lists, and these flags may prevent that.
654 =item B<-cti=n>, B<--closing-token-indentation>
656 The B<-cti=n> flag controls the indentation of a line beginning with
657 a C<)>, C<]>, or a non-block C<}>. Such a line receives:
659 -cti = 0 no extra indentation (default)
660 -cti = 1 extra indentation such that the closing token
661 aligns with its opening token.
662 -cti = 2 one extra indentation level if the line looks like:
664 -cti = 3 one extra indentation level always
666 The flags B<-cti=1> and B<-cti=2> work well with the B<-lp> flag (previous
669 # perltidy -lp -cti=1
671 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
672 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
675 # perltidy -lp -cti=2
677 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
678 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'
681 These flags are merely hints to the formatter and they may not always be
682 followed. In particular, if -lp is not being used, the indentation for
683 B<cti=1> is constrained to be no more than one indentation level.
685 If desired, this control can be applied independently to each of the
686 closing container token types. In fact, B<-cti=n> is merely an
687 abbreviation for B<-cpi=n -csbi=n -cbi=n>, where:
688 B<-cpi> or B<--closing-paren-indentation> controls B<)>'s,
689 B<-csbi> or B<--closing-square-bracket-indentation> controls B<]>'s,
690 B<-cbi> or B<--closing-brace-indentation> controls non-block B<}>'s.
692 =item B<-icp>, B<--indent-closing-paren>
694 The B<-icp> flag is equivalent to
695 B<-cti=2>, described in the previous section. The B<-nicp> flag is
696 equivalent B<-cti=0>. They are included for backwards compatibility.
698 =item B<-icb>, B<--indent-closing-brace>
700 The B<-icb> option gives one extra level of indentation to a brace which
701 terminates a code block . For example,
710 The default is not to do this, indicated by B<-nicb>.
712 =item B<-olq>, B<--outdent-long-quotes>
714 When B<-olq> is set, lines which is a quoted string longer than the
715 value B<maximum-line-length> will have their indentation removed to make
716 them more readable. This is the default. To prevent such out-denting,
717 use B<-nolq> or B<--nooutdent-long-lines>.
719 =item B<-oll>, B<--outdent-long-lines>
721 This command is equivalent to B<--outdent-long-quotes> and
722 B<--outdent-long-comments>, and it is included for compatibility with previous
723 versions of perltidy. The negation of this also works, B<-noll> or
724 B<--nooutdent-long-lines>, and is equivalent to setting B<-nolq> and B<-nolc>.
726 =item Outdenting Labels: B<-ola>, B<--outdent-labels>
728 This command will cause labels to be outdented by 2 spaces (or whatever B<-ci>
729 has been set to), if possible. This is the default. For example:
732 LOOP: while ( $i = <FOTOS> ) {
738 Use B<-nola> to not outdent labels.
740 =item Outdenting Keywords
744 =item B<-okw>, B<--outdent-keywords>
746 The command B<-okw> will cause certain leading control keywords to
747 be outdented by 2 spaces (or whatever B<-ci> has been set to), if
748 possible. By default, these keywords are C<redo>, C<next>, C<last>,
749 C<goto>, and C<return>. The intention is to make these control keywords
750 easier to see. To change this list of keywords being outdented, see
753 For example, using C<perltidy -okw> on the previous example gives:
756 LOOP: while ( $i = <FOTOS> ) {
762 The default is not to do this.
764 =item Specifying Outdented Keywords: B<-okwl=string>, B<--outdent-keyword-list=string>
766 This command can be used to change the keywords which are outdented with
767 the B<-okw> command. The parameter B<string> is a required list of perl
768 keywords, which should be placed in quotes if there are more than one.
769 By itself, it does not cause any outdenting to occur, so the B<-okw>
770 command is still required.
772 For example, the commands C<-okwl="next last redo goto" -okw> will cause
773 those four keywords to be outdented. It is probably simplest to place
774 any B<-okwl> command in a F<.perltidyrc> file.
780 =head2 Whitespace Control
782 Whitespace refers to the blank space between variables, operators,
783 and other code tokens.
787 =item B<-fws>, B<--freeze-whitespace>
789 This flag causes your original whitespace to remain unchanged, and
790 causes the rest of the whitespace commands in this section, the
791 Code Indentation section, and
792 the Comment Control section to be ignored.
794 =item Tightness of curly braces, parentheses, and square brackets.
796 Here the term "tightness" will mean the closeness with which
797 pairs of enclosing tokens, such as parentheses, contain the quantities
798 within. A numerical value of 0, 1, or 2 defines the tightness, with
799 0 being least tight and 2 being most tight. Spaces within containers
800 are always symmetric, so if there is a space after a C<(> then there
801 will be a space before the corresponding C<)>.
803 The B<-pt=n> or B<--paren-tightness=n> parameter controls the space within
804 parens. The example below shows the effect of the three possible
807 if ( ( my $len_tab = length( $tabstr ) ) > 0 ) { # -pt=0
808 if ( ( my $len_tab = length($tabstr) ) > 0 ) { # -pt=1 (default)
809 if ((my $len_tab = length($tabstr)) > 0) { # -pt=2
811 When n is 0, there is always a space to the right of a '(' and to the left
812 of a ')'. For n=2 there is never a space. For n=1, the default, there
813 is a space unless the quantity within the parens is a single token, such
814 as an identifier or quoted string.
816 Likewise, the parameter B<-sbt=n> or B<--square-bracket-tightness=n>
817 controls the space within square brackets, as illustrated below.
819 $width = $col[ $j + $k ] - $col[ $j ]; # -sbt=0
820 $width = $col[ $j + $k ] - $col[$j]; # -sbt=1 (default)
821 $width = $col[$j + $k] - $col[$j]; # -sbt=2
823 Curly braces which do not contain code blocks are controlled by
824 the parameter B<-bt=n> or B<--brace-tightness=n>.
826 $obj->{ $parsed_sql->{ 'table' }[0] }; # -bt=0
827 $obj->{ $parsed_sql->{'table'}[0] }; # -bt=1 (default)
828 $obj->{$parsed_sql->{'table'}[0]}; # -bt=2
830 And finally, curly braces which contain blocks of code are controlled by the
831 parameter B<-bbt=n> or B<--block-brace-tightness=n> as illustrated in the
834 %bf = map { $_ => -M $_ } grep { /\.deb$/ } dirents '.'; # -bbt=0 (default)
835 %bf = map { $_ => -M $_ } grep {/\.deb$/} dirents '.'; # -bbt=1
836 %bf = map {$_ => -M $_} grep {/\.deb$/} dirents '.'; # -bbt=2
838 To simplify input in the case that all of the tightness flags have the same
839 value <n>, the parameter <-act=n> or B<--all-containers-tightness=n> is an
840 abbreviation for the combination <-pt=n -sbt=n -bt=n -bbt=n>.
843 =item B<-tso>, B<--tight-secret-operators>
845 The flag B<-tso> causes certain perl token sequences (secret operators)
846 which might be considered to be a single operator to be formatted "tightly"
847 (without spaces). The operators currently modified by this flag are:
849 0+ +0 ()x!! ~~<> ,=> =( )=
851 For example the sequence B<0 +>, which converts a string to a number,
852 would be formatted without a space: B<0+> when the B<-tso> flag is set. This
853 flag is off by default.
855 =item B<-sts>, B<--space-terminal-semicolon>
857 Some programmers prefer a space before all terminal semicolons. The
858 default is for no such space, and is indicated with B<-nsts> or
859 B<--nospace-terminal-semicolon>.
862 $i = 1; # -nsts (default)
864 =item B<-sfs>, B<--space-for-semicolon>
866 Semicolons within B<for> loops may sometimes be hard to see,
867 particularly when commas are also present. This option places spaces on
868 both sides of these special semicolons, and is the default. Use
869 B<-nsfs> or B<--nospace-for-semicolon> to deactivate it.
871 for ( @a = @$ap, $u = shift @a ; @a ; $u = $v ) { # -sfs (default)
872 for ( @a = @$ap, $u = shift @a; @a; $u = $v ) { # -nsfs
874 =item B<-asc>, B<--add-semicolons>
876 Setting B<-asc> allows perltidy to add any missing optional semicolon at the end
877 of a line which is followed by a closing curly brace on the next line. This
878 is the default, and may be deactivated with B<-nasc> or B<--noadd-semicolons>.
880 =item B<-dsm>, B<--delete-semicolons>
882 Setting B<-dsm> allows perltidy to delete extra semicolons which are
883 simply empty statements. This is the default, and may be deactivated
884 with B<-ndsm> or B<--nodelete-semicolons>. (Such semicolons are not
885 deleted, however, if they would promote a side comment to a block
888 =item B<-aws>, B<--add-whitespace>
890 Setting this option allows perltidy to add certain whitespace improve
891 code readability. This is the default. If you do not want any
892 whitespace added, but are willing to have some whitespace deleted, use
893 B<-naws>. (Use B<-fws> to leave whitespace completely unchanged).
895 =item B<-dws>, B<--delete-old-whitespace>
897 Setting this option allows perltidy to remove some old whitespace
898 between characters, if necessary. This is the default. If you
899 do not want any old whitespace removed, use B<-ndws> or
900 B<--nodelete-old-whitespace>.
902 =item Detailed whitespace controls around tokens
904 For those who want more detailed control over the whitespace around
905 tokens, there are four parameters which can directly modify the default
906 whitespace rules built into perltidy for any token. They are:
908 B<-wls=s> or B<--want-left-space=s>,
910 B<-nwls=s> or B<--nowant-left-space=s>,
912 B<-wrs=s> or B<--want-right-space=s>,
914 B<-nwrs=s> or B<--nowant-right-space=s>.
916 These parameters are each followed by a quoted string, B<s>, containing a
917 list of token types. No more than one of each of these parameters
918 should be specified, because repeating a command-line parameter
919 always overwrites the previous one before perltidy ever sees it.
921 To illustrate how these are used, suppose it is desired that there be no
922 space on either side of the token types B<= + - / *>. The following two
923 parameters would specify this desire:
925 -nwls="= + - / *" -nwrs="= + - / *"
927 (Note that the token types are in quotes, and that they are separated by
928 spaces). With these modified whitespace rules, the following line of math:
930 $root = -$b + sqrt( $b * $b - 4. * $a * $c ) / ( 2. * $a );
934 $root=-$b+sqrt( $b*$b-4.*$a*$c )/( 2.*$a );
936 These parameters should be considered to be hints to perltidy rather
937 than fixed rules, because perltidy must try to resolve conflicts that
938 arise between them and all of the other rules that it uses. One
939 conflict that can arise is if, between two tokens, the left token wants
940 a space and the right one doesn't. In this case, the token not wanting
941 a space takes priority.
943 It is necessary to have a list of all token types in order to create
944 this type of input. Such a list can be obtained by the command
945 B<--dump-token-types>. Also try the B<-D> flag on a short snippet of code
946 and look at the .DEBUG file to see the tokenization.
948 B<WARNING> Be sure to put these tokens in quotes to avoid having them
949 misinterpreted by your command shell.
951 =item Space between specific keywords and opening paren
953 When an opening paren follows a Perl keyword, no space is introduced after the
954 keyword, unless it is (by default) one of these:
956 my local our and or eq ne if else elsif until unless
957 while for foreach return switch case given when
959 These defaults can be modified with two commands:
961 B<-sak=s> or B<--space-after-keyword=s> adds keywords.
963 B<-nsak=s> or B<--nospace-after-keyword=s> removes keywords.
965 where B<s> is a list of keywords (in quotes if necessary). For example,
967 my ( $a, $b, $c ) = @_; # default
968 my( $a, $b, $c ) = @_; # -nsak="my local our"
970 The abbreviation B<-nsak='*'> is equivalent to including all of the
971 keywords in the above list.
973 When both B<-nsak=s> and B<-sak=s> commands are included, the B<-nsak=s>
974 command is executed first. For example, to have space after only the
975 keywords (my, local, our) you could use B<-nsak="*" -sak="my local our">.
977 To put a space after all keywords, see the next item.
979 =item Space between all keywords and opening parens
981 When an opening paren follows a function or keyword, no space is introduced
982 after the keyword except for the keywords noted in the previous item. To
983 always put a space between a function or keyword and its opening paren,
986 B<-skp> or B<--space-keyword-paren>
988 You will probably also want to use the flag B<-sfp> (next item) too.
990 =item Space between all function names and opening parens
992 When an opening paren follows a function the default is not to introduce
993 a space. To cause a space to be introduced use:
995 B<-sfp> or B<--space-function-paren>
997 myfunc( $a, $b, $c ); # default
998 myfunc ( $a, $b, $c ); # -sfp
1000 You will probably also want to use the flag B<-skp> (previous item) too.
1002 =item Trimming whitespace around C<qw> quotes
1004 B<-tqw> or B<--trim-qw> provide the default behavior of trimming
1005 spaces around multi-line C<qw> quotes and indenting them appropriately.
1007 B<-ntqw> or B<--notrim-qw> cause leading and trailing whitespace around
1008 multi-line C<qw> quotes to be left unchanged. This option will not
1009 normally be necessary, but was added for testing purposes, because in
1010 some versions of perl, trimming C<qw> quotes changes the syntax tree.
1012 =item Trimming trailing whitespace from lines of POD
1014 B<-trp> or B<--trim-pod> will remove trailing whitespace from lines of POD.
1015 The default is not to do this.
1019 =head2 Comment Controls
1021 Perltidy has a number of ways to control the appearance of both block comments
1022 and side comments. The term B<block comment> here refers to a full-line
1023 comment, whereas B<side comment> will refer to a comment which appears on a
1024 line to the right of some code.
1028 =item B<-ibc>, B<--indent-block-comments>
1030 Block comments normally look best when they are indented to the same
1031 level as the code which follows them. This is the default behavior, but
1032 you may use B<-nibc> to keep block comments left-justified. Here is an
1035 # this comment is indented (-ibc, default)
1036 if ($task) { yyy(); }
1038 The alternative is B<-nibc>:
1040 # this comment is not indented (-nibc)
1041 if ($task) { yyy(); }
1043 See also the next item, B<-isbc>, as well as B<-sbc>, for other ways to
1044 have some indented and some outdented block comments.
1046 =item B<-isbc>, B<--indent-spaced-block-comments>
1048 If there is no leading space on the line, then the comment will not be
1049 indented, and otherwise it may be.
1051 If both B<-ibc> and B<-isbc> are set, then B<-isbc> takes priority.
1053 =item B<-olc>, B<--outdent-long-comments>
1055 When B<-olc> is set, lines which are full-line (block) comments longer
1056 than the value B<maximum-line-length> will have their indentation
1057 removed. This is the default; use B<-nolc> to prevent outdenting.
1059 =item B<-msc=n>, B<--minimum-space-to-comment=n>
1061 Side comments look best when lined up several spaces to the right of
1062 code. Perltidy will try to keep comments at least n spaces to the
1063 right. The default is n=4 spaces.
1065 =item B<-fpsc=n>, B<--fixed-position-side-comment=n>
1067 This parameter tells perltidy to line up side comments in column number B<n>
1068 whenever possible. The default, n=0, will not do this.
1070 =item B<-iscl>, B<--ignore-side-comment-lengths>
1072 This parameter causes perltidy to ignore the length of side comments when
1073 setting line breaks. The default, B<-niscl>, is to include the length of
1074 side comments when breaking lines to stay within the length prescribed
1075 by the B<-l=n> maximum line length parameter. For example, the following
1076 long single line would remain intact with -l=80 and -iscl:
1078 perltidy -l=80 -iscl
1079 $vmsfile =~ s/;[\d\-]*$//; # Clip off version number; we can use a newer version as well
1081 whereas without the -iscl flag the line will be broken:
1084 $vmsfile =~ s/;[\d\-]*$//
1085 ; # Clip off version number; we can use a newer version as well
1088 =item B<-hsc>, B<--hanging-side-comments>
1090 By default, perltidy tries to identify and align "hanging side
1091 comments", which are something like this:
1093 my $IGNORE = 0; # This is a side comment
1094 # This is a hanging side comment
1097 A comment is considered to be a hanging side comment if (1) it immediately
1098 follows a line with a side comment, or another hanging side comment, and
1099 (2) there is some leading whitespace on the line.
1100 To deactivate this feature, use B<-nhsc> or B<--nohanging-side-comments>.
1101 If block comments are preceded by a blank line, or have no leading
1102 whitespace, they will not be mistaken as hanging side comments.
1104 =item Closing Side Comments
1106 A closing side comment is a special comment which perltidy can
1107 automatically create and place after the closing brace of a code block.
1108 They can be useful for code maintenance and debugging. The command
1109 B<-csc> (or B<--closing-side-comments>) adds or updates closing side
1110 comments. For example, here is a small code snippet
1113 if ( !defined( $_[0] ) ) {
1114 print("Hello, World\n");
1117 print( $_[0], "\n" );
1121 And here is the result of processing with C<perltidy -csc>:
1124 if ( !defined( $_[0] ) ) {
1125 print("Hello, World\n");
1128 print( $_[0], "\n" );
1130 } ## end sub message
1132 A closing side comment was added for C<sub message> in this case, but not
1133 for the C<if> and C<else> blocks, because they were below the 6 line
1134 cutoff limit for adding closing side comments. This limit may be
1135 changed with the B<-csci> command, described below.
1137 The command B<-dcsc> (or B<--delete-closing-side-comments>) reverses this
1138 process and removes these comments.
1140 Several commands are available to modify the behavior of these two basic
1141 commands, B<-csc> and B<-dcsc>:
1145 =item B<-csci=n>, or B<--closing-side-comment-interval=n>
1147 where C<n> is the minimum number of lines that a block must have in
1148 order for a closing side comment to be added. The default value is
1149 C<n=6>. To illustrate:
1151 # perltidy -csci=2 -csc
1153 if ( !defined( $_[0] ) ) {
1154 print("Hello, World\n");
1155 } ## end if ( !defined( $_[0] ))
1157 print( $_[0], "\n" );
1158 } ## end else [ if ( !defined( $_[0] ))
1159 } ## end sub message
1161 Now the C<if> and C<else> blocks are commented. However, now this has
1162 become very cluttered.
1164 =item B<-cscp=string>, or B<--closing-side-comment-prefix=string>
1166 where string is the prefix used before the name of the block type. The
1167 default prefix, shown above, is C<## end>. This string will be added to
1168 closing side comments, and it will also be used to recognize them in
1169 order to update, delete, and format them. Any comment identified as a
1170 closing side comment will be placed just a single space to the right of
1173 =item B<-cscl=string>, or B<--closing-side-comment-list-string>
1175 where C<string> is a list of block types to be tagged with closing side
1176 comments. By default, all code block types preceded by a keyword or
1177 label (such as C<if>, C<sub>, and so on) will be tagged. The B<-cscl>
1178 command changes the default list to be any selected block types; see
1179 L<Specifying Block Types>.
1180 For example, the following command
1181 requests that only C<sub>'s, labels, C<BEGIN>, and C<END> blocks be
1182 affected by any B<-csc> or B<-dcsc> operation:
1184 -cscl="sub : BEGIN END"
1186 =item B<-csct=n>, or B<--closing-side-comment-maximum-text=n>
1188 The text appended to certain block types, such as an C<if> block, is
1189 whatever lies between the keyword introducing the block, such as C<if>,
1190 and the opening brace. Since this might be too much text for a side
1191 comment, there needs to be a limit, and that is the purpose of this
1192 parameter. The default value is C<n=20>, meaning that no additional
1193 tokens will be appended to this text after its length reaches 20
1194 characters. Omitted text is indicated with C<...>. (Tokens, including
1195 sub names, are never truncated, however, so actual lengths may exceed
1196 this). To illustrate, in the above example, the appended text of the
1197 first block is C< ( !defined( $_[0] )...>. The existing limit of
1198 C<n=20> caused this text to be truncated, as indicated by the C<...>. See
1199 the next flag for additional control of the abbreviated text.
1201 =item B<-cscb>, or B<--closing-side-comments-balanced>
1203 As discussed in the previous item, when the
1204 closing-side-comment-maximum-text limit is exceeded the comment text must
1205 be truncated. Older versions of perltidy terminated with three dots, and this
1206 can still be achieved with -ncscb:
1208 perltidy -csc -ncscb
1209 } ## end foreach my $foo (sort { $b cmp $a ...
1211 However this causes a problem with editors which cannot recognize
1212 comments or are not configured to do so because they cannot "bounce" around in
1213 the text correctly. The B<-cscb> flag has been added to
1214 help them by appending appropriate balancing structure:
1217 } ## end foreach my $foo (sort { $b cmp $a ... })
1219 The default is B<-cscb>.
1221 =item B<-csce=n>, or B<--closing-side-comment-else-flag=n>
1223 The default, B<n=0>, places the text of the opening C<if> statement after any
1226 If B<n=2> is used, then each C<elsif> is also given the text of the opening
1227 C<if> statement. Also, an C<else> will include the text of a preceding
1228 C<elsif> statement. Note that this may result some long closing
1231 If B<n=1> is used, the results will be the same as B<n=2> whenever the
1232 resulting line length is less than the maximum allowed.
1233 =item B<-cscb>, or B<--closing-side-comments-balanced>
1235 When using closing-side-comments, and the closing-side-comment-maximum-text
1236 limit is exceeded, then the comment text must be abbreviated.
1237 It is terminated with three dots if the B<-cscb> flag is negated:
1239 perltidy -csc -ncscb
1240 } ## end foreach my $foo (sort { $b cmp $a ...
1242 This causes a problem with older editors which do not recognize comments
1243 because they cannot "bounce" around in the text correctly. The B<-cscb>
1244 flag tries to help them by appending appropriate terminal balancing structures:
1247 } ## end foreach my $foo (sort { $b cmp $a ... })
1249 The default is B<-cscb>.
1252 =item B<-cscw>, or B<--closing-side-comment-warnings>
1254 This parameter is intended to help make the initial transition to the use of
1255 closing side comments.
1257 things to happen if a closing side comment replaces an existing, different
1258 closing side comment: first, an error message will be issued, and second, the
1259 original side comment will be placed alone on a new specially marked comment
1260 line for later attention.
1262 The intent is to avoid clobbering existing hand-written side comments
1263 which happen to match the pattern of closing side comments. This flag
1264 should only be needed on the first run with B<-csc>.
1268 B<Important Notes on Closing Side Comments:>
1274 Closing side comments are only placed on lines terminated with a closing
1275 brace. Certain closing styles, such as the use of cuddled elses
1276 (B<-ce>), preclude the generation of some closing side comments.
1280 Please note that adding or deleting of closing side comments takes
1281 place only through the commands B<-csc> or B<-dcsc>. The other commands,
1282 if used, merely modify the behavior of these two commands.
1286 It is recommended that the B<-cscw> flag be used along with B<-csc> on
1287 the first use of perltidy on a given file. This will prevent loss of
1288 any existing side comment data which happens to have the csc prefix.
1292 Once you use B<-csc>, you should continue to use it so that any
1293 closing side comments remain correct as code changes. Otherwise, these
1294 comments will become incorrect as the code is updated.
1298 If you edit the closing side comments generated by perltidy, you must also
1299 change the prefix to be different from the closing side comment prefix.
1300 Otherwise, your edits will be lost when you rerun perltidy with B<-csc>. For
1301 example, you could simply change C<## end> to be C<## End>, since the test is
1302 case sensitive. You may also want to use the B<-ssc> flag to keep these
1303 modified closing side comments spaced the same as actual closing side comments.
1307 Temporarily generating closing side comments is a useful technique for
1308 exploring and/or debugging a perl script, especially one written by someone
1309 else. You can always remove them with B<-dcsc>.
1313 =item Static Block Comments
1315 Static block comments are block comments with a special leading pattern,
1316 C<##> by default, which will be treated slightly differently from other
1317 block comments. They effectively behave as if they had glue along their
1318 left and top edges, because they stick to the left edge and previous line
1319 when there is no blank spaces in those places. This option is
1320 particularly useful for controlling how commented code is displayed.
1324 =item B<-sbc>, B<--static-block-comments>
1326 When B<-sbc> is used, a block comment with a special leading pattern, C<##> by
1327 default, will be treated specially.
1329 Comments so identified are treated as follows:
1335 If there is no leading space on the line, then the comment will not
1336 be indented, and otherwise it may be,
1340 no new blank line will be
1341 inserted before such a comment, and
1345 such a comment will never become
1346 a hanging side comment.
1350 For example, assuming C<@month_of_year> is
1353 @month_of_year = ( # -sbc (default)
1354 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1358 Without this convention, the above code would become
1360 @month_of_year = ( # -nsbc
1361 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct',
1367 which is not as clear.
1368 The default is to use B<-sbc>. This may be deactivated with B<-nsbc>.
1370 =item B<-sbcp=string>, B<--static-block-comment-prefix=string>
1372 This parameter defines the prefix used to identify static block comments
1373 when the B<-sbc> parameter is set. The default prefix is C<##>,
1374 corresponding to C<-sbcp=##>. The prefix is actually part of a perl
1375 pattern used to match lines and it must either begin with C<#> or C<^#>.
1376 In the first case a prefix ^\s* will be added to match any leading
1377 whitespace, while in the second case the pattern will match only
1378 comments with no leading whitespace. For example, to
1379 identify all comments as static block comments, one would use C<-sbcp=#>.
1380 To identify all left-adjusted comments as static block comments, use C<-sbcp='^#'>.
1382 Please note that B<-sbcp> merely defines the pattern used to identify static
1383 block comments; it will not be used unless the switch B<-sbc> is set. Also,
1384 please be aware that since this string is used in a perl regular expression
1385 which identifies these comments, it must enable a valid regular expression to
1388 A pattern which can be useful is:
1392 This pattern requires a static block comment to have at least one character
1393 which is neither a # nor a space. It allows a line containing only '#'
1394 characters to be rejected as a static block comment. Such lines are often used
1395 at the start and end of header information in subroutines and should not be
1396 separated from the intervening comments, which typically begin with just a
1399 =item B<-osbc>, B<--outdent-static-block-comments>
1401 The command B<-osbc> will cause static block comments to be outdented by 2
1402 spaces (or whatever B<-ci=n> has been set to), if possible.
1406 =item Static Side Comments
1408 Static side comments are side comments with a special leading pattern.
1409 This option can be useful for controlling how commented code is displayed
1410 when it is a side comment.
1414 =item B<-ssc>, B<--static-side-comments>
1416 When B<-ssc> is used, a side comment with a static leading pattern, which is
1417 C<##> by default, will be spaced only a single space from previous
1418 character, and it will not be vertically aligned with other side comments.
1420 The default is B<-nssc>.
1422 =item B<-sscp=string>, B<--static-side-comment-prefix=string>
1424 This parameter defines the prefix used to identify static side comments
1425 when the B<-ssc> parameter is set. The default prefix is C<##>,
1426 corresponding to C<-sscp=##>.
1428 Please note that B<-sscp> merely defines the pattern used to identify
1429 static side comments; it will not be used unless the switch B<-ssc> is
1430 set. Also, note that this string is used in a perl regular expression
1431 which identifies these comments, so it must enable a valid regular
1432 expression to be formed.
1439 =head2 Skipping Selected Sections of Code
1441 Selected lines of code may be passed verbatim to the output without any
1442 formatting. This feature is enabled by default but can be disabled with
1443 the B<--noformat-skipping> or B<-nfs> flag. It should be used sparingly to
1444 avoid littering code with markers, but it might be helpful for working
1445 around occasional problems. For example it might be useful for keeping
1446 the indentation of old commented code unchanged, keeping indentation of
1447 long blocks of aligned comments unchanged, keeping certain list
1448 formatting unchanged, or working around a glitch in perltidy.
1452 =item B<-fs>, B<--format-skipping>
1454 This flag, which is enabled by default, causes any code between
1455 special beginning and ending comment markers to be passed to the
1456 output without formatting. The default beginning marker is #<<<
1457 and the default ending marker is #>>> but they
1458 may be changed (see next items below). Additional text may appear on
1459 these special comment lines provided that it is separated from the
1460 marker by at least one space. For example
1462 #<<< do not let perltidy touch this
1470 The comment markers may be placed at any location that a block comment may
1471 appear. If they do not appear to be working, use the -log flag and examine the
1472 F<.LOG> file. Use B<-nfs> to disable this feature.
1474 =item B<-fsb=string>, B<--format-skipping-begin=string>
1476 The B<-fsb=string> parameter may be used to change the beginning marker for
1477 format skipping. The default is equivalent to -fsb='#<<<'. The string that
1478 you enter must begin with a # and should be in quotes as necessary to get past
1479 the command shell of your system. It is actually the leading text of a pattern
1480 that is constructed by appending a '\s', so you must also include backslashes
1481 for characters to be taken literally rather than as patterns.
1483 Some examples show how example strings become patterns:
1485 -fsb='#\{\{\{' becomes /^#\{\{\{\s/ which matches #{{{ but not #{{{{
1486 -fsb='#\*\*' becomes /^#\*\*\s/ which matches #** but not #***
1487 -fsb='#\*{2,}' becomes /^#\*{2,}\s/ which matches #** and #*****
1489 =item B<-fse=string>, B<--format-skipping-end=string>
1491 The B<-fsb=string> is the corresponding parameter used to change the
1492 ending marker for format skipping. The default is equivalent to
1497 =head2 Line Break Control
1499 The parameters in this section control breaks after
1500 non-blank lines of code. Blank lines are controlled
1501 separately by parameters in the section L<Blank Line
1506 =item B<-fnl>, B<--freeze-newlines>
1508 If you do not want any changes to the line breaks within
1509 lines of code in your script, set
1510 B<-fnl>, and they will remain fixed, and the rest of the commands in
1511 this section and sections
1512 L<Controlling List Formatting>,
1513 L<Retaining or Ignoring Existing Line Breaks>.
1514 You may want to use B<-noll> with this.
1516 Note: If you also want to keep your blank lines exactly
1517 as they are, you can use the B<-fbl> flag which is described
1518 in the section L<Blank Line Control>.
1520 =item B<-ce>, B<--cuddled-else>
1522 Enable the "cuddled else" style, in which C<else> and C<elsif> are
1523 follow immediately after the curly brace closing the previous block.
1524 The default is not to use cuddled elses, and is indicated with the flag
1525 B<-nce> or B<--nocuddled-else>. Here is a comparison of the
1537 else { # -nce (default)
1541 =item B<-bl>, B<--opening-brace-on-new-line>
1543 Use the flag B<-bl> to place the opening brace on a new line:
1545 if ( $input_file eq '-' ) # -bl
1547 important_function();
1550 This flag applies to all structural blocks, including named sub's (unless
1551 the B<-sbl> flag is set -- see next item).
1553 The default style, B<-nbl>, places an opening brace on the same line as
1554 the keyword introducing it. For example,
1556 if ( $input_file eq '-' ) { # -nbl (default)
1558 =item B<-sbl>, B<--opening-sub-brace-on-new-line>
1560 The flag B<-sbl> can be used to override the value of B<-bl> for
1561 the opening braces of named sub's. For example,
1565 produces this result:
1569 if (!defined($_[0])) {
1570 print("Hello, World\n");
1577 This flag is negated with B<-nsbl>. If B<-sbl> is not specified,
1578 the value of B<-bl> is used.
1580 =item B<-asbl>, B<--opening-anonymous-sub-brace-on-new-line>
1582 The flag B<-asbl> is like the B<-sbl> flag except that it applies
1583 to anonymous sub's instead of named subs. For example
1587 produces this result:
1591 if ( !defined( $_[0] ) ) {
1592 print("Hello, World\n");
1595 print( $_[0], "\n" );
1599 This flag is negated with B<-nasbl>, and the default is B<-nasbl>.
1601 =item B<-bli>, B<--brace-left-and-indent>
1603 The flag B<-bli> is the same as B<-bl> but in addition it causes one
1604 unit of continuation indentation ( see B<-ci> ) to be placed before
1605 an opening and closing block braces.
1609 if ( $input_file eq '-' ) # -bli
1611 important_function();
1614 By default, this extra indentation occurs for blocks of type:
1615 B<if>, B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>,
1616 B<while>, B<until>, and also with a preceding label. The next item
1617 shows how to change this.
1619 =item B<-blil=s>, B<--brace-left-and-indent-list=s>
1621 Use this parameter to change the types of block braces for which the
1622 B<-bli> flag applies; see L<Specifying Block Types>. For example,
1623 B<-blil='if elsif else'> would apply it to only C<if/elsif/else> blocks.
1625 =item B<-bar>, B<--opening-brace-always-on-right>
1627 The default style, B<-nbl> places the opening code block brace on a new
1628 line if it does not fit on the same line as the opening keyword, like
1631 if ( $bigwasteofspace1 && $bigwasteofspace2
1632 || $bigwasteofspace3 && $bigwasteofspace4 )
1634 big_waste_of_time();
1637 To force the opening brace to always be on the right, use the B<-bar>
1638 flag. In this case, the above example becomes
1640 if ( $bigwasteofspace1 && $bigwasteofspace2
1641 || $bigwasteofspace3 && $bigwasteofspace4 ) {
1642 big_waste_of_time();
1645 A conflict occurs if both B<-bl> and B<-bar> are specified.
1647 =item B<-otr>, B<--opening-token-right> and related flags
1649 The B<-otr> flag is a hint that perltidy should not place a break between a
1650 comma and an opening token. For example:
1652 # default formatting
1653 push @{ $self->{$module}{$key} },
1655 accno => $ref->{accno},
1656 description => $ref->{description}
1660 push @{ $self->{$module}{$key} }, {
1661 accno => $ref->{accno},
1662 description => $ref->{description}
1665 The flag B<-otr> is actually an abbreviation for three other flags
1666 which can be used to control parens, hash braces, and square brackets
1667 separately if desired:
1669 -opr or --opening-paren-right
1670 -ohbr or --opening-hash-brace-right
1671 -osbr or --opening-square-bracket-right
1673 =item Vertical tightness of non-block curly braces, parentheses, and square brackets.
1675 These parameters control what shall be called vertical tightness. Here are the
1682 Opening tokens (except for block braces) are controlled by B<-vt=n>, or
1683 B<--vertical-tightness=n>, where
1685 -vt=0 always break a line after opening token (default).
1686 -vt=1 do not break unless this would produce more than one
1687 step in indentation in a line.
1688 -vt=2 never break a line after opening token
1692 You must also use the B<-lp> flag when you use the B<-vt> flag; the
1693 reason is explained below.
1697 Closing tokens (except for block braces) are controlled by B<-vtc=n>, or
1698 B<--vertical-tightness-closing=n>, where
1700 -vtc=0 always break a line before a closing token (default),
1701 -vtc=1 do not break before a closing token which is followed
1702 by a semicolon or another closing token, and is not in
1704 -vtc=2 never break before a closing token.
1706 The rules for B<-vtc=1> are designed to maintain a reasonable balance
1707 between tightness and readability in complex lists.
1711 Different controls may be applied to different token types,
1712 and it is also possible to control block braces; see below.
1716 Finally, please note that these vertical tightness flags are merely
1717 hints to the formatter, and it cannot always follow them. Things which
1718 make it difficult or impossible include comments, blank lines, blocks of
1719 code within a list, and possibly the lack of the B<-lp> parameter.
1720 Also, these flags may be ignored for very small lists (2 or 3 lines in
1725 Here are some examples:
1727 # perltidy -lp -vt=0 -vtc=0
1735 # perltidy -lp -vt=1 -vtc=0
1736 %romanNumerals = ( one => 'I',
1742 # perltidy -lp -vt=1 -vtc=1
1743 %romanNumerals = ( one => 'I',
1748 The difference between B<-vt=1> and B<-vt=2> is shown here:
1750 # perltidy -lp -vt=1
1752 mysprintf( "(void)find_threadsv(%s);",
1753 cstring( $threadsv_names[ $op->targ ] )
1757 # perltidy -lp -vt=2
1758 $init->add( mysprintf( "(void)find_threadsv(%s);",
1759 cstring( $threadsv_names[ $op->targ ] )
1763 With B<-vt=1>, the line ending in C<add(> does not combine with the next
1764 line because the next line is not balanced. This can help with
1765 readability, but B<-vt=2> can be used to ignore this rule.
1767 The tightest, and least readable, code is produced with both C<-vt=2> and
1770 # perltidy -lp -vt=2 -vtc=2
1771 $init->add( mysprintf( "(void)find_threadsv(%s);",
1772 cstring( $threadsv_names[ $op->targ ] ) ) );
1774 Notice how the code in all of these examples collapses vertically as
1775 B<-vt> increases, but the indentation remains unchanged. This is
1776 because perltidy implements the B<-vt> parameter by first formatting as
1777 if B<-vt=0>, and then simply overwriting one output line on top of the
1778 next, if possible, to achieve the desired vertical tightness. The
1779 B<-lp> indentation style has been designed to allow this vertical
1780 collapse to occur, which is why it is required for the B<-vt> parameter.
1782 The B<-vt=n> and B<-vtc=n> parameters apply to each type of container
1783 token. If desired, vertical tightness controls can be applied
1784 independently to each of the closing container token types.
1786 The parameters for controlling parentheses are B<-pvt=n> or
1787 B<--paren-vertical-tightness=n>, and B<-pcvt=n> or
1788 B<--paren-vertical-tightness-closing=n>.
1790 Likewise, the parameters for square brackets are B<-sbvt=n> or
1791 B<--square-bracket-vertical-tightness=n>, and B<-sbcvt=n> or
1792 B<--square-bracket-vertical-tightness-closing=n>.
1794 Finally, the parameters for controlling non-code block braces are
1795 B<-bvt=n> or B<--brace-vertical-tightness=n>, and B<-bcvt=n> or
1796 B<--brace-vertical-tightness-closing=n>.
1798 In fact, the parameter B<-vt=n> is actually just an abbreviation for
1799 B<-pvt=n -bvt=n sbvt=n>, and likewise B<-vtc=n> is an abbreviation
1800 for B<-pvtc=n -bvtc=n sbvtc=n>.
1802 =item B<-bbvt=n> or B<--block-brace-vertical-tightness=n>
1804 The B<-bbvt=n> flag is just like the B<-vt=n> flag but applies
1805 to opening code block braces.
1807 -bbvt=0 break after opening block brace (default).
1808 -bbvt=1 do not break unless this would produce more than one
1809 step in indentation in a line.
1810 -bbvt=2 do not break after opening block brace.
1812 It is necessary to also use either B<-bl> or B<-bli> for this to work,
1813 because, as with other vertical tightness controls, it is implemented by
1814 simply overwriting a line ending with an opening block brace with the
1815 subsequent line. For example:
1817 # perltidy -bli -bbvt=0
1818 if ( open( FILE, "< $File" ) )
1820 while ( $File = <FILE> )
1828 # perltidy -bli -bbvt=1
1829 if ( open( FILE, "< $File" ) )
1830 { while ( $File = <FILE> )
1837 By default this applies to blocks associated with keywords B<if>,
1838 B<elsif>, B<else>, B<unless>, B<for>, B<foreach>, B<sub>, B<while>,
1839 B<until>, and also with a preceding label. This can be changed with
1840 the parameter B<-bbvtl=string>, or
1841 B<--block-brace-vertical-tightness-list=string>, where B<string> is a
1842 space-separated list of block types. For more information on the
1843 possible values of this string, see L<Specifying Block Types>
1845 For example, if we want to just apply this style to C<if>,
1846 C<elsif>, and C<else> blocks, we could use
1847 C<perltidy -bli -bbvt=1 -bbvtl='if elsif else'>.
1849 There is no vertical tightness control for closing block braces; with
1850 one exception they will be placed on separate lines.
1851 The exception is that a cascade of closing block braces may
1852 be stacked on a single line. See B<-scbb>.
1854 =item B<-sot>, B<--stack-opening-tokens> and related flags
1856 The B<-sot> flag tells perltidy to "stack" opening tokens
1857 when possible to avoid lines with isolated opening tokens.
1862 $opt_c = Text::CSV_XS->new(
1871 $opt_c = Text::CSV_XS->new( {
1878 For detailed control of individual closing tokens the following
1879 controls can be used:
1881 -sop or --stack-opening-paren
1882 -sohb or --stack-opening-hash-brace
1883 -sosb or --stack-opening-square-bracket
1884 -sobb or --stack-opening-block-brace
1886 The flag B<-sot> is an abbreviation for B<-sop -sohb -sosb>.
1888 The flag B<-sobb> is a abbreviation for B<-bbvt=2 -bbvtl='*'>. This
1889 will case a cascade of opening block braces to appear on a single line,
1890 although this an uncommon occurrence except in test scripts.
1892 =item B<-sct>, B<--stack-closing-tokens> and related flags
1894 The B<-sct> flag tells perltidy to "stack" closing tokens
1895 when possible to avoid lines with isolated closing tokens.
1900 $opt_c = Text::CSV_XS->new(
1909 $opt_c = Text::CSV_XS->new(
1916 The B<-sct> flag is somewhat similar to the B<-vtc> flags, and in some
1917 cases it can give a similar result. The difference is that the B<-vtc>
1918 flags try to avoid lines with leading opening tokens by "hiding" them at
1919 the end of a previous line, whereas the B<-sct> flag merely tries to
1920 reduce the number of lines with isolated closing tokens by stacking them
1921 but does not try to hide them. For example:
1924 $opt_c = Text::CSV_XS->new(
1928 always_quote => 1, } );
1930 For detailed control of the stacking of individual closing tokens the
1931 following controls can be used:
1933 -scp or --stack-closing-paren
1934 -schb or --stack-closing-hash-brace
1935 -scsb or --stack-closing-square-bracket
1936 -scbb or --stack-closing-block-brace
1938 The flag B<-sct> is an abbreviation for stacking the non-block closing
1939 tokens, B<-scp -schb -scsb>.
1941 Stacking of closing block braces, B<-scbb>, causes a cascade of isolated
1942 closing block braces to be combined into a single line as in the following
1950 push( @lines, "$w1 $w2 $w3 $w4\n" );
1953 To simplify input even further for the case in which both opening and closing
1954 non-block containers are stacked, the flag B<-sac> or B<--stack-all-containers>
1955 is an abbreviation for B<-sot -sot>.
1957 =item B<-dnl>, B<--delete-old-newlines>
1959 By default, perltidy first deletes all old line break locations, and then it
1960 looks for good break points to match the desired line length. Use B<-ndnl>
1961 or B<--nodelete-old-newlines> to force perltidy to retain all old line break
1964 =item B<-anl>, B<--add-newlines>
1966 By default, perltidy will add line breaks when necessary to create
1967 continuations of long lines and to improve the script appearance. Use
1968 B<-nanl> or B<--noadd-newlines> to prevent any new line breaks.
1970 This flag does not prevent perltidy from eliminating existing line
1971 breaks; see B<--freeze-newlines> to completely prevent changes to line
1974 =item Controlling whether perltidy breaks before or after operators
1976 Four command line parameters provide some control over whether
1977 a line break should be before or after specific token types.
1978 Two parameters give detailed control:
1980 B<-wba=s> or B<--want-break-after=s>, and
1982 B<-wbb=s> or B<--want-break-before=s>.
1984 These parameters are each followed by a quoted string, B<s>, containing
1985 a list of token types (separated only by spaces). No more than one of each
1986 of these parameters should be specified, because repeating a
1987 command-line parameter always overwrites the previous one before
1988 perltidy ever sees it.
1990 By default, perltidy breaks B<after> these token types:
1991 % + - * / x != == >= <= =~ !~ < > | &
1992 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
1994 And perltidy breaks B<before> these token types by default:
1997 To illustrate, to cause a break after a concatenation operator, C<'.'>,
1998 rather than before it, the command line would be
2002 As another example, the following command would cause a break before
2003 math operators C<'+'>, C<'-'>, C<'/'>, and C<'*'>:
2007 These commands should work well for most of the token types that perltidy uses
2008 (use B<--dump-token-types> for a list). Also try the B<-D> flag on a short
2009 snippet of code and look at the .DEBUG file to see the tokenization. However,
2010 for a few token types there may be conflicts with hardwired logic which cause
2011 unexpected results. One example is curly braces, which should be controlled
2012 with the parameter B<bl> provided for that purpose.
2014 B<WARNING> Be sure to put these tokens in quotes to avoid having them
2015 misinterpreted by your command shell.
2017 Two additional parameters are available which, though they provide no further
2018 capability, can simplify input are:
2020 B<-baao> or B<--break-after-all-operators>,
2022 B<-bbao> or B<--break-before-all-operators>.
2024 The -baao sets the default to be to break after all of the following operators:
2026 % + - * / x != == >= <= =~ !~ < > | &
2027 = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
2028 . : ? && || and or err xor
2030 and the B<-bbao> flag sets the default to break before all of these operators.
2031 These can be used to define an initial break preference which can be fine-tuned
2032 with the B<-wba> and B<-wbb> flags. For example, to break before all operators
2033 except an B<=> one could use --bbao -wba='=' rather than listing every
2034 single perl operator except B<=> on a -wbb flag.
2038 =head2 Controlling List Formatting
2040 Perltidy attempts to place comma-separated arrays of values in tables
2041 which look good. Its default algorithms usually work well, and they
2042 have been improving with each release, but several parameters are
2043 available to control list formatting.
2047 =item B<-boc>, B<--break-at-old-comma-breakpoints>
2049 This flag tells perltidy to try to break at all old commas. This is not
2050 the default. Normally, perltidy makes a best guess at list formatting,
2051 and seldom uses old comma breakpoints. Usually this works well,
2060 The default formatting will flatten this down to one line:
2062 # perltidy (default)
2063 my @list = ( 1, 1, 1, 1, 2, 1, 1, 3, 3, 1, 1, 4, 6, 4, 1, );
2065 which hides the structure. Using B<-boc>, plus additional flags
2066 to retain the original style, yields
2068 # perltidy -boc -lp -pt=2 -vt=1 -vtc=1
2075 A disadvantage of this flag is that all tables in the file
2076 must already be nicely formatted. For another possibility see
2077 the -fs flag in L<Skipping Selected Sections of Code>.
2079 =item B<-mft=n>, B<--maximum-fields-per-table=n>
2081 If the computed number of fields for any table exceeds B<n>, then it
2082 will be reduced to B<n>. The default value for B<n> is a large number,
2083 40. While this value should probably be left unchanged as a general
2084 rule, it might be used on a small section of code to force a list to
2085 have a particular number of fields per line, and then either the B<-boc>
2086 flag could be used to retain this formatting, or a single comment could
2087 be introduced somewhere to freeze the formatting in future applications
2100 =item B<-cab=n>, B<--comma-arrow-breakpoints=n>
2102 A comma which follows a comma arrow, '=>', is given special
2103 consideration. In a long list, it is common to break at all such
2104 commas. This parameter can be used to control how perltidy breaks at
2105 these commas. (However, it will have no effect if old comma breaks are
2106 being forced because B<-boc> is used). The possible values of B<n> are:
2108 n=0 break at all commas after =>
2109 n=1 stable: break at all commas after => if container is open,
2110 EXCEPT FOR one-line containers
2111 n=2 break at all commas after =>, BUT try to form the maximum
2112 maximum one-line container lengths
2113 n=3 do not treat commas after => specially at all
2114 n=4 break everything: like n=0 but ALSO break a short container with
2115 a => not followed by a comma when -vt=0 is used
2116 n=5 stable: like n=1 but ALSO break at open one-line containers when
2117 -vt=0 is used (default)
2119 For example, given the following single line, perltidy by default will
2120 not add any line breaks because it would break the existing one-line
2123 bless { B => $B, Root => $Root } => $package;
2125 Using B<-cab=0> will force a break after each comma-arrow item:
2133 If perltidy is subsequently run with this container broken, then by
2134 default it will break after each '=>' because the container is now
2135 broken. To reform a one-line container, the parameter B<-cab=2> could
2138 The flag B<-cab=3> can be used to prevent these commas from being
2139 treated specially. In this case, an item such as "01" => 31 is
2140 treated as a single item in a table. The number of fields in this table
2141 will be determined by the same rules that are used for any other table.
2146 "01" => 31, "02" => 29, "03" => 31, "04" => 30,
2147 "05" => 31, "06" => 30, "07" => 31, "08" => 31,
2148 "09" => 30, "10" => 31, "11" => 30, "12" => 31
2153 =head2 Retaining or Ignoring Existing Line Breaks
2155 Several additional parameters are available for controlling the extent
2156 to which line breaks in the input script influence the output script.
2157 In most cases, the default parameter values are set so that, if a choice
2158 is possible, the output style follows the input style. For example, if
2159 a short logical container is broken in the input script, then the
2160 default behavior is for it to remain broken in the output script.
2162 Most of the parameters in this section would only be required for a
2163 one-time conversion of a script from short container lengths to longer
2164 container lengths. The opposite effect, of converting long container
2165 lengths to shorter lengths, can be obtained by temporarily using a short
2166 maximum line length.
2170 =item B<-bol>, B<--break-at-old-logical-breakpoints>
2172 By default, if a logical expression is broken at a C<&&>, C<||>, C<and>,
2173 or C<or>, then the container will remain broken. Also, breaks
2174 at internal keywords C<if> and C<unless> will normally be retained.
2175 To prevent this, and thus form longer lines, use B<-nbol>.
2177 =item B<-bok>, B<--break-at-old-keyword-breakpoints>
2179 By default, perltidy will retain a breakpoint before keywords which may
2180 return lists, such as C<sort> and <map>. This allows chains of these
2181 operators to be displayed one per line. Use B<-nbok> to prevent
2182 retaining these breakpoints.
2184 =item B<-bot>, B<--break-at-old-ternary-breakpoints>
2186 By default, if a conditional (ternary) operator is broken at a C<:>,
2187 then it will remain broken. To prevent this, and thereby
2188 form longer lines, use B<-nbot>.
2190 =item B<-boa>, B<--break-at-old-attribute-breakpoints>
2192 By default, if an attribute list is broken at a C<:> in the source file, then
2193 it will remain broken. For example, given the following code, the line breaks
2194 at the ':'s will be retained:
2199 : Get('Name' => 'foo') : Set('Name');
2201 If the attributes are on a single line in the source code then they will remain
2202 on a single line if possible.
2204 To prevent this, and thereby always form longer lines, use B<-nboa>.
2206 =item B<-iob>, B<--ignore-old-breakpoints>
2208 Use this flag to tell perltidy to ignore existing line breaks to the
2209 maximum extent possible. This will tend to produce the longest possible
2210 containers, regardless of type, which do not exceed the line length
2213 =item B<-kis>, B<--keep-interior-semicolons>
2215 Use the B<-kis> flag to prevent breaking at a semicolon if
2216 there was no break there in the input file. Normally
2217 perltidy places a newline after each semicolon which
2218 terminates a statement unless several statements are
2219 contained within a one-line brace block. To illustrate,
2220 consider the following input lines:
2222 dbmclose(%verb_delim); undef %verb_delim;
2223 dbmclose(%expanded); undef %expanded;
2225 The default is to break after each statement, giving
2227 dbmclose(%verb_delim);
2229 dbmclose(%expanded);
2232 With B<perltidy -kis> the multiple statements are retained:
2234 dbmclose(%verb_delim); undef %verb_delim;
2235 dbmclose(%expanded); undef %expanded;
2237 The statements are still subject to the specified value
2238 of B<maximum-line-length> and will be broken if this
2239 maximum is exceeded.
2243 =head2 Blank Line Control
2245 Blank lines can improve the readability of a script if they are carefully
2246 placed. Perltidy has several commands for controlling the insertion,
2247 retention, and removal of blank lines.
2251 =item B<-fbl>, B<--freeze-blank-lines>
2253 Set B<-fbl> if you want to the blank lines in your script to
2254 remain exactly as they are. The rest of the parameters in
2255 this section may then be ignored. (Note: setting the B<-fbl> flag
2256 is equivalent to setting B<-mbl=0> and B<-kbl=2>).
2258 =item B<-bbc>, B<--blanks-before-comments>
2260 A blank line will be introduced before a full-line comment. This is the
2261 default. Use B<-nbbc> or B<--noblanks-before-comments> to prevent
2262 such blank lines from being introduced.
2264 =item B<-blbs=n>, B<--blank-lines-before-subs=n>
2266 The parameter B<-blbs=n> requests that least B<n> blank lines precede a sub
2267 definition which does not follow a comment and which is more than one-line
2268 long. The default is <-blbs=1>. B<BEGIN> and B<END> blocks are included.
2270 The requested number of blanks statement will be inserted regardless of the
2271 value of B<--maximum-consecutive-blank-lines=n> (B<-mbl=n>) with the exception
2272 that if B<-mbl=0> then no blanks will be output.
2274 This parameter interacts with the value B<k> of the parameter B<--maximum-consecutive-blank-lines=k> (B<-mbl=k>) as follows:
2276 1. If B<-mbl=0> then no blanks will be output. This allows all blanks to be suppressed with a single parameter. Otherwise,
2278 2. If the number of old blank lines in the script is less than B<n> then
2279 additional blanks will be inserted to make the total B<n> regardless of the
2282 3. If the number of old blank lines in the script equals or exceeds B<n> then
2283 this parameter has no effect, however the total will not exceed
2284 value specified on the B<-mbl=k> flag.
2287 =item B<-blbp=n>, B<--blank-lines-before-packages=n>
2289 The parameter B<-blbp=n> requests that least B<n> blank lines precede a package
2290 which does not follow a comment. The default is <-blbp=1>.
2292 This parameter interacts with the value B<k> of the parameter
2293 B<--maximum-consecutive-blank-lines=k> (B<-mbl=k>) in the same way as described
2294 for the previous item B<-blbs=n>.
2297 =item B<-bbs>, B<--blanks-before-subs>
2299 For compatibility with previous versions, B<-bbs> or B<--blanks-before-subs>
2300 is equivalent to F<-blbp=1> and F<-blbs=1>.
2302 Likewise, B<-nbbs> or B<--noblanks-before-subs>
2303 is equivalent to F<-blbp=0> and F<-blbs=0>.
2305 =item B<-bbb>, B<--blanks-before-blocks>
2307 A blank line will be introduced before blocks of coding delimited by
2308 B<for>, B<foreach>, B<while>, B<until>, and B<if>, B<unless>, in the following
2315 The block is not preceded by a comment.
2319 The block is not a one-line block.
2323 The number of consecutive non-blank lines at the current indentation depth is at least B<-lbl>
2328 This is the default. The intention of this option is to introduce
2329 some space within dense coding.
2330 This is negated with B<-nbbb> or B<--noblanks-before-blocks>.
2332 =item B<-lbl=n> B<--long-block-line-count=n>
2334 This controls how often perltidy is allowed to add blank lines before
2335 certain block types (see previous section). The default is 8. Entering
2336 a value of B<0> is equivalent to entering a very large number.
2338 =item B<-mbl=n> B<--maximum-consecutive-blank-lines=n>
2340 This parameter specifies the maximum number of consecutive blank lines which
2341 will be output within code sections of a script. The default is n=1. If the
2342 input file has more than n consecutive blank lines, the number will be reduced
2343 to n except as noted above for the B<-blbp> and B<-blbs> parameters. If B<n=0>
2344 then no blank lines will be output (unless all old blank lines are retained
2345 with the B<-kbl=2> flag of the next section).
2347 This flag obviously does not apply to pod sections,
2348 here-documents, and quotes.
2350 =item B<-kbl=n>, B<--keep-old-blank-lines=n>
2352 The B<-kbl=n> flag gives you control over how your existing blank lines are
2355 The possible values of B<n> are:
2357 n=0 ignore all old blank lines
2358 n=1 stable: keep old blanks, but limited by the value of the B<-mbl=n> flag
2359 n=2 keep all old blank lines, regardless of the value of the B<-mbl=n> flag
2361 The default is B<n=1>.
2363 =item B<-sob>, B<--swallow-optional-blank-lines>
2365 This is equivalent to B<kbl=0> and is included for compatibility with
2368 =item B<-nsob>, B<--noswallow-optional-blank-lines>
2370 This is equivalent to B<kbl=1> and is included for compatibility with
2377 A style refers to a convenient collection of existing parameters.
2381 =item B<-gnu>, B<--gnu-style>
2383 B<-gnu> gives an approximation to the GNU Coding Standards (which do
2384 not apply to perl) as they are sometimes implemented. At present, this
2385 style overrides the default style with the following parameters:
2387 -lp -bl -noll -pt=2 -bt=2 -sbt=2 -icp
2389 =item B<-pbp>, B<--perl-best-practices>
2391 B<-pbp> is an abbreviation for the parameters in the book B<Perl Best Practices>
2394 -l=78 -i=4 -ci=4 -st -se -vt=2 -cti=0 -pt=1 -bt=1 -sbt=1 -bbt=1 -nsfs -nolq
2395 -wbb="% + - * / x != == >= <= =~ !~ < > | & =
2396 **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x="
2398 Please note that this parameter set includes -st and -se flags, which make
2399 perltidy act as a filter on one file only. These can be overridden by placing
2400 B<-nst> and/or B<-nse> after the -pbp parameter.
2402 Also note that the value of continuation indentation, -ci=4, is equal to the
2403 value of the full indentation, -i=4. In some complex statements perltidy will
2404 produce nicer results with -ci=2. This can be implemented by including -ci=2
2405 after the -pbp parameter. For example,
2410 : $type eq 'item' ? "the $section entry"
2411 : "the section on $section"
2415 ? ( $section ? ' in ' : '' ) . "the $page$page_ext manpage"
2416 : ' elsewhere in this document'
2419 # perltidy -pbp -ci=2
2422 : $type eq 'item' ? "the $section entry"
2423 : "the section on $section"
2427 ? ( $section ? ' in ' : '' ) . "the $page$page_ext manpage"
2428 : ' elsewhere in this document'
2433 =head2 Other Controls
2437 =item Deleting selected text
2439 Perltidy can selectively delete comments and/or pod documentation. The
2440 command B<-dac> or B<--delete-all-comments> will delete all comments
2441 B<and> all pod documentation, leaving just code and any leading system
2444 The command B<-dp> or B<--delete-pod> will remove all pod documentation
2447 Two commands which remove comments (but not pod) are: B<-dbc> or
2448 B<--delete-block-comments> and B<-dsc> or B<--delete-side-comments>.
2449 (Hanging side comments will be deleted with block comments here.)
2451 The negatives of these commands also work, and are the defaults. When
2452 block comments are deleted, any leading 'hash-bang' will be retained.
2453 Also, if the B<-x> flag is used, any system commands before a leading
2454 hash-bang will be retained (even if they are in the form of comments).
2456 =item Writing selected text to a file
2458 When perltidy writes a formatted text file, it has the ability to also
2459 send selected text to a file with a F<.TEE> extension. This text can
2460 include comments and pod documentation.
2462 The command B<-tac> or B<--tee-all-comments> will write all comments
2463 B<and> all pod documentation.
2465 The command B<-tp> or B<--tee-pod> will write all pod documentation (but
2468 The commands which write comments (but not pod) are: B<-tbc> or
2469 B<--tee-block-comments> and B<-tsc> or B<--tee-side-comments>.
2470 (Hanging side comments will be written with block comments here.)
2472 The negatives of these commands also work, and are the defaults.
2474 =item Using a F<.perltidyrc> command file
2476 If you use perltidy frequently, you probably won't be happy until you
2477 create a F<.perltidyrc> file to avoid typing commonly-used parameters.
2478 Perltidy will first look in your current directory for a command file
2479 named F<.perltidyrc>. If it does not find one, it will continue looking
2480 for one in other standard locations.
2482 These other locations are system-dependent, and may be displayed with
2483 the command C<perltidy -dpro>. Under Unix systems, it will first look
2484 for an environment variable B<PERLTIDY>. Then it will look for a
2485 F<.perltidyrc> file in the home directory, and then for a system-wide
2486 file F</usr/local/etc/perltidyrc>, and then it will look for
2487 F</etc/perltidyrc>. Note that these last two system-wide files do not
2488 have a leading dot. Further system-dependent information will be found
2489 in the INSTALL file distributed with perltidy.
2491 Under Windows, perltidy will also search for a configuration file named perltidy.ini since Windows does not allow files with a leading period (.).
2492 Use C<perltidy -dpro> to see the possible locations for your system.
2493 An example might be F<C:\Documents and Settings\All Users\perltidy.ini>.
2495 Another option is the use of the PERLTIDY environment variable.
2496 The method for setting environment variables depends upon the version of
2497 Windows that you are using. Instructions for Windows 95 and later versions can
2500 http://www.netmanage.com/000/20021101_005_tcm21-6336.pdf
2502 Under Windows NT / 2000 / XP the PERLTIDY environment variable can be placed in
2503 either the user section or the system section. The later makes the
2504 configuration file common to all users on the machine. Be sure to enter the
2505 full path of the configuration file in the value of the environment variable.
2506 Ex. PERLTIDY=C:\Documents and Settings\perltidy.ini
2508 The configuration file is free format, and simply a list of parameters, just as
2509 they would be entered on a command line. Any number of lines may be used, with
2510 any number of parameters per line, although it may be easiest to read with one
2511 parameter per line. Comment text begins with a #, and there must
2512 also be a space before the # for side comments. It is a good idea to
2513 put complex parameters in either single or double quotes.
2515 Here is an example of a F<.perltidyrc> file:
2517 # This is a simple of a .perltidyrc configuration file
2518 # This implements a highly spaced style
2519 -se # errors to standard error output
2520 -w # show all warnings
2521 -bl # braces on new lines
2522 -pt=0 # parens not tight at all
2523 -bt=0 # braces not tight
2524 -sbt=0 # square brackets not tight
2526 The parameters in the F<.perltidyrc> file are installed first, so any
2527 parameters given on the command line will have priority over them.
2529 To avoid confusion, perltidy ignores any command in the .perltidyrc
2530 file which would cause some kind of dump and an exit. These are:
2532 -h -v -ddf -dln -dop -dsn -dtt -dwls -dwrs -ss
2534 There are several options may be helpful in debugging a F<.perltidyrc>
2541 A very helpful command is B<--dump-profile> or B<-dpro>. It writes a
2542 list of all configuration filenames tested to standard output, and
2543 if a file is found, it dumps the content to standard output before
2544 exiting. So, to find out where perltidy looks for its configuration
2545 files, and which one if any it selects, just enter
2551 It may be simplest to develop and test configuration files with
2552 alternative names, and invoke them with B<-pro=filename> on the command
2553 line. Then rename the desired file to F<.perltidyrc> when finished.
2557 The parameters in the F<.perltidyrc> file can be switched off with
2558 the B<-npro> option.
2562 The commands B<--dump-options>, B<--dump-defaults>, B<--dump-long-names>,
2563 and B<--dump-short-names>, all described below, may all be helpful.
2567 =item Creating a new abbreviation
2569 A special notation is available for use in a F<.perltidyrc> file
2570 for creating an abbreviation for a group
2571 of options. This can be used to create a
2572 shorthand for one or more styles which are frequently, but not always,
2573 used. The notation is to group the options within curly braces which
2574 are preceded by the name of the alias (without leading dashes), like this:
2581 where B<newword> is the abbreviation, and B<opt1>, etc, are existing parameters
2582 I<or other abbreviations>. The main syntax requirement is that
2583 the new abbreviation must begin on a new line.
2584 Space before and after the curly braces is optional.
2586 specific example, the following line
2588 airy {-bl -pt=0 -bt=0 -sbt=0}
2590 could be placed in a F<.perltidyrc> file, and then invoked at will with
2592 perltidy -airy somefile.pl
2594 (Either C<-airy> or C<--airy> may be used).
2596 =item Skipping leading non-perl commands with B<-x> or B<--look-for-hash-bang>
2598 If your script has leading lines of system commands or other text which
2599 are not valid perl code, and which are separated from the start of the
2600 perl code by a "hash-bang" line, ( a line of the form C<#!...perl> ),
2601 you must use the B<-x> flag to tell perltidy not to parse and format any
2602 lines before the "hash-bang" line. This option also invokes perl with a
2603 -x flag when checking the syntax. This option was originally added to
2604 allow perltidy to parse interactive VMS scripts, but it should be used
2605 for any script which is normally invoked with C<perl -x>.
2607 =item Making a file unreadable
2609 The goal of perltidy is to improve the readability of files, but there
2610 are two commands which have the opposite effect, B<--mangle> and
2611 B<--extrude>. They are actually
2612 merely aliases for combinations of other parameters. Both of these
2613 strip all possible whitespace, but leave comments and pod documents,
2614 so that they are essentially reversible. The
2615 difference between these is that B<--mangle> puts the fewest possible
2616 line breaks in a script while B<--extrude> puts the maximum possible.
2617 Note that these options do not provided any meaningful obfuscation, because
2618 perltidy can be used to reformat the files. They were originally
2619 developed to help test the tokenization logic of perltidy, but they
2621 One use for B<--mangle> is the following:
2623 perltidy --mangle myfile.pl -st | perltidy -o myfile.pl.new
2625 This will form the maximum possible number of one-line blocks (see next
2626 section), and can sometimes help clean up a badly formatted script.
2628 A similar technique can be used with B<--extrude> instead of B<--mangle>
2629 to make the minimum number of one-line blocks.
2631 Another use for B<--mangle> is to combine it with B<-dac> to reduce
2632 the file size of a perl script.
2634 =item One-line blocks
2636 There are a few points to note regarding one-line blocks. A one-line
2637 block is something like this,
2639 if ($x > 0) { $y = 1 / $x }
2641 where the contents within the curly braces is short enough to fit
2644 With few exceptions, perltidy retains existing one-line blocks, if it
2645 is possible within the line-length constraint, but it does not attempt
2646 to form new ones. In other words, perltidy will try to follow the
2647 one-line block style of the input file.
2649 If an existing one-line block is longer than the maximum line length,
2650 however, it will be broken into multiple lines. When this happens, perltidy
2651 checks for and adds any optional terminating semicolon (unless the B<-nasc>
2652 option is used) if the block is a code block.
2654 The main exception is that perltidy will attempt to form new one-line
2655 blocks following the keywords C<map>, C<eval>, and C<sort>, because
2656 these code blocks are often small and most clearly displayed in a single
2659 One-line block rules can conflict with the cuddled-else option. When
2660 the cuddled-else option is used, perltidy retains existing one-line
2661 blocks, even if they do not obey cuddled-else formatting.
2663 Occasionally, when one-line blocks get broken because they exceed the
2664 available line length, the formatting will violate the requested brace style.
2665 If this happens, reformatting the script a second time should correct
2670 The following flags are available for debugging:
2672 B<--dump-defaults> or B<-ddf> will write the default option set to standard output and quit
2674 B<--dump-profile> or B<-dpro> will write the name of the current
2675 configuration file and its contents to standard output and quit.
2677 B<--dump-options> or B<-dop> will write current option set to standard
2680 B<--dump-long-names> or B<-dln> will write all command line long names (passed
2681 to Get_options) to standard output and quit.
2683 B<--dump-short-names> or B<-dsn> will write all command line short names
2684 to standard output and quit.
2686 B<--dump-token-types> or B<-dtt> will write a list of all token types
2687 to standard output and quit.
2689 B<--dump-want-left-space> or B<-dwls> will write the hash %want_left_space
2690 to standard output and quit. See the section on controlling whitespace
2693 B<--dump-want-right-space> or B<-dwrs> will write the hash %want_right_space
2694 to standard output and quit. See the section on controlling whitespace
2697 B<--no-memoize> or B<-nmem> will turn of memoizing.
2698 Memoization can reduce run time when running perltidy repeatedly in a
2699 single process. It is on by default but can be deactivated for
2700 testing with B<-nmem>.
2702 B<-DEBUG> will write a file with extension F<.DEBUG> for each input file
2703 showing the tokenization of all lines of code.
2705 =item Working with MakeMaker, AutoLoader and SelfLoader
2707 The first $VERSION line of a file which might be eval'd by MakeMaker
2708 is passed through unchanged except for indentation.
2709 Use B<--nopass-version-line>, or B<-npvl>, to deactivate this feature.
2711 If the AutoLoader module is used, perltidy will continue formatting
2712 code after seeing an __END__ line.
2713 Use B<--nolook-for-autoloader>, or B<-nlal>, to deactivate this feature.
2715 Likewise, if the SelfLoader module is used, perltidy will continue formatting
2716 code after seeing a __DATA__ line.
2717 Use B<--nolook-for-selfloader>, or B<-nlsl>, to deactivate this feature.
2719 =item Working around problems with older version of Perl
2721 Perltidy contains a number of rules which help avoid known subtleties
2722 and problems with older versions of perl, and these rules always
2723 take priority over whatever formatting flags have been set. For example,
2724 perltidy will usually avoid starting a new line with a bareword, because
2725 this might cause problems if C<use strict> is active.
2727 There is no way to override these rules.
2735 =item The B<-html> master switch
2737 The flag B<-html> causes perltidy to write an html file with extension
2738 F<.html>. So, for example, the following command
2740 perltidy -html somefile.pl
2742 will produce a syntax-colored html file named F<somefile.pl.html>
2743 which may be viewed with a browser.
2745 B<Please Note>: In this case, perltidy does not do any formatting to the
2746 input file, and it does not write a formatted file with extension
2747 F<.tdy>. This means that two perltidy runs are required to create a
2748 fully reformatted, html copy of a script.
2750 =item The B<-pre> flag for code snippets
2752 When the B<-pre> flag is given, only the pre-formatted section, within
2753 the <PRE> and </PRE> tags, will be output. This simplifies inclusion
2754 of the output in other files. The default is to output a complete
2757 =item The B<-nnn> flag for line numbering
2759 When the B<-nnn> flag is given, the output lines will be numbered.
2761 =item The B<-toc>, or B<--html-table-of-contents> flag
2763 By default, a table of contents to packages and subroutines will be
2764 written at the start of html output. Use B<-ntoc> to prevent this.
2765 This might be useful, for example, for a pod document which contains a
2766 number of unrelated code snippets. This flag only influences the code
2767 table of contents; it has no effect on any table of contents produced by
2768 pod2html (see next item).
2770 =item The B<-pod>, or B<--pod2html> flag
2772 There are two options for formatting pod documentation. The default is
2773 to pass the pod through the Pod::Html module (which forms the basis of
2774 the pod2html utility). Any code sections are formatted by perltidy, and
2775 the results then merged. Note: perltidy creates a temporary file when
2776 Pod::Html is used; see L<"FILES">. Also, Pod::Html creates temporary
2777 files for its cache.
2779 NOTE: Perltidy counts the number of C<=cut> lines, and either moves the
2780 pod text to the top of the html file if there is one C<=cut>, or leaves
2781 the pod text in its original order (interleaved with code) otherwise.
2783 Most of the flags accepted by pod2html may be included in the perltidy
2784 command line, and they will be passed to pod2html. In some cases,
2785 the flags have a prefix C<pod> to emphasize that they are for the
2786 pod2html, and this prefix will be removed before they are passed to
2787 pod2html. The flags which have the additional C<pod> prefix are:
2789 --[no]podheader --[no]podindex --[no]podrecurse --[no]podquiet
2790 --[no]podverbose --podflush
2792 The flags which are unchanged from their use in pod2html are:
2794 --backlink=s --cachedir=s --htmlroot=s --libpods=s --title=s
2795 --podpath=s --podroot=s
2797 where 's' is an appropriate character string. Not all of these flags are
2798 available in older versions of Pod::Html. See your Pod::Html documentation for
2801 The alternative, indicated with B<-npod>, is not to use Pod::Html, but
2802 rather to format pod text in italics (or whatever the stylesheet
2803 indicates), without special html markup. This is useful, for example,
2804 if pod is being used as an alternative way to write comments.
2806 =item The B<-frm>, or B<--frames> flag
2808 By default, a single html output file is produced. This can be changed
2809 with the B<-frm> option, which creates a frame holding a table of
2810 contents in the left panel and the source code in the right side. This
2811 simplifies code browsing. Assume, for example, that the input file is
2812 F<MyModule.pm>. Then, for default file extension choices, these three
2813 files will be created:
2815 MyModule.pm.html - the frame
2816 MyModule.pm.toc.html - the table of contents
2817 MyModule.pm.src.html - the formatted source code
2819 Obviously this file naming scheme requires that output be directed to a real
2820 file (as opposed to, say, standard output). If this is not the
2821 case, or if the file extension is unknown, the B<-frm> option will be
2824 =item The B<-text=s>, or B<--html-toc-extension> flag
2826 Use this flag to specify the extra file extension of the table of contents file
2827 when html frames are used. The default is "toc".
2828 See L<Specifying File Extensions>.
2830 =item The B<-sext=s>, or B<--html-src-extension> flag
2832 Use this flag to specify the extra file extension of the content file when html
2833 frames are used. The default is "src".
2834 See L<Specifying File Extensions>.
2836 =item The B<-hent>, or B<--html-entities> flag
2838 This flag controls the use of Html::Entities for html formatting. By
2839 default, the module Html::Entities is used to encode special symbols.
2840 This may not be the right thing for some browser/language
2841 combinations. Use --nohtml-entities or -nhent to prevent this.
2845 Style sheets make it very convenient to control and adjust the
2846 appearance of html pages. The default behavior is to write a page of
2847 html with an embedded style sheet.
2849 An alternative to an embedded style sheet is to create a page with a
2850 link to an external style sheet. This is indicated with the
2851 B<-css=filename>, where the external style sheet is F<filename>. The
2852 external style sheet F<filename> will be created if and only if it does
2853 not exist. This option is useful for controlling multiple pages from a
2856 To cause perltidy to write a style sheet to standard output and exit,
2857 use the B<-ss>, or B<--stylesheet>, flag. This is useful if the style
2858 sheet could not be written for some reason, such as if the B<-pre> flag
2859 was used. Thus, for example,
2861 perltidy -html -ss >mystyle.css
2863 will write a style sheet with the default properties to file
2866 The use of style sheets is encouraged, but a web page without a style
2867 sheets can be created with the flag B<-nss>. Use this option if you
2868 must to be sure that older browsers (roughly speaking, versions prior to
2869 4.0 of Netscape Navigator and Internet Explorer) can display the
2870 syntax-coloring of the html files.
2872 =item Controlling HTML properties
2874 Note: It is usually more convenient to accept the default properties
2875 and then edit the stylesheet which is produced. However, this section
2876 shows how to control the properties with flags to perltidy.
2878 Syntax colors may be changed from their default values by flags of the either
2879 the long form, B<-html-color-xxxxxx=n>, or more conveniently the short form,
2880 B<-hcx=n>, where B<xxxxxx> is one of the following words, and B<x> is the
2881 corresponding abbreviation:
2884 ---------- -------- --
2887 identifier identifier i
2888 bareword, function bareword w
2890 quite, pattern quote q
2891 here doc text here-doc-text h
2892 here doc target here-doc-target hh
2893 punctuation punctuation pu
2895 structural braces structure s
2896 semicolon semicolon sc
2900 sub definition name subroutine m
2901 pod text pod-text pd
2903 A default set of colors has been defined, but they may be changed by providing
2904 values to any of the following parameters, where B<n> is either a 6 digit
2905 hex RGB color value or an ascii name for a color, such as 'red'.
2907 To illustrate, the following command will produce an html
2908 file F<somefile.pl.html> with "aqua" keywords:
2910 perltidy -html -hck=00ffff somefile.pl
2912 and this should be equivalent for most browsers:
2914 perltidy -html -hck=aqua somefile.pl
2916 Perltidy merely writes any non-hex names that it sees in the html file.
2917 The following 16 color names are defined in the HTML 3.2 standard:
2936 Many more names are supported in specific browsers, but it is safest
2937 to use the hex codes for other colors. Helpful color tables can be
2938 located with an internet search for "HTML color tables".
2940 Besides color, two other character attributes may be set: bold, and italics.
2941 To set a token type to use bold, use the flag
2942 B<--html-bold-xxxxxx> or B<-hbx>, where B<xxxxxx> or B<x> are the long
2943 or short names from the above table. Conversely, to set a token type to
2944 NOT use bold, use B<--nohtml-bold-xxxxxx> or B<-nhbx>.
2946 Likewise, to set a token type to use an italic font, use the flag
2947 B<--html-italic-xxxxxx> or B<-hix>, where again B<xxxxxx> or B<x> are the
2948 long or short names from the above table. And to set a token type to
2949 NOT use italics, use B<--nohtml-italic-xxxxxx> or B<-nhix>.
2951 For example, to use bold braces and lime color, non-bold, italics keywords the
2952 following command would be used:
2954 perltidy -html -hbs -hck=00FF00 -nhbk -hik somefile.pl
2956 The background color can be specified with B<--html-color-background=n>,
2957 or B<-hcbg=n> for short, where n is a 6 character hex RGB value. The
2958 default color of text is the value given to B<punctuation>, which is
2961 Here are some notes and hints:
2963 1. If you find a preferred set of these parameters, you may want
2964 to create a F<.perltidyrc> file containing them. See the perltidy man
2965 page for an explanation.
2967 2. Rather than specifying values for these parameters, it is probably
2968 easier to accept the defaults and then edit a style sheet. The style
2969 sheet contains comments which should make this easy.
2971 3. The syntax-colored html files can be very large, so it may be best to
2972 split large files into smaller pieces to improve download times.
2976 =head1 SOME COMMON INPUT CONVENTIONS
2978 =head2 Specifying Block Types
2980 Several parameters which refer to code block types may be customized by also
2981 specifying an associated list of block types. The type of a block is the name
2982 of the keyword which introduces that block, such as B<if>, B<else>, or B<sub>.
2983 An exception is a labeled block, which has no keyword, and should be specified
2984 with just a colon. To specify all blocks use B<'*'>.
2986 For example, the following parameter specifies C<sub>, labels, C<BEGIN>, and
2989 -cscl="sub : BEGIN END"
2991 (the meaning of the -cscl parameter is described above.) Note that
2992 quotes are required around the list of block types because of the
2993 spaces. For another example, the following list specifies all block types
2994 for vertical tightness:
2998 =head2 Specifying File Extensions
3000 Several parameters allow default file extensions to be overridden. For
3001 example, a backup file extension may be specified with B<-bext=ext>,
3002 where B<ext> is some new extension. In order to provides the user some
3003 flexibility, the following convention is used in all cases to decide if
3004 a leading '.' should be used. If the extension C<ext> begins with
3005 C<A-Z>, C<a-z>, or C<0-9>, then it will be appended to the filename with
3006 an intermediate '.' (or perhaps an '_' on VMS systems). Otherwise, it
3007 will be appended directly.
3009 For example, suppose the file is F<somefile.pl>. For C<-bext=old>, a '.' is
3010 added to give F<somefile.pl.old>. For C<-bext=.old>, no additional '.' is
3011 added, so again the backup file is F<somefile.pl.old>. For C<-bext=~>, then no
3012 dot is added, and the backup file will be F<somefile.pl~> .
3014 =head1 SWITCHES WHICH MAY BE NEGATED
3016 The following list shows all short parameter names which allow a prefix
3017 'n' to produce the negated form:
3019 D anl asc aws b bbb bbc bbs bl bli boc bok bol bot ce
3020 csc dac dbc dcsc ddf dln dnl dop dp dpro dsc dsm dsn dtt dwls
3021 dwrs dws f fll frm fs hsc html ibc icb icp iob isbc lal log
3022 lp lsl ohbr okw ola oll opr opt osbr otr ple pod pvl q
3023 sbc sbl schb scp scsb sct se sfp sfs skp sob sohb sop sosb sot
3024 ssc st sts syn t tac tbc toc tp tqw tsc w x bar kis
3026 Equivalently, the prefix 'no' or 'no-' on the corresponding long names may be
3033 =item Parsing Limitations
3035 Perltidy should work properly on most perl scripts. It does a lot of
3036 self-checking, but still, it is possible that an error could be
3037 introduced and go undetected. Therefore, it is essential to make
3038 careful backups and to test reformatted scripts.
3040 The main current limitation is that perltidy does not scan modules
3041 included with 'use' statements. This makes it necessary to guess the
3042 context of any bare words introduced by such modules. Perltidy has good
3043 guessing algorithms, but they are not infallible. When it must guess,
3044 it leaves a message in the log file.
3046 If you encounter a bug, please report it.
3048 =item What perltidy does not parse and format
3050 Perltidy indents but does not reformat comments and C<qw> quotes.
3051 Perltidy does not in any way modify the contents of here documents or
3052 quoted text, even if they contain source code. (You could, however,
3053 reformat them separately). Perltidy does not format 'format' sections
3054 in any way. And, of course, it does not modify pod documents.
3062 =item Temporary files
3064 Under the -html option with the default --pod2html flag, a temporary file is
3065 required to pass text to Pod::Html. Unix systems will try to use the POSIX
3066 tmpnam() function. Otherwise the file F<perltidy.TMP> will be temporarily
3067 created in the current working directory.
3069 =item Special files when standard input is used
3071 When standard input is used, the log file, if saved, is F<perltidy.LOG>,
3072 and any errors are written to F<perltidy.ERR> unless the B<-se> flag is
3073 set. These are saved in the current working directory.
3075 =item Files overwritten
3077 The following file extensions are used by perltidy, and files with these
3078 extensions may be overwritten or deleted: F<.ERR>, F<.LOG>, F<.TEE>,
3079 and/or F<.tdy>, F<.html>, and F<.bak>, depending on the run type and
3082 =item Files extensions limitations
3084 Perltidy does not operate on files for which the run could produce a file with
3085 a duplicated file extension. These extensions include F<.LOG>, F<.ERR>,
3086 F<.TEE>, and perhaps F<.tdy> and F<.bak>, depending on the run type. The
3087 purpose of this rule is to prevent generating confusing filenames such as
3088 F<somefile.tdy.tdy.tdy>.
3094 perlstyle(1), Perl::Tidy(3)
3098 This man page documents perltidy version 20130922.
3102 Michael Cartmell supplied code for adaptation to VMS and helped with
3105 Yves Orton supplied code for adaptation to the various versions
3108 Axel Rose supplied a patch for MacPerl.
3110 Hugh S. Myers designed and implemented the initial Perl::Tidy module interface.
3112 Many others have supplied key ideas, suggestions, and bug reports;
3113 see the CHANGES file.
3118 email: perltidy at users.sourceforge.net
3119 http://perltidy.sourceforge.net
3123 Copyright (c) 2000-2012 by Steve Hancock
3127 This package is free software; you can redistribute it and/or modify it
3128 under the terms of the "GNU General Public License".
3130 Please refer to the file "COPYING" for details.
3134 This package is distributed in the hope that it will be useful,
3135 but WITHOUT ANY WARRANTY; without even the implied warranty of
3136 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
3138 See the "GNU General Public License" for more details.