The London Perl and Raku Workshop takes place on 26th Oct 2024. If your company depends on Perl, please consider sponsoring and/or attending.
 / 
Text-FixEOL-1.01
River stage zero No dependents
1 ++
/ Text::FixEOL

NAME

Text::FixEOL - Canonicalizes text to a specified EOL/EOF convention, repairing any 'mixed' usages

SYNOPSIS

 use Text::FixEOL;

 my $fixer = Text::FixEOL->new({ EOL     => 'platform|asis|mac|dos|unix|vms|netware|network|os2|cygwin|crlf|cr|lf|literal:$value',
                                 EOF     => 'platform|add|asis|remove|mac|dos|unix|vms|netware|cygwin|network|os2',
                                 FixLast => 'platform│yes│no│mac│dos│unix│vms│netware│network│os2│cygwin',
             });
 my $fixed_text = $fixer->fix_eol($string);

 my $mac_text     = $fixer->eol_to_mac($string);
 my $windows_text = $fixer->eol_to_dos($string);
 my $unix_text    = $fixer->eol_to_unix($string);
 my $crlf_text    = $fixer->to_crlf($string);

DESCRIPTION

Converts the EOL and EOF conventions in the passed string to a canonicalization form that handles 'mixed' EOL conventions.

It canonicalizes EOL as \n (the platform defined EOL) if it does not know the particular platform. Can also 'fix' the end-of-file mark if needed and ensure that the last line of the string is EOL terminated.

CHANGES

1.00 2005.05.02 - Initial Release

1.01 2005.05.03 - Changes to Build.PL to have Test::NoWarnings and Test::More as prerequisite modules and changes to META.yml to try to make CPANPLUS happy. Small documentation change to 'new'. No functional changes.

METHODS

new({ [EOL => 'platform|asis|mac|dos|unix|vms|netware|network|os2|cygwin|crlf|cr|lf|literal:$value', ] [EOF => 'platform|add|asis|remove|mac|dos|unix|vms|netware|network|os2|cygwin', ] [FixLast => 'platform|yes|no|mac│dos│unix│vms│netware│network│os2│cygwin', ] });

Instantates a new Text::FixEOL object and (optionally) allows setting the modes of operation.

fix_eol($input);

Converts the EOLs/EOF in the passed text to a canonicalization form. This includes fixing EOLs to the appropriate values, as well as handling EOF issues. Specifically, the presence of terminal Ctrl-Z characters (EOF) and whether or not the last line of the string is terminated with EOL (FixLast).

When running in default mode (all setting at 'platform') on an unrecognized platform, it will attempt to convert to the local machine conventions using \n for EOL, 'remove' for EOF and 'yes' for FixLast.

It attempts to 'Do What I Mean' (DWIM) for mixed EOL values (text where there is a mixture of differing EOL conventions such as can happen when a document is editted in more than one environment).

Example:

   my $fixed_string = $fixer->fix_eol($string);

The full grid of platform defaults is as follows:

                   EOL        EOF       FixLast
 unix              \012      remove       yes
 dos           \015\012        asis       yes
 windows       \015\012        asis       yes
 mswin32       \015\012        asis       yes
 mac               \015      remove       yes
 macos             \015      remove       yes
 vms           \015\012      remove       yes
 os2           \015\012        asis       yes
 netware       \015\012        asis       yes
 cygwin        \015\012        asis       yes
 network       \015\012      remove       yes

The EOL canonicalization form is premised on the following example map which attempts to 'DWIM' for all cases of between 1 and 4 consecutive CR/LF characters. The values shown in the table are for illustrative purposes only. The actual values depend on the conversion mode(s) specified.

  ["\012" => "\012"],
  ["\015" => "\012"],

  ["\012\015" => "\012"],
  ["\015\012" => "\012"],
  ["\015\012\015" => "\012\012"],
  ["\012\015\012" => "\012\012"],

  ["\012a\012b\012" => "\012a\012b\012"],
  ["\012a\012b\015" => "\012a\012b\012"],
  ["\012a\015b\012" => "\012a\012b\012"],
  ["\012a\015b\015" => "\012a\012b\012"],
  ["\015a\012b\012" => "\012a\012b\012"],
  ["\015a\012b\015" => "\012a\012b\012"],
  ["\015a\015b\012" => "\012a\012b\012"],
  ["\015a\015b\015" => "\012a\012b\012"],

  ["\012\015a\012\015b\012\015" => "\012a\012b\012"],
  ["\012\015a\012\015b\015"     => "\012a\012b\012"],
  ["\012\015a\015b\012\015"     => "\012a\012b\012"],
  ["\012\015a\015b\015"         => "\012a\012b\012"],
  ["\015a\012\015b\012\015"     => "\012a\012b\012"],
  ["\015a\012\015b\015"         => "\012a\012b\012"],
  ["\015a\015b\012\015"         => "\012a\012b\012"],

  ["\015\012a\015\012b\015\012" => "\012a\012b\012"],
  ["\015\012a\015\012b\015"     => "\012a\012b\012"],
  ["\015\012a\015b\015\012"     => "\012a\012b\012"],
  ["\015\012a\015b\015"         => "\012a\012b\012"],
  ["\015a\015\012b\015\012"     => "\012a\012b\012"],
  ["\015a\015\012b\015"         => "\012a\012b\012"],
  ["\015a\015b\015\012"         => "\012a\012b\012"],
  ["\015\012\015a\015\012\015b\015\012\015" => "\012\012a\012\012b\012\012"],
  ["\015\012\015a\015\012\015b\015"         => "\012\012a\012\012b\012"],
  ["\015\012\015a\015b\015\012\015"         => "\012\012a\012b\012\012"],
  ["\015\012\015a\015b\015"                 => "\012\012a\012b\012"],
  ["\015a\015\012\015b\015\012\015"         => "\012a\012\012b\012\012"],
  ["\015a\015\012\015b\015"                 => "\012a\012\012b\012"],
  ["\015a\015b\015\012\015"                 => "\012a\012b\012\012"],

  ["\012\012\012a\012\012\012b\012\012\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\012\012\015a\012\012\015b\012\012\015" => "\012\012a\012\012b\012\012"],
  ["\012\015\012a\012\015\012b\012\015\012" => "\012\012a\012\012b\012\012"],
  ["\012\015\015a\012\015\015b\012\015\015" => "\012\012a\012\012b\012\012"],
  ["\015\012\012a\015\012\012b\015\012\012" => "\012\012a\012\012b\012\012"],
  ["\015\012\015a\015\012\015b\015\012\015" => "\012\012a\012\012b\012\012"],
  ["\015\015\012a\015\015\012b\015\015\012" => "\012\012a\012\012b\012\012"],
  ["\015\015\015a\015\015\015b\015\015\015" => "\012\012\012a\012\012\012b\012\012\012"],


  ["\012\012\012\012a\012\012\012\012b\012\012\012\012" => "\012\012\012\012a\012\012\012\012b\012\012\012\012"],
  ["\012\012\012\015a\012\012\012\015b\012\012\012\015" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\012\012\015\012a\012\012\015\012b\012\012\015\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\012\012\015\015a\012\012\015\015b\012\012\015\015" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\012\015\012\012a\012\015\012\012b\012\015\012\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\012\015\012\015a\012\015\012\015b\012\015\012\015" => "\012\012a\012\012b\012\012"],
  ["\012\015\015\012a\012\015\015\012b\012\015\015\012" => "\012\012a\012\012b\012\012"],
  ["\012\015\015\015a\012\015\015\015b\012\015\015\015" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\012\012\012a\015\012\012\012b\015\012\012\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\012\012\015a\015\012\012\015b\015\012\012\015" => "\012\012a\012\012b\012\012"],
  ["\015\012\015\012a\015\012\015\012b\015\012\015\012" => "\012\012a\012\012b\012\012"],
  ["\015\012\015\015a\015\012\015\015b\015\012\015\015" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\015\012\012a\015\015\012\012b\015\015\012\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\015\012\015a\015\015\012\015b\015\015\012\015" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\015\015\012a\015\015\015\012b\015\015\015\012" => "\012\012\012a\012\012\012b\012\012\012"],
  ["\015\015\015\015a\015\015\015\015b\015\015\015\015" => "\012\012\012\012a\012\012\012\012b\012\012\012\012"],

Additionally any Ctrl-Z (EOF) characters are processed according to the EOF setting.

eol_to_unix($string);

Converts the passed string to Unix EOL/EOF conventions. This is equivalent to using fix_eol with

   EOL     => 'unix',
   EOF     => 'unix',
   FixLast => 'unix'

The Unix EOL convention terminates all lines with \012. Ctrl-Z (EOF) characters at the end of the string are removed.

eol_to_dos($string);

Converts the passed string to Windows/DOS EOL/EOF conventions This is equivalent to using fix_eol with

   EOL     => 'dos',
   EOF     => 'dos',
   FixLast => 'dos'

The DOS EOL convention terminates all lines with \015\012. Ctrl-Z (EOF) characters at the end of the string are left alone if present.

eol_to_mac($string);

Converts the passed string to Macintosh EOL/EOF conventions This is equivalent to using fix_eol with

   EOL     => 'mac',
   EOF     => 'mac',
   FixLast => 'mac'

The Mac EOL convention terminates all lines with \012. Ctrl-Z (EOF) characters at the end of the string are removed.

eol_to_network($string);

Converts the passed string to network EOL/EOF conventions. This is equivalent to using fix_eol with

   EOL     => 'network',
   EOF     => 'network',
   FixLast => 'network'

The network EOL convention terminates all lines with \015\012. Ctrl-Z (EOF) characters at the end of the string are removed.

eol_to_crlf($string);

Converts the passed string to CRLF format without otherwise changing it.

This is equivalent to using fix_eol with

   EOL     => 'crlf',
   EOF     => 'asis',
   FixLast => 'no'
fix_last_handling(['platform|yes|no|mac│dos│unix│vms│netware│network│os2│cygwin']);

Get/set accessor that specifies whether or not the last line of the string should have an EOL added if missing. Default is 'platform'.

fix_last_mode;

Returns the current mode ('yes' or 'no') for the handling of the last line of the string (the actual 'yes'/'no' value rather than the mnemomic used for configuration).

'yes' indicates the processor will ensure the last line is terminated with an EOL value, appending one if needed.

'no' indicates the processor will not append a EOL if it is missing.

eol_handling(['platform|asis|mac|dos|unix|vms|netware|network|os2|cygwin|crlf|cr|lf|literal:$value']);

Sets/gets the end-of-line character handling property. The default is 'platform'.

The full list of supported settings is as follows:

  mac      - use Macintosh default (\015)
  macos    - use Macintosh default (\015)
  dos      - use DOS default       (\015\012)
  windows  - use DOS default       (\015\012)
  mswin32  - use DOS default       (\015\012)
  unix     - use Unix default      (\012)
  vms      - use vms default       (\015\012)
  netware  - use netware default   (\015\012)
  network  - use network default   (\015\012)
  os2      - use OS/2 default      (\015\012)
  cygwin   - use CygWin default    (\015\012)
  platform - use the currect execution environment default.
  asis     - leave EOLs alone.
  crlf     - use CRLF              (\015\012)
  cr       - use CR                (\015)
  lf       - use LF                (\012)

 'literal:$value' where you replace '$value' with a string value
        instructs the converter to use the literal you specify
        for EOL.
eol_mode;

Returns the current EOL string (the actual string value(s) for the EOL). This is the literal string value, not the mnemomic used for configuration.

eof_handling(['platform|add|asis|remove|mac|dos|unix|vms|netware|network|os2|cygwin']);

Specifies what is to be done with DOS end-of-file characters (control-Z).

 'add'      a Ctrl-Z (EOF) will be appended if not already present.
 'asis'     the EOF (whatever it is) will be left alone.
 'remove'   any trailing Ctrl-Z (EOF) characters will be removed.
 'platform' use the current execution environment platform's default for EOF handling.

 The platforms have the following default EOF handling behaviors:

             EOF Handling
 unix           remove
 dos              asis
 windows          asis
 mswin32          asis
 mac            remove
 macos          remove
 vms            remove
 os2              asis
 netware          asis
 cygwin           asis
 network        remove
eof_mode;

Returns the current EOF mode ('asis', 'remove' or 'add'). This is the actual mode rather than the mnemomic used for configuration.

COPYRIGHT

2005 Benjamin Franz <snowhare@nihongo.org>

LICENSE

Artistic

BUGS

None known.

TODO

Everything

VERSION

1.01 - 2004.05.03 Documentation and CPANPLUS compatibility changes.

1 POD Error

The following errors were encountered while parsing the POD:

Around line 12:

Non-ASCII character seen before =encoding in ''platform│yes│no│mac│dos│unix│vms│netware│network│os2│cygwin','. Assuming UTF-8