StringEOLToCRLF() - Convert EOL (@CR or @LF) to @CRLF.

[guinness 1,480]

Convert EOL characters @CR and @LF to @CRLF. This is ideal if you have a document that has a mismatch of @CR and @LF line endings.

#include <MsgBoxConstants.au3>

; Convert all line endings to @CRLF.
Local $sString = StringEOLToCRLF('This is a sentence ' & @CR & Chr(12) & 'with ' & Chr(13) & 'whitespace.' & @CRLF) ; 12 = Form Feed, 13 = Carriage Return.

; Display the converted line endings.
MsgBox($MB_SYSTEMMODAL, '', $sString)

; Visually display that @CR and @LF have converted to @CRLF.
$sString = StringReplace($sString, @CRLF, '@CRLF')
MsgBox($MB_SYSTEMMODAL, '', $sString)

Func StringEOLToCRLF($sString) ; Initial Regular expression by Melba23 with a new suggestion by Ascend4nt and modified By guinness.
    Return StringRegExpReplace($sString, '(*BSR_ANYCRLF)\R', @CRLF) ; Idea by Ascend4nt
EndFunc   ;==>StringEOLToCRLF

Edited May 25, 2014 by guinness

[Ascend4nt 122]

Or simply:

StringRegExpReplace($sString, '\R', @CRLF)

 

See the Newline Sequences section of the PCRE Pattern specification.  It's a nice shortcut

[guinness 1,480]

Ah, good to know. Thanks.

Wait, that includes f e.g.

Local $sString = StringEOLToCRLF("This is a sentence " & @CR & Chr(12) & "with " & Chr(13) & "whitespace." & @CRLF)

Edited July 21, 2013 by guinness

[Ascend4nt 122]

True, its not just linefeed and carriage-returns..  I suppose for me it doesn't matter though, as they are all ending the current line.

[guinness 1,480]

True, its not just linefeed and carriage-returns..  I suppose for me it doesn't matter though, as they are all ending the current line.

Me neither. It's a good shortcut.

[jchd 1,055]

FYI you can still restrict what R matches.

"(*BSR_ANYCRLF)R" is a pattern for matching only CR or LF or CRLF

On the contrary and by default with compiled options currently used:

"(*BSR_UNICODE)R" matches any Unicode newline sequence, i.e. is equivalent to "(?>rn|n|x0b|f|r|x85)"

BSR acronym stands for BackSlash R.

These two options only apply to what R matches and are independant of what PCRE regards as being a "line". "Line" breaks are themselves ruled by another set of options which govern what $ and ^ mean:

(*CR) carriage return

(*LF) linefeed

(*CRLF) carriage return, followed by linefeed

(*ANYCRLF) any of the three above

(*ANY) all Unicode newline sequences

All (*????) option must appear before all other options and in uppercase. Refer to PCRE docs for more details.

[guinness 1,480]

Ah! When I read BSR_ANYCRLF and such-like, I thought it had to be done a Jon's end. Go it jchd. Thanks.

[jchd 1,055]

No its an old feature, already available long prior 3.3.8.1

[guinness 1,480]

Updated the function to using R.