Regexp, replacing line and insert before CRLF

[Myicq]

I have a somewhat strange regexp problem I can't seem to come over. Hope someone can point me in the right direction.

Task: 

I have some lines (objects) stored in an array. The lines are of the format

4000-1=Some Text[cr][lf]
4001-1=Other text[cr][lf]

 

The [cr][lf] is not literal, but Windows line endings. I put here to clarify.

Now I would like to comment on these lines, to end up with:

4000-1=Some Text             ; This is a comment for first line [cr][lf]
4001-1=Other text             ; This is a comment for second line  [cr][lf]

 

As the comment is determined by the number, I have made a keys file, like

4000=this is comment for 4000 lines
4001=this is comment for 4001 lines

 

..and I read that file with inireadsection into $keys. Works great.

 

Here is my problem: 

If I have the following code: 

 

for $k = 1 to $keys[0][0]
$regexp = $keys[$k][0] & "-(.*)$"
$new = $keys[$k][0] & "-$1" & @TAB & @TAB & ";" & $keys[$k][1]
$objects[0] = stringregexpreplace($objects[0],  $regexp , $new)

 

..the output ($objects[0] ) end up with the comment AFTER the crlf pair. 

 

4000-1=Some Text
            ; This is a comment for first line [cr][lf]
4001-1=Other text
            ; This is a comment for second line  [cr][lf]

[czardas]

I can't test anything with the system I'm on right now but you could try using (rn), (*CRLF) or (*ANYCRLF) in a second group and backreference "$1" & $keys[$k][1] & "$2" ==> something like that. You need to remove the CRLF characters from the end of $keys[$k][0] for it to work, although I'm not exactly sure how they ended up in there.

What I wrote is probably a mess, but I think it's going in the right direction.

Edited May 12, 2014 by czardas

[guinness]

Replace lines 2 and 3 with this...

$regexp = $keys[$k][0] & "-(.*)(\R)"
$new = $keys[$k][0] & "-$1" & @TAB & @TAB & ";" & $keys[$k][1] & "$2"

[Myicq]

@guiness: thanks for the suggestion. Initially did not work, but I went over the files again.

Turned out that KEYS file (the inifile) was made by accident in unix file format, with endings = @LF. 

I can't see how this should mess up things for the regexpreplace, but apparently AutoIT does not like to work with fileread on two different endings. The inners of that is beyond my understanding.

I am just happy all works. Thanks for your help!

[guinness]

R should take of that.

Edit: Ah the INI file, OK nothing to do my regular expression. By the way R looks for both r, n or rn. So it doesn't matter about the line endings in your other file.

Edited May 12, 2014 by guinness

[Myicq]

R should take of that.

Edit: Ah the INI file, OK nothing to do my regular expression. By the way R looks for both r, n or rn. So it doesn't matter about the line endings in your other file.

 

No, nothing to do with your regexp. But.. Actually it turned out it had a major importance because when I load the ini file with a different encoding, the regexpreplace gave different result. So some internal mixup of line endings. Anyway, all good. Once again, thank you for your help.