JREPL.BAT v8.6 - regex text processor with support for text highlighting and alternate character sets

[bobax]

Thanks a lot for feedback, that is much appreciated.
I am pretty new to this so I will give it a try and experiment on it...

Riko

[GEORGE300]

Hello there,

For a dailly usage case in which all files in a folder need to be converted from utf-8 to utf-16 is this script suited ?
So, just run manually script when needed.
Thanks a lot

[dbenham]

Converting utf-8 to utf-16 could be done with JREPL, but you would get much better performance with aGerman's CONVERT.exe. Its sole purpose is to convert text files from one encoding to another.

Both utilities process only one file, so you would have to process an entire folder using a FOR loop either way.

[aGerman]

JREPL might be a real good choice because it's a script which works with on-board resources.

Code: Select all

for %%i in ("*.txt") do cmd /c jrepl "^" "" /f "%%~i|UTF-8" /o "-|UTF-16"

CONVERTCP has the disadvantage of being a 3rd party executable. However, if performance matters (e.g. due to big files or a lot of text in total) it might be more suitable for you.

Code: Select all

for %%i in ("*.txt") do convertcp "UTF-8" "UTF-16" /b /i "%%~i" /o "%%~i.tmp~" && move /y "%%~i.tmp~" "%%~i"

(Direct overwriting of the original file is not supported on purpose.)


Beware of running those scripts twice on the same files!

Steffen

[Snazz]

Hi,

I'm new to JREPL and this level of regex, forgive me if this has been covered in the previous 484 posts.

I've got an expression working on regex101 here but it doesn't seem to work with JREPL.

Code: Select all

(hello_worldi[0-9]+e[0-9]+)

It worked when there was only one "[0-9]+" but having a second one after the "e" seems to break it.

Essentially I'm trying to add a setting to a text file, where the settings only exist there if they're different to default.
So I find the setting that is always next to it, which has dynamic numbers before and after the "e" character.

[Hazado]

https://regex101.com/r/UnSIRS/1

Im attempting to make this regex work. Does work with regex101.

Regex - <AppSet>[\r\s]*?<Applicator>[\r\s]*?<plMatrixChannelApplicator>[\r\s]*?.*Pinky((?!<\/AppSet>))[\s\S]*?<\/AppSet>

Any Help is appreciated. Thank you

[FreakyBird]

Hello,

I'm struggling with a difficulty to use JREPL.bat for my goal. Hope someone can help me

I have this code :

Code: Select all

<config>
        <bool name="audio.bgmusic" value="false" />
	<bool name="global.achievements" value="true" />
	<string name="listViewStyle" value="Spin" />
	<string name="HiddenFeatures" value="Freaky bird;joy;ekett;prorata;great;super;beautiful" />
	<string name="Language" value="us_US" />
</config>

What i want is to find the hiddenFeatures line and replace the "Freaky bird;joy;ekett;prorata;great;super;beautiful" that is between quote. This value change everytime so actually i've tried :

Code: Select all

setlocal enableDelayedExpansion
set "newValue=myNewValue"
type "settings.cfg"|jrepl "(<string name="HiddenFeatures" value=").*(" />)" "$1!newValue!$2" >fileName.cfg.new
Move /y "fileName.cfg.new" "settings.cfg"

But it doesn't work. Can somenone help ?

Thanks in adavance,

[aGerman]

Use extra escape sequences (option /XSEQ) in order to to be able to use \q rather than quotes in your pattern.
Also, JREPL supports overwriting of the input file.

Code: Select all

setlocal enableDelayedExpansion
set "newValue=myNewValue"
cmd /c jrepl "(<string name=\qHiddenFeatures\q value=\q).*(\q />)" "$1!newValue!$2" /XSEQ /F "settings.cfg" /O -

Steffen

[FreakyBird]

It works perfectly, thanks !

[medp7060]

Hi there.

I've got this batch to extract the text between wordA and wordB from a text file: wordA----------text1-----------wordB

Code: Select all

@echo off
type infile |jrepl ".*wordA(.*)wordB.*" "$1" > outfile

It works when there are only one set of wordA and wordB. It returns i.e. "----------text1-----------"

However, if the wordB appears more than once, e.g. wordA----------text1-----------wordB-------text2-------------wordB

then output is all the text between the first wordA and the last wordB, i.e. "----------text1-----------wordB-------text2-------------"

How can I make it return the text between the first "wordA" and the first "wordB"?

Example:

Code: Select all

{"info":"OK","content":"This is my text","status":200,"error_code":0}

If I set wordA to content":" and wordB to "
The output of the above script is: This is my text","status":200,"error_code
What I want is: This is my text

Your help is much appreciated.

[aGerman]

Code: Select all

cmd /c jrepl.bat ".*\qcontent\q:\q(.+?(?=\q)).*" "$1" /XSEQ /F "infile" /O "outfile"

https://regex101.com/r/kbvFkJ/1

Steffen

[medp7060]

Many thanks, Steffen. It works as expected!!!

[IanC]

I'm struggling to do a JPREL operation on a simple XML (GPX) file. The section I want to delete (complete with line-feeds) is as follows:

Code: Select all

<extensions>
<mmttimezone>+01:00</mmttimezone>
</extensions>

So what am I doing wrong here?

Code: Select all

Call jrepl "\<extensions\>\s.*?\s\<\/extensions\>\s" "" /M/f MMT.gpx /o -

I think it's something to do with line-feeds. I've tried \r \n amongst others, and I believe the file is UTF-8. Not sure if I have the /M switch in the correct place, or even if I need it?

[aGerman]

If you want to use \r and \n you have to pass option /XSEQ.
That should work:

Code: Select all

Call jrepl "\<extensions\>\r\n.*?\r\n\<\/extensions\>\r\n" "" /XSEQ /M /F "MMT.gpx" /O -

Steffen

[IanC]

If you want to use \r and \n you have to pass option /XSEQ.
That should work:

Code: Select all

Call jrepl "\<extensions\>\r\n.*?\r\n\<\/extensions\>\r\n" "" /XSEQ /M /F "MMT.gpx" /O -

Steffen

Many thanks, that did it!