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

[mirrormirror]

Thank you - it mostly works after testing on one file but one line fails:

Code: Select all

call findrepl /o:13:13 < "%file%"|jrepl ".*" "SET \qmyVar2=$1\q" /x /i

produces:

Code: Select all

SET "myVar2=$1"SET "myVar2=$1"

Line 13 is:
"second.string here!"
- no "SET" or anything, just a string enclosed in quotes. The output needs to strip the quotes, then output this format
[SET "][myVAR2=][unquoted string][EndQuote]
like this:

Code: Select all

SET "myVAR2=second.string here!"

[mirrormirror]

Actually, I may have resolved it - it's been a while since I've used regex but I remembered a few things. - I changed the line from this:

Code: Select all

call findrepl /o:13:13 < "%file%"|jrepl ".*" "SET \qmyVar2=$1\q" /x /i

to:

Code: Select all

call findrepl /o:13:13 <"%file%" |jrepl "\q(.*)\q" "SET \qmyVar2=$1\q" /x /i

and I seem to be getting the correct output - at least from the command-line. I am going to test the whole thing on a few more files...

[foxidrive]

I edited my code on the last post too mirrormirror.

It's a little different and just removes the first and last character without checking for a quote.

[zimxavier]

Hi !

I have an issue with some lines with space(s) in excess:

call jrepl "(^\t)([a-zA-Z0-9_]*)" $2 /jmatch /f input.txt /o output.txt

input.txt

Code: Select all

   events = { 910 911 }
    action_clan_sentiment = {
   potential = {

output.txt

events
potential

I wish it extract "action_clan_sentiment" too, but there is a space before it (and after a tab). How can I write a code which catch all three like that :

events
action_clan_sentiment
potential

i.e. no take account of one or two space(s) in excess and take account of each tab.

PS: jrepl is very powerful, thank you

[dbenham]

You are asking about basic regex functionality.

You can use the shorthand \w to represent a "word" character, equivalent to [a-zA-Z0-9_]

And you can use \s to represent whitespace, including space and tab.

I believe you should be looking for this pattern (in English):

• Beginning of line
• followed by 0 or more whitespace characters
• followed by 1 or more "word" characters
• followed by 0 or more whitespace characters
• followed by =

And you want to capture the "word" characters, so you only need one set of parentheses:

Code: Select all

call jrepl "^\s*(\w+)\s*=" $1 /jmatch /f input.txt /o output.txt

Dave Benham

[Aacini]

I am sorry, but I can't resist the temptation of post this reply...

Both dbenham's JREPL.BAT and my own FindRepl.bat programs are extensive and complex applications designed to perform advanced and very sophisticated text replacement operations. However, I never have seen that anyone of these programs be used at least at a small fraction of their capabilities! These programs are mainly sub-used, and the resources invested in they are wasted...

For example, although FindRepl.bat is very capable of extract the line # 13 of a file, IMHO it is just absurd to use a 950 lines program to achieve a result that can be obtained in a simpler way; like this one:

Code: Select all

for /F "usebackq skip=12 delims=" %%a in ("%file%") do set "line13=%%a" & goto break
:break

... or this one, that don't use a GOTO:

Code: Select all

set "line13="
for /F "usebackq skip=12 delims=" %%a in ("%file%") do if not defined line13 set "line13=%%a"

In a similar way, the replace operation shown in previous reply is finally executed in a very simple function in just one line of JScript code, so certainly you don't need a 850 lines program (like JREPL.BAT) in order to write and execute the same line in a small Batch-JScript hybrid file; for example:

Code: Select all

@set @a=0 //  &  cscript //nologo //E:JScript "%~F0"  &  goto :EOF

WScript.Stdout.Write(WScript.Stdin.ReadAll().replace(/^\s*(\w+).+/gm,"$1\r\n"));

Output example:

Code: Select all

C:\> type input.txt
   events = { 910 911 }
    action_clan_sentiment = {
   potential = {

C:\> test.bat < input.txt
events
action_clan_sentiment
potential

You may review other examples of not-so-simple replacements that can be achieved in a few JSCript lines here or here...


Nor JREPL.BAT neither FindRepl.bat can avoid that the user needs to learn regular expressions in order to successfully use these programs

Antonio


PS - You may use the "Search an extra string in a block of lines (/B switch)" FindRepl's feature to perform a replacement in line 13 of the file:

Code: Select all

call findrepl "" /O:13:13 /Q:' /B:"'(.*)'" "SET 'myVar2=$1'" < "%file%"

[zimxavier]

Thanks to both of you
I have now a list of regex : https://msdn.microsoft.com/en-us/librar ... 80%29.aspx
It will take time to assimilate all of these regex, but it's very useful for my work already.

I have a new question. From input.txt :

Code: Select all

   option = {

      prestige = -20
      custom_tooltip = { text = no_own_ball_tooltip }
      name = EVTOPTB_REP_104
      wealth = -100
      narrative_event = { id = REP.105 days = 14 tooltip = EVTTOOLTIP_REP_105 } 

I wish to extract all characters between "option =" and the end of the first following line containing "name ="

ouput.txt:

Code: Select all

   option = {

      prestige = -20
      custom_tooltip = { text = no_own_ball_tooltip }
      name = EVTOPTB_REP_104

I tried several codes but it doesn't work.
This one seems work only if "name" is in the line just after.

Code: Select all

call jrepl "option.*(\r\n)(.*name.*)" "$0" /jmatch /m /f input.txt /o output.txt

---------
@Aacini I found several posts about extract , remove or replace line #n but i would need to add a line. For example, a first line and/or a last line in a file.

[foxidrive]

I have a new question. Extract from input.txt :

Code: Select all

   option = {

      prestige = -20
      custom_tooltip = { text = no_own_ball_tooltip }
      name = EVTOPTB_REP_104
      wealth = -100
      narrative_event = { id = REP.105 days = 14 tooltip = EVTTOOLTIP_REP_105 } 

I wish to extract all characters between "option =" and the end of the first following line containing "name ="

With Aacini's findrepl.bat

Code: Select all

call findrepl "option =" /e:"name.*" <"input.txt"

This uses a native Windows batch script called Findrepl.bat written by Aacini, which uses jscript to make it robust and swift.
viewtopic.php?f=3&t=4697

Place Findrepl.bat in the same folder as the batch file, or in a folder that is on the system path.

There is also copy on Dropbox (unblock it after downloading):
https://www.dropbox.com/s/rfdldmcb6vwi9xc/findrepl.bat

[dbenham]

I wish to extract all characters between "option =" and the end of the first following line containing "name ="

You need to bone up on your regular expression knowledge.

Code: Select all

call jrepl "^[ \t]*option =[\s\S]*?^\s*name =.*$" $0 /m /jmatch /f input.txt /o output.txt

Dave Benham

[dbenham]

Here is version 3.8 with the following bug fixes.

1) Given certain options, I had some unintended global variables that could interfere with user defined JScript variables: i, libs, lib, and rtn. I moved them behind the _g opaque object.

2) Work around the %~f0 bug that crops up when a command is quoted

3) Substitute variable /OPTIONS for OPTIONS in the option parser so that it is unlikely to interfere with /V option variables.

JREPL3.8.zip

(9.01 KiB) Downloaded 592 times

Dave Benham

[foxidrive]

Nice one.

[vincs2]

Hello thank you for this amazing work, I have the following question :
In a text file I have lines like that :

Code: Select all

C\MyFolder\Folder\File 31549115
C\MyFolder\Folder\File2 31544445
C\MyFolder\Folder\File3 31545555

and would like to look through the file and erase all the "C\MyFolder\" part to obtain something like that :

Code: Select all

Folder\File 31549115
Folder\File3 1549115
Folder\File2 31544445

Is that possible with your script?
Thanks !

[vincs2]

Hello thank you for this amazing work, I have a question :

I have a txt file with paths of files and their size at every lines similar to this :

Code: Select all

\MyFolder\Folder\File01 31549115 
\MyFolder\Folder\File02 22512271  
\MyFolder\Folder\File03 19837047
 

I want to delete the \MyFolder\ part that it look like the following :

Code: Select all

Folder\File01 31549115 
Folder\File02 22512271 
Folder\File03 19837047

Is it possible with your script ?

Thank you

[foxidrive]

This works with your sample text, and is very literal with what you've shown. It can be made more general, but it would be useful to give a sample of actual text format, if it is different.

The - means to overwrite the original file, so use a sample file to test it with.

Code: Select all

call jrepl "^..*?\\" "" /f "textfile.txt" /o -

[catalinnc]

@dbenham

1st thanks a lot for this very useful script...

i need a little help...

please, be kind and provide a line that will replace this characters with a space...

Code: Select all

=
%
"
^
!
<
>
&
|

if this is posible all in one line it will be perfect...

input file is "input.txt" and output file is "output.txt"...

thanks a lot...
_