Page 1 of 1

Trying to batch replace with unicode characters.

Posted: Mon Jan 31, 2005 5:32 pm
by nakashian
I'm trying to use Advanced Find and Replace in Batch Replace mode to replace unicode characters with non-unicode characters, I tried what the help documentation shows \x{0543} but I get a popup dialog error
"search For" partHex Code After \x is Too Big (pos 6)

\x{nnnn} seems to work in the search field in Batch Replace mode on hex number as large as \x{00FF}.

could use some help.

Re: Trying to batch replace with unicode characters.

Posted: Mon Jan 31, 2005 5:32 pm
by Abacre
nakashian wrote:\x{nnnn} seems to work in the search field in Batch Replace mode on hex number as large as \x{00FF}.

Yes, it's an issue. We will correct it next week.

Posted: Tue Feb 15, 2005 7:26 am
by nakashian
Has any patch code been release on this bug yet?

/Pierre

Posted: Tue Feb 15, 2005 9:12 pm
by Abacre
Hi Pierre,

Yes, it's fixed now.

Please download the latest version of the program:
http://www.abacre.com/download/afrexe.zip

This is only updated EXE file. You need to extract executable file from
zip archive into folder where the program is installed. So you will
replace old executable file by this new one.

Tell me does it work now?

Posted: Thu Feb 17, 2005 3:31 am
by nakashian
Thanks for the quick response, I tried it out with the "Use regular expressions" check box selected, I still get the same error

"Search For" partHex Code After \x Is Too Big (pos 6)
Please, correct syntax of this regular expression"

in the search for I had \x{0578}

This error is occuring before the file is even read.

The file I dowloaded had 487,936 bytes, larger than my original AFR.exe
487,424;
When I start the new executable it still reports the same version 2.2 as my current version, but that may not mean anything much.

/Pierre

Posted: Thu Feb 17, 2005 11:09 am
by Abacre
In last post we corrected it in replace with part, but not in search for part.
Please download again:
http://www.abacre.com/download/afrexe.zip

Posted: Fri Feb 18, 2005 4:58 am
by nakashian
Thanks again for the response. I tried it again. I think I'm running into a different problem now. The file I tried to run AFR against was UTF-8 format, I think UTF-8 files have some characters that are 1 byte, and some characters that are 2 bytes (like my character \x{0578}). The resulting file I had didn't seem to get substitutions but rather the 2 byte unicode characters were split to 2 1 byte characters (doubling number of characters in the text file, where the words don't quite make sense). I think I'll wait for future release of AFR if it will support reading UTF-8 files.

Thanks for the help
/Pierre

Posted: Fri Feb 18, 2005 12:54 pm
by Abacre
Could you send us a sample file having \x{0578} chars to support@abacre.com?