TextCrawler VERY slow in batch (binary mode) vs manual

[rodp]

Hi All,

Please could someone help. I have 50 files or so I need to run through textcrawler batch process. It's a complex 8 step process and has to cater for various scenarios. When I process it manually, using the 'regular expression' tab within textcrawler, each of the 8 steps are done pretty quick. However, when I go to use the batch method on the same file, it takes such a long time - we're talking 30 - 45 mins!!

I've attached the files for someone else to have a go or if possible could the developer please take a look.

I've tried rebooting and am using version 2.2. It seems to be slightly quicker when not in binary mode but unfortunately, I need to use binary mode or else it messes up the file.

Many thanks in advance

RodP

[DigitalVolcano]

Hi
The batch mode replace is performing a replacement count (for the conformation popup) for each step. The problem is that it is doing this count for each operation on the complete original file, not the changed/replaced file as it would be at that particular step. As it is running the later regex steps on the complete file rather than the trimmed down one this is killing performance. (also the estimated replacement count is totally wrong)

A workaround for this is to disable the 'Display number of replacements and prompt for confirmation' checkbox in the Options->Replace screen. This should now run fast.

I'll probably need to fix this by removing the replace counts prompt for batch mode.

hope this helps!