Regex working in wizard but not during conversion

[madbologna]

When I test many regexes with the regex builder wizard, they work perfectly, but when I execute them, they often aren't functioning. The only way I've been able to get them to work is by breaking the regex up into very small pieces.

Here's how the regex would look ideally

<i>Book title Here</i>,  Page [\d]+<br>
.1990 PUBLISHING COMPANY, Revised 7/99, www.website.org<br>
<hr/>

It matches correctly for every occurence in the wizard. Each match spans three lines. Input format is pdf, output format mobi.

If I try this regex when converting, I get no matches and the text remains.

Has anyone seen this before?

[kovidgoyal]

you can always use the debug pipeline section of the conversion dialog to see exactly what html the conversion system sees, to figure outwhy the regex is not working.

[madbologna]

Originally Posted by kovidgoyal

you can always use the debug pipeline section of the conversion dialog to see exactly what html the conversion system sees, to figure outwhy the regex is not working.

Thanks so much, that did the trick! There were unprintable characters that were screwing up the matching. Unfortunately the wizard ignores them but the actual processor does not.