Am Sonntag 29 Juni 2008 17:51:18 schrieb Chris Watkins: That is possible with some regular expression magic: python replace.py -regex "(?s)foo(.*$)" "bar\\1" -page:Fubar Maybe you have to type \1 instead of \\1 if you're using Windows. You can run this: python replace.py foo bar -page:Fubar -exceptinsidetag:link -exceptinsidetag:hyperlink This will exclude wikilinks and URLs. There are some more things that can be excluded, see the source code of the method replaceExcept() in wikipedia.py (look at the exceptionRegexes dictionary). I have just added a regular expression for section headers for you, so if you're running the SVN version, you can use this parameter: -exceptinsidetag:header Cheers Daniel

Next question: If I have a set of synonyms in the wiki, and they redirect to the same page (say, cow and cattle), is it possible to find the first time any of these terms occur on a page, and link only that? So cow would be changed to [[cow]], or cattle to [[cattle]], but not both on the same page, if both terms occur. Many thanks Daniel, this has been a great help - and adding the -exceptinsidetag:header to regex thanks to my request is very good of you. I have been doing test runs (without saving, as my bot has a problem at the moment) and I am able to successfully use all these terms that you gave. This is wonderful. Chris On Mon, Jun 30, 2008 at 3:11 PM, Daniel Herding <DHerding(a)gmx.de> wrote: -- Chris Watkins (a.k.a. Chriswaterguy) Appropedia.org - Sharing knowledge to build rich, sustainable lives. Blog: chriswaterguy.livejournal.com/ Buying at Amazon, eBay etc? Start at http://appropedia.maatiam.com and a percentage of your purchase supports Appropedia - at no extra cost. Where men are the most sure and arrogant, they are commonly the most mistaken, and have there given reins to passion, without that proper deliberation and suspense, which can alone secure them from the grossest absurdities. -- David Hume