For the Firefox add-on, do nothing if trying to toggle back to an original URL that doesn’t start “http”. This can happen if, for example, you’d typed “bbc.co.uk” on the Textise home page. Textise can deal with this sort of abbreviated URL and, in fact, so can Firefox, but for some reason it doesn’t like using an incomplete URL as a location.href.
Change the copyright to extend to 2011 (on home page).
Check and, if necessary, fix the HTTP method when coming back from the Options page. I have a suspicion that POST is used, which was originally how Textise worked. However, I’ve since changed to GET so that bookmarking is possible (i.e. the URL contains all the necessary info, including the URL of the page you’re trying to Textise).
Have a look at the code to remove images. I’ve seen a couple of Out Of Memory exception (on the Kindle, as it happens) that refer to that part of the code.
Double-check the release status of the published code – shouldn’t be DEBUG!
Have a think about removing the actual .NET error description from error messages. The error message could simply ask the user to report the URL they tried to Textise, not cut and paste the whole thing. This might make Textise seem a bit more pro (!). I might still like to see the errors, of course, just in case I can’t replicate them, so I’m also wondering about creating an error RSS feed that I can subscribe to…
Fix internal bookmarks (e.g. “<div id=’mainContent’>”. At the moment, Textise reduces these to just “<div>” because sometimes they also contain formatting. If I could keep just the bookmark info (the “id=” bit) it would really help with sites that have reams of links at the top of each page. Might slow the program down somewhat, though, as it’s another trawl through the whole page.