Because the spell check has been borrowed from an app designed to write letters and documents, it is not well suited to spellchecking web form entries. For example, it doesn't very much like html or bbcode tags, which are used very regularly by bloggers, who I think are the primary target of the spellchecker. eg. <a href="http://mozilla.org">Mozilla rocks</a> [img]http://www.google.com/intl/en/images/logo.gif[/img] The spellchecker doesn't like href, http, img, url, etc. etc. etc.
Is it not better to ignore the tags as a whole? So everything between "<" and ">" and "[" and "]"
Seems like a valid request, marking NEW for investigation. It's probably not about adding all possible tags to the dictionary, though. A more reasonable solution would be to recognize markup and handle it in a smarter way, as mentioned in comment #2. It would also work across other languages. Changing dependencies as it's most likely not about adding new words to the dictionary.
This isn't on the 2.0 radar, so I'm removing it from the 2.0 tracking bug. It would indeed be nice to support different input types, but that takes far more time than I have now, and nobody else is working on the spellchecker. This bug will not block 2.0. Ancestor: why did you change the component from Spellchecker to Firefox general? Spellchecker is not Firefox-specific.
(In reply to comment #3) > Ancestor: why did you change the component from Spellchecker to Firefox > general? Spellchecker is not Firefox-specific. > Sorry about that. I got a little confused since many of spellchecking bugs are in Fx -> General. Sorry about bugspam.
could bug 339127 comment 2 help here (switching off the spell-checker if accept is text/html) ?
*** Bug 341107 has been marked as a duplicate of this bug. ***
(In reply to comment #5) > could bug 339127 comment 2 help here (switching off the spell-checker if accept > is text/html) ? I'd still like to be able to spell check words in those fields, just ignore the tags. If we just ignore things that are between < and >, spelling errors could still remain in areas of html which are designed for text, eg alt="" and title="" Ignoring all characters inside html tags is a temporary solution, but not a complete one.
You can't just ignore everything between < and >, because those characters can also be used in normal text. Just because html happens to use them doesn't mean that they always indicate html tags.
*** Bug 350391 has been marked as a duplicate of this bug. ***
2 years ago
11 months ago