Strangely I can't find any documentation on this...
Does the browser cache XSL files? Does it just cache the resulting transformation? What about the url's within the xsl file pointing to other xml files? Does anyone know any links with details on this?
07-13-2008, 05:58 PM
actually it doesn't matter what the content is all that matters for caching is if u are using the right http-verb (GET) and sending the correct http headers for caching
07-13-2008, 06:14 PM
At least with Internet Explorer, Opera and Firefox 2 (Firefox 3 does not support well XSLT).
In most cases you can work offline smoothly once the file has been loaded.
(Firefox 3 does not support well XSLT).
Again, that is false. Post an example in the other thread and we can figure out what is wrong.
After playing with it a bit, it does seem to cache everything regardless of its depth within the files. (Quite amazing speed improvement I might add).
As for the xsl issues I've seen with firefox 3, I think its due to ignorance of the security changes:
<?xml-stylesheet type="text/xsl" href="Foo.xsl"?>
<?xml-stylesheet type="text/xsl" href="sub/Foo.xsl"?>
This fails if on the client machine:
<?xml-stylesheet type="text/xsl" href="../Foo.xsl"?>
<?xml-stylesheet type="text/xsl" href="../sub/Foo.xsl"?>
Which is obviously due to the disallowing of relative URLs:
07-14-2008, 05:48 PM
TNO: No, that may be a very relevant piece of information, worth filing a bug on bugzilla.mozilla.org about.
07-16-2008, 01:13 AM
Alex: The bug is here https://bugzilla.mozilla.org/show_bug.cgi?id=230606
It has been a big discussion over the years but somehow it was decided to leave the restriction by default.
jkd: As you can see it was not false at all, you can also read the discussion in bugzilla to realise that from the beginning my point of view was shared with many others. I may suggest to think instead of field before make a disqualification, the informatique is a technique, not a religion.