Upgrade StorageSnapshot encoding
Are you guys (Dan, Simon, etc.) doing something with the character encoding of snapshotted files?
I ask because if I go to this page it displays correctly. But the snapshotted page does not, unless I change the viewed character encoding to utf-8.
So based on that I am guessing that somehow the encoding gets changed, but not the declaration?
I ask because if I go to this page it displays correctly. But the snapshotted page does not, unless I change the viewed character encoding to utf-8.
So based on that I am guessing that somehow the encoding gets changed, but not the declaration?
In Beta 4 we're now using WebPageDump (based on ScrapBook code) to save snapshots. WebPageDump converts everything to UTF-8 and changes the META Content-Type accordingly.
The standard Mozilla saveDocument() functionality used in Beta 3 saves the page as is. That particular page is UTF-8 but has "charset=iso-8859-1" in its META Content-Type declaration. When pulled over the wire, Mozilla uses the Content-Type header from the server, which is set to UTF-8. When loading the page locally it uses the META declaration, which is incorrect and will cause the page to display incorrectly. You should get the same results using File -> Save Page As....
Re-save in Beta 4 and it should look right. This is one of the reasons we switched to WebPageDump.