[vocaro]

In past versions of OmniWeb, if I browsed to a plain text file, such as this one, I could then save the file as-is to my hard drive. In 5.5b2, however, when I try to save the file, OmniWeb insists on saving it as a web archive. Needless to say, this makes the file useless to me. OmniWeb should only save HTML documents as web archives, not all documents.

[Forrest]

You can change the pulldown to "source."

[vocaro]

Originally Posted by Forrest

You can change the pulldown to "source."

Ah, forgot about that. But surely that should be the default for plain-text files...

[Forrest]

In that case, here's a couple other related threads:
http://forums.omnigroup.com/showthread.php?t=949
http://forums.omnigroup.com/showthread.php?t=989

In your case, it's a .bib file. AFAICT, that's not always text. I'm sure there's more to this.

[vocaro]

Originally Posted by Forrest

In your case, it's a .bib file. AFAICT, that's not always text. I'm sure there's more to this.

No, BibTeX files are always plain-text. Or do you mean .bib could be a completely different (possibly binary) format? Even then, I don't see how it would matter. WebArchive files are for web files only. That is, anything HTML. So, OmniWeb doesn't have to know whether a file is binary or text or whatever; it just has to check whether the file is HTML or not, and that should be easy enough.

[Forrest]

Originally Posted by vocaro

No, BibTeX files are always plain-text. Or do you mean .bib could be a completely different (possibly binary) format?

I searched for files that use the .bib extension, there are more than BibTeX.

WebArchive files are for web files only. That is, anything HTML.

There's so much more on the web than HTML.

[vocaro]

Originally Posted by Forrest

I searched for files that use the .bib extension, there are more than BibTeX.

As I stated earlier, an extension of .bib may indeed be binary, yes. Or it might be plain text. It doesn't matter. If WebKit doesn't know how to render it, then it makes no sense to save it as a WebArchive file.

Originally Posted by Forrest

There's so much more on the web than HTML.

You can find a gazillion file formats on the web, but WebArchives are intended for only one, and that's HTML. If it's not HTML, it should be saved as source, not as a WebArchive. Otherwise, you're locking up the file into a format that only WebKit knows how to read, preventing you from doing anything with it in other applications.

[Forrest]

Originally Posted by vocaro

If WebKit doesn't know how to render it, then it makes no sense to save it as a WebArchive file.

Unfortunately, that's completely subjective. I can make a .bib file that displays perfectly fine as a web page. The content/mime type of the file also comes into play.

WebArchives are intended for only one, and that's HTML.

I disagree. There are multiple file formats commonly used on the web that are very useful being saved as webarchive.

Overall, I think the saving options need some work, though I still disagree with your reasons on how it should be changed. To me it doesn't seem like the way it's supposed to be working is working (as noted in one of the posts I linked to above.) Hopefully someone who has a reproducible example of it not working correctly has sent in a bug report.

Added two new defaults to OmniWeb to track whether the user prefers to save 'simple' documents (those without sub-resources), and 'complex' documents (those with sub-resources), as webarchives, or as source. The defaults are 'OWSaveTypeForComplexDocument' and 'OWSaveTypeForSimpleDocument', and valid entries are 'webarchive' and 'source'.
These are automatically written with whichever option you choose each time you save, so that OmniWeb will remember your last choice for each save type.

[vocaro]

Originally Posted by Forrest

I disagree. There are multiple file formats commonly used on the web that are very useful being saved as webarchive.

Such as...?

[Forrest]

Originally Posted by vocaro

Such as...?

XHTML for one. Pages can also be completely comprised of JavaScript too, though that should be avoided. Flash is another one. If it's saved just as the swf, it can be missing linked files which is one of the advantages of a webarchive. SMIL is another than can be, though typically requires a plugin and that should really be embedded in HTML, but isn't always.