Summary: | kword anchors not being saved | ||
---|---|---|---|
Product: | [Unmaintained] kword | Reporter: | Clarence Dang <dang> |
Component: | general | Assignee: | Thomas Zander <zander> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | aragorn |
Priority: | NOR | ||
Version First Reported In: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Compiled Sources | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: |
Description
Clarence Dang
2003-06-24 09:18:04 UTC
Subject: Re: New: kword anchors not being saved
On Tuesday 24 June 2003 09:18, you wrote:
> Type some text.
> Insert an image inline.
> Type some more text.
> Save the document.
> Open it back.
>
> The ANCHOR hasn't been written out to the file.
> So you get '#' in place of where the image is supposed to be and the image is no longer inline.
I can't reproduce that.
And there was no change in that area at all.
I changed libkotext in a binary-incompatible way though, so maybe you
didn't recompile kotext+kword and this leads to a wrong behaviour (wrong
virtual method being called)?
Please try recompiling koffice - does it still happen?
6 hours after cvs-clean and recompilation, it still doesn't work :( I recompiled abs everything all over again, deleted my $KDEHOME and I can still reproduce. So it's not a BIC problem. David, are you using qt-copy (QT 3.2 beta 2)? I think it may be a QT bug... Saving a file with just a single inline image, KWTextParag::save() correctly calls KWAnchor::save() (via "customItem->save( formatElem )") but "parentElem.ownerDocument().isNull()" reports "true" in KWAchor::save(), which doesn't sound right at all. Depressingly, "doc.isNull()" returns "false" but "formatElem.ownerDocument().isNull()" returns "true" (when they should both return the same thing, which should be "false" [1]), where these tests are placed right after "customItem->save". [1] Disclaimer: I generally don't use QT's XML classes because I don't have a use for them so what I'm saying may be complete rubbish... *** Bug 60337 has been marked as a duplicate of this bug. *** So it was a QT bug after all. Thanks for fixing in qt-copy, David. Everything works now. |