Load this document (from bug#38619) - the charts on page 9 end up with some (massive) scaling issue - the clipping region luckily works well, but the text and objects inside the WMF are the wrong size.
The same issue appears to be a problem for:
Created attachment 77121 [details]
EMF embedded in the second attachment
Timur - how did you generate the 2nd attachment here ? :-)
I just fixed bug#62834 - and would -love- to create an automated regression / unit test for this.
Unfortunately, that document is a bit big to check into git - I'll also need another one for the scaling issue that is identified there.
Any chance of a small document - preferably a table inside an EMF with two entries: "Hello" and "World" (or something) that exhibits the same bug#62834 - so that I can write a unit test harness for this sort of thing.
I've tried generating that with LibreOffice, but everything works - so it must be related to the unusual structure of the meta-file :-)
(In reply to comment #2)
I don't know, I received it. What I know, it's source is Oracle database, but I don't know whether it's an export or just some printscreen. I asked the author, hopefully he will reply.
Created attachment 77152 [details]
document created by the same procedure, but seems OK and prints fine
The explanation is: table from Excel is pasted in Word, with Paste Special..as Picture (EMF), maybe resized after that.
Attached document is created by the same procedure, but it seems OK.
So, we stay we the original document, which is 30KB.
I'd like to put this in MAB 4.0 unless someone objects?
sounds reasonable to me.
This is fixed in master - it may well be down to:
Author: Radek Doulik <email@example.com>
Date: Mon Apr 29 00:00:00 2013 +0200
Fix bnc#795857 Use correct sizes for EMF+ bitmap rendering
Fix pdf export wrong size issues for embedded EMF+ images.
I'm looking at back-porting that fix back to -4-1 - just doing a test build.
Good news. Will it be in 22.214.171.124?
Although this discussion doesn't belong here, this is the first time I noticed that there are also bnc, i, lp, rhbz apart from fdo in fixed bugs. It's really interesting, I haven't seen it explained somewhere.
> Good news. Will it be in 126.96.36.199?
If that's in tripple review mode now, I suspect not - that's a lot of resource to ask.
> Although this discussion doesn't belong here, this is the first time I
> noticed that there are also bnc, i, lp, rhbz apart from fdo in fixed
> bugs. It's really interesting, I haven't seen it explained somewhere.
These are all abbreviations; bugzilla.novell.com, redhat bugzilla, issues (from OO) and FreeDesktop.org etc. :-)
Will this be backported to 4.x? There is no target now. Thank you.
Sure it is in 4.0 - should be in 4.0.5 - but play with a snapshot I guess :-)
This is not in the list for 188.8.131.52 and also not in 184.108.40.206 (which doesn't seems complete).