[cairo] postscript backend cut my text

[Charles Doutriaux]

pdf,svg,png work great

Just in case I'm also attaching the full C code (not just the snipet) 
but the snipet shows the bug really well and the good and bad ps 
geneterated (test_1.6.4.ps and test_1.7.6.ps ) Notice that there is a 
lot more XXXX in the 1.6.4 as if 1.7.6 was cutting the output for 
portrait orientation

C.


Behdad Esfahbod wrote:
> Charles Doutriaux wrote:
>   
>> Doh!!!
>> I'm sorry I forgot I rebuild with 1.6.4 last night.....
>>
>> Ok the snipet I sent yesterday might have missed a cairo_save (i 
>> probably forgot a copy/paste) but it is definately still missing a bunch 
>> of XXXX
>> when linking against 1.76.  and it works fine against 1.6.4 so i'll be 
>> suing this for now
>>     
>
> Can you send us before/after PS files?  Also, what happens if you generate PDF
> or SVG instead?  What about image?  Three of us tried the code you sent and
> couldn't reproduce the problem.
>
> behdad
>
>
>   
>> hope this helps, and again sorry for the false hapy ending :(
>>
>> For now i'm going to revert 1.6.4 for our stuff
>>
>> thanks,
>>
>> C.
>>
>>
>> Charles Doutriaux wrote:
>>     
>>> Hi Chris,
>>>
>>> Yes! that was it, adding a cairo_save(cr) right after creating the 
>>> surface fixed it!
>>>
>>> Hum.. I'm not sure I understand why I need to do a save before i do 
>>> anything else (since there's nothing on the page yet) but hey it works!
>>>
>>> Thanks a lot,
>>>
>>> C.
>>>
>>> PS: Thanks also for the pointer to cairo_trace
>>>
>>> C.
>>>
>>> Chris Wilson wrote:
>>>   
>>>       
>>>> On Tue, 2008-09-23 at 15:38 -0700, Charles Doutriaux wrote:
>>>>   
>>>>     
>>>>         
>>>>> I'm running inot a weird feature under cairo 1.7.6 and 1.7.2  (don't 
>>>>> think it was in 1.6.4)
>>>>>
>>>>> Considering the attached piece of code, this used to work in previous 
>>>>> version
>>>>>     
>>>>>       
>>>>>           
>>>> Behdad spotted that the snippet was missing an initial cairo_save().
>>>> After adding that, the output looks like the axes for a graph, and
>>>> importantly look the same for the image and PS backends, using both
>>>> cairo-1.6 and master. This is the first hint of a regression with 1.7 so
>>>> its important for us that we are able to resolve it before 1.8 (which
>>>> Behdad keeps insisting is tomorrow ;-) If you please can verify the test
>>>> case and also grab a couple of screenshots to serve as a reference.
>>>>
>>>> <plug>cairo-trace is a very useful tool for capture/replay of bugs:
>>>> http://   cgit.freedesktop.org/~ickle/cairo-trace/</plug>
>>>>   
>>>>     
>>>>         
>>> _______________________________________________
>>> cairo mailing list
>>> cairo at cairographics.org
>>> http://  lists.cairographics.org/mailman/listinfo/cairo
>>>
>>>
>>>   
>>>       
>> _______________________________________________
>> cairo mailing list
>> cairo at cairographics.org
>> http:// lists.cairographics.org/mailman/listinfo/cairo
>>
>>     
>
>
>   

-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_cairo_1.7.6.c
Type: text/x-csrc
Size: 92607 bytes
Desc: not available
Url : http://lists.cairographics.org/archives/cairo/attachments/20080924/ddcf0f99/attachment-0001.c 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.1.7.6.ps
Type: application/postscript
Size: 46401 bytes
Desc: not available
Url : http://lists.cairographics.org/archives/cairo/attachments/20080924/ddcf0f99/attachment-0002.ps 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.1.6.4.ps
Type: application/postscript
Size: 57019 bytes
Desc: not available
Url : http://lists.cairographics.org/archives/cairo/attachments/20080924/ddcf0f99/attachment-0003.ps