Re: patch to be applied to default or stable ?

[John W. Eaton]

On 17-Jan-2012, Ben Abbott wrote:

| On Jan 17, 2012, at 9:14 AM, John W. Eaton wrote:
| 
| > On 17-Jan-2012, Ben Abbott wrote:
| > 
| > | 
| > | On Jan 15, 2012, at 10:12 PM, Ben Abbott wrote:
| > | 
| > | > On Jan 15, 2012, at 8:28 PM, "John W. Eaton" <address@hidden> wrote:
| > | > 
| > | >> On 15-Jan-2012, Ben Abbott wrote:
| > | >> 
| > | >> | I've fixed a bug where bit map output was being saved in the 
improper size.
| > | >> | 
| > | >> |    https://savannah.gnu.org/bugs/index.php?35307
| > | >> | 
| > | >> | Its not pushed yet, as I'm not sure which branch it should be 
applied to.
| > | >> | 
| > | >> | If you don't have a preference, I'll push it to default.
| > | >> 
| > | >> Did this feature work correctly in previous versions of Octave, or has
| > | >> this bug always been present?
| > | >> 
| > | >> jwe
| > | > 
| > | > This bug existed for the entire 3.4.x series. It was introduced when we 
began to use gs to convert eps to bitmap formats. The 3.2.x series would have 
worked correctly.
| > | > 
| > | > Ben
| > | 
| > | I pushed to default.
| > | 
| > |   http://hg.savannah.gnu.org/hgweb/octave/rev/df695e37d404
| > 
| > I'm sorry I didn't reply earlier.  I do want to be somewhat
| > conservative about what is applied to the stable branch but this
| > change doesn't look that risky and it fixes a regression so it might
| > be best to apply it to the stable branch.  I can transplant it if you
| > agree.
| > 
| > jwe
| 
| Ok. I haven't transplanted before, so I think it best that you do it.

Looking at this today, if I do

  print test.eps '-S1024,768'

the bounding box in the eps file is set to 1024x768, but the size of
the plot does not fill the bounding box.  Instead, it is in the top
left corner of the page and occupies about 1/4 of the page area.

Is that the correct behavior?

jwe