Re: true landscape orientation

From: Rob Nicholas <rnicholas_at_nyahnyahspammersnyahnyah>
Date: Wed, 14 Jan 2009 13:45:53 -0800

Fred and Mary,

I also get portrait orientation using Preview.app and gs (on my Mac).
However, using gv (on a Linux machine), I get landscape orientation
(actually, it's probably "seascape" but I can't tell for sure). I can
get a little further (correct in Preview.app, upside down in gv)
making simple text "slides" in LaTeX; perhaps some of the viewers look
at the dominant text direction to choose an orientation.

Looks like there may not be a good solution to this (admittedly minor)
problem -- thanks for taking the time to look into it, though.

~Rob

On Sun, Jan 11, 2009 at 3:02 PM, Fred Clare <fredclare_at_mac.com> wrote:
> Rob,
>
> Thanks for your input. In looking at my Postscript Reference
> Manual your observation about inserting a Postscript comment:
>
> %%Orientation: Landscape
>
> in the initial comments section of a Postscript file should
> indeed afford interpreters the ability to position the plots
> as you desire. However, neither of the Postscript viewers I am using
> on my Mac ("gs" and "preview") seems to respond to having this comment
> in a file. Let me know what you get with the trivial Postscript
> file attached. Given your experience with GV it would appear
> that Postscript interpreters are not rock solid in their honoring
> the orientation comment. :-) How would you expect a Postscript
> printer to react to the landscape orientation comment?
>
> I could easily insert an orientation comment in the lower level
> Postscript driver when a Postscript landscape workstation
> is opened. I think things get a little tricky when considering
> Postscript created from NCL. For example, when the gsnMaximize
> resource is set to True, then the resource gsnPaperOrientation can
> be applied on a frame-by-frame basis. To have orientation
> comments in the output Postscript properly reflect this whould
> require inserting %%PageOrientation comments at the Postscript
> page level. Effecting this at the NCL level would require
> implementing a function call for the Postscript driver to
> respond to, and NCL would have to make that call where appropriate.
> Assuming that we can settle for a global orientation comment,
> there are still considerations for NCL. Since the resource
> gsnMaximize can be set on a frame-by-frame basis, and that
> may or may not result in an individual frame being produced
> in either landscape or portrait mode, we are back in the same
> situation as with the above considerations for gsnPaperOrientation.
>
> I see nothing in my PDF Reference Manual the would be the
> equivalent to the Postscript orientation comment. But my
> manual is eight years old, so maybe there has been an addition.
>
> Before proceding, I will wait for Mary to weigh in, since
> she knows the details of how NCL interfaces with the the lower
> level codes in this area.
>
> Regards,
>
> Fred Clare
>
>
>
>
>
>
>
> On Jan 9, 2009, at 6:32 PM, Rob Nicholas wrote:
>
>> Adding "%%Orientation: Landscape" to the header of the PostScript file
>> seems to get you part way there -- GV displays the page in landscape
>> mode but upside down.
>>
>>
>> On Fri, Jan 9, 2009 at 12:14 PM, Fred Clare <fredclare_at_mac.com> wrote:
>>>
>>> On Jan 9, 2009, at 12:36 PM, Rob Nicholas wrote:
>>>
>>>> Mary,
>>>>
>>>> On Fri, Jan 9, 2009 at 8:18 AM, Mary Haley <haley_at_ucar.edu> wrote:
>>>>>
>>>>> Maybe there's a flag that one can put in a PS/PDF file that tells
>>>>> other PS/PDF viewing applications that the image has been rotated, and
>>>>> hence it should automatically show it in landscape mode. I'll ask our
>>>>> PS/PDF expert about this.
>>>>
>>>> I'm almost certain this is true -- most other applications I've used
>>>> that produce PostScript output seem to get this right. Unless it's PDF
>>>> (rather than PostScript) I'm thinking of (although the problem exists
>>>> with NCL's PDF output, too).
>>>
>>> I know of no Postscript or PDF operator that would specifically
>>> indicate to a viewing application to view the image as being rotated.
>>> Each language has a rotate operator, but that simply rotates the
>>> image within the context of the internal coordinate systems.
>>> Perhaps you could supply us with an example file that seems to
>>> have this property and I could look in there to see what is being
>>> used.
>>>
>>> Fred Clare
>>>
>>>>
>>>>
>>>> On Fri, Jan 9, 2009 at 8:18 AM, Mary Haley <haley_at_ucar.edu> wrote:
>>>>>
>>>>> For now, can you use the "Rotate" button on your viewer to rotate
>>>>> the image as desired?
>>>>
>>>> Sure, though it gets to be a bit of a chore when you're producing 20
>>>> or 50 plots at a time. Let me know if you unearth a solution.
>>>>
>>>> Thanks,
>>>> Rob
>>>> _______________________________________________
>>>> ncl-talk mailing list
>>>> List instructions, subscriber options, unsubscribe:
>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>>
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
>
Received on Wed Jan 14 2009 - 14:45:53 MST

This archive was generated by hypermail 2.2.0 : Fri Jan 16 2009 - 14:05:56 MST