[GRASSLIST:10576] Re: Labels and ps.map

[Michael Barton]

The text layer in the new GIS Manager (start with gis.m&) is drawn by the
TclTk canvas so that it is postscript text in eps output. So it might be
what you are looking for. However, it is only for single labels, manually
placed.

Michael
__________________________________________
Michael Barton, Professor of Anthropology
School of Human Evolution and Social Change
Arizona State University
Tempe, AZ 85287-2402

phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton



> From: Nick Cahill <ndcahill at facstaff.wisc.edu>
> Date: Fri, 24 Feb 2006 15:15:37 +0200
> To: Multiple recipients of list <GRASSLIST at baylor.edu>
> Subject: [GRASSLIST:10550] Re: Labels and ps.map
> 
> Dear Michael,
> 
> Yes, I'm talking about the quick-and-dirty display labels from
> d.vect. These are incredibly useful, as you point out, and again,
> very intuitive; they are an attribute of the vector file rather than
> some separate file which you have to remember about. Thank you for
> the test images. However, as you point out, they are bitmaps; and I'm
> looking for postscript output which will allow me to move and edit
> the labels for a published map. Ps.map does that just fine. I would
> still wish that ps.map could draw directly on the vector attributes
> for drawing labels, like gis.m and g.m do, rather than going through
> an external label file.
> 
> Best wishes,
> 
> Nick
> 
> 
> 
> On Feb 24, 2006, at 9:18 AM, Michael Barton wrote:
> 
>> Nick,
>> 
>> I see that you are talking about the display labels created by
>> d.vect--rather than those created by d.labels.
>> 
>> The d.vect labels are a 'quick and dirty' vector labeler, but that is
>> exactly what is needed at times.
>> 
>> I just did a test with the new gis.m and attach 3 small files here.
>> In two
>> cases (making the eps and pdf), I used the print function dialog.
>> In the
>> third case (making a png), I used the save output function.
>> 
>> As you can see, there are labels in all 3 images.
>> 
>> Michael
>> __________________________________________
>> Michael Barton, Professor of Anthropology
>> School of Human Evolution and Social Change
>> Arizona State University
>> Tempe, AZ 85287-2402
>> 
>> phone: 480-965-6213
>> fax: 480-965-7671
>> www: http://www.public.asu.edu/~cmbarton
>> 
>> 
>> 
>>> From: Nick Cahill <ndcahill at facstaff.wisc.edu>
>>> Date: Fri, 24 Feb 2006 07:34:35 +0200
>>> To: Multiple recipients of list <GRASSLIST at baylor.edu>
>>> Subject: [GRASSLIST:10541] Re: Labels and ps.map
>>> 
>>> Dear Michael,
>>> 
>>> In d.m and gis.m, there is a radio button for showing labels, you
>>> choose what column to display and its parameters, left, right, size,
>>> etc. Does this make a temorary label file for display? I was guessing
>>> that it was somehow able to draw directly from the database, since
>>> there is no place to specify a label file. This all seems very
>>> intuitive; but when printed to postscript, those labels do not appear
>>> (for me at least; because I don't have ghostscript, perhaps?). So
>>> I've been working through ps.map; which works fine, but involves the
>>> extra step of creating a label file. The postscript from ps.map comes
>>> out as proper editable text, not as a raster, which is important to
>>> me as I'll need to move the labels around later. Creating the label
>>> file is not much of a hassle, except that it requires knowing in
>>> advance how large the labels should be; you can't resize them on the
>>> fly and immediately see what the results are. I just wondered whether
>>> a more direct method was in store for the future.
>>> 
>>> Thanks,
>>> 
>>> Nick
>>> 
>>> 
>>> On Feb 24, 2006, at 6:39 AM, Michael Barton wrote:
>>> 
>>>> Nick,
>>>> 
>>>> Point vector attribute display in d.m and gis.m is done by
>>>> d.labels--with
>>>> original labels file prepared by v.label (or by hand).
>>>> 
>>>> There is a fair amount of control over the display possible.
>>>> However, the
>>>> labels ultimately are rendered to the screen as rasters. Gis.m can
>>>> output
>>>> the screen to EPS or a postscript device now (if you have
>>>> ghostscript
>>>> installed). However, it is output as a postscript bitmap/raster
>>>> image. If
>>>> you make your map display larger, it might give you higher
>>>> resolution in the
>>>> final EPS.
>>>> 
>>>> Michael
>>>> __________________________________________
>>>> Michael Barton, Professor of Anthropology
>>>> School of Human Evolution and Social Change
>>>> Arizona State University
>>>> Tempe, AZ 85287-2402
>>>> 
>>>> phone: 480-965-6213
>>>> fax: 480-965-7671
>>>> www: http://www.public.asu.edu/~cmbarton
>>>> 
>>>> 
>>>> 
>>>>> From: Nick Cahill <ndcahill at facstaff.wisc.edu>
>>>>> Date: Tue, 21 Feb 2006 20:15:03 +0200
>>>>> To: GRASSLIST international <GRASSLIST at baylor.edu>
>>>>> Subject: [GRASSLIST:10492] Labels and ps.map
>>>>> 
>>>>> Dear Grass Gurus,
>>>>> 
>>>>> Is there a reason why d.m and gis.m can display point vector
>>>>> attributes as labels, but this can't be output to a postscript
>>>>> file;
>>>>> and ps.map has to rely on the - I assume, somewhat outdated -
>>>>> method
>>>>> of using a separate label file? Currently, to get labels to come
>>>>> out
>>>>> correctly in ps.map, I have to run ps.map, look at the postscript
>>>>> output, find that the labels are a bit too big or something, make a
>>>>> whole new label file with different parameters, and start again;
>>>>> and
>>>>> this usually takes a number of iterations. It would be much more
>>>>> convenient (for me at least, as a user) to have site labels be part
>>>>> of vector display within ps.map, so you could have e.g.
>>>>> 
>>>>> vpoints vector
>>>>> type point or/and centroid
>>>>> layer # (layer number used with cats/where/sizecol options)
>>>>> cats list of categories (e.g. 1,3,5-7)
>>>>> where SQL where statement like: vlastnik = 'Cimrman'
>>>>> masked [y|n]
>>>>> color color
>>>>> fcolor color
>>>>> **attlabel (attribute column)
>>>>> **attlabelsize (in points or whatever)
>>>>> **font, color, etc.
>>>>> 
>>>>> 
>>>>> Or am I missing something obvious?
>>>>> 
>>>>> Thank you,
>>>>> 
>>>>> Nick Cahill
>>>> 
>> 
>> <labeltest_eps.pdf>
>> <labeltest_gs.pdf>
>> <labeltest_png.png>