Re: [Qgis-developer] processing.runalg() issue with grass algorithms

[Eric Goddard]

Thanks, I knew they returned dictionaries, but I forgot to change the
call to runalg to have lpf as the output argument. The script executes
with the SAGA algorithm.



On Tue, Mar 25, 2014 at 5:25 PM, Victor Olaya  wrote:
> The runalg method does not return a layer. It returns a dict with the
> outputs of the algorithms. Keys are output names, values are paths to
> output objects
>
> Hope this helps
>
> 2014-03-25 22:36 GMT+01:00 Eric Goddard :
>> Alright, latest master (9e42c5b), with processing version
>> 2.0-20131120. Running a custom tool that consists of the following
>> code results in an error:
>>
>> ##dem=raster
>> ##radius=number 4
>> ##lpf=output raster
>>
>> lpf = processing.runalg("saga:simplefilter", dem, 0, 0, radius, None)
>>
>>
>> An error has occured while executing Python code:
>>
>> Traceback (most recent call last):
>>   File 
>> "/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
>> line 256, in accept
>> self.finish()
>>   File 
>> "/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
>> line 281, in finish
>> Postprocessing.handleAlgorithmResults(self.alg, self, not keepOpen)
>>   File "/home/eric/.qgis2/python/plugins/processing/gui/Postprocessing.py",
>> line 77, in handleAlgorithmResults
>> dlg = CouldNotLoadResultsDialog(wrongLayers, alg)
>>   File 
>> "/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
>> line 41, in __init__
>> self.setupUi()
>>   File 
>> "/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
>> line 50, in setupUi
>> html = self.alg.getPostProcessingErrorMessage(self.wrongLayers)
>>   File "/home/eric/.qgis2/python/plugins/processing/core/GeoAlgorithm.py",
>> line 520, in getPostProcessingErrorMessage
>> + layer.value + '\n'
>> TypeError: cannot concatenate 'str' and 'dict' objects
>>
>>
>> Python version:
>> 2.7.6 (default, Feb 26 2014, 12:07:17)
>> [GCC 4.8.2 20140206 (prerelease)]
>>
>>
>> QGIS version:
>> 2.3.0-Master Master, 9e42c5b
>>
>> Python path: ['/home/eric/.qgis2/python/plugins/processing',
>> '/usr/share/qgis/python', u'/home/eric/.qgis2/python',
>> u'/home/eric/.qgis2/python/plugins', '/usr/share/qgis/python/plugins',
>> '/usr/lib/python27.zip', '/usr/lib/python2.7',
>> '/usr/lib/python2.7/plat-linux2', '/usr/lib/python2.7/lib-tk',
>> '/usr/lib/python2.7/lib-old', '/usr/lib/python2.7/lib-dynload',
>> '/usr/lib/python2.7/site-packages',
>> '/usr/lib/python2.7/site-packages/gtk-2.0',
>> '/usr/lib/python2.7/site-packages/wx-3.0-gtk2',
>> '/usr/share/qgis/python/plugins/fTools/tools', '/home/eric/Documents']
>>
>>
>> Running the code in the console completes successfully. However, I'm
>> still unable to run a grass algorithm when the grass_region_parameter
>> and grass_region_cellsize_parameter are set to None.
>>
>> Thanks,
>> Eric
>>
>> On Mon, Mar 24, 2014 at 10:13 PM, Eric Goddard  
>> wrote:
>>> Oops, my mistake. I have ##lrm=output raster in the header, but lpf = ... Is
>>> the only line in my code.
>>>
>>> I don't get the error message after changing lrm to lpf in the header. I do
>>> get another error message, but I think I broke my processing by trying to
>>> force an update. I'll test it tomorrow. Sorry about the confusion!
>>>
>>> Eric
>>>
>>> On Mar 22, 2014 5:06 PM, "Eric Goddard"  wrote:

 Ah, I didn't know a file could he specified without a path. Thanks for the
 clarification!

 On Mar 22, 2014 4:37 PM, "Victor Olaya"  wrote:
>
> > If I run the same line in the python console, it works correctly:
>  radius = 4
>  lpf = processing.runalg("saga:simplefilter", "SR_50M_alaska_nad", 0,
>  0, radius, None)
>  processing.load(lpf['RESULT'])
> >
> > Results in the map being loaded.
>
> Not sure why this is happening. I will try to reproduce the error and
> see if it can be fixed
> Thanks for the report
>
> >
> > Also, if I change the output folder in the processing settings to a
> > different location (in this case, tmp/processing in my home folder)
> > and run another algorithm in the python console, the RESULT key still
> > points to /tmp/processing/...
>
> This is not a bug. If you do not enter a filename, the output will
> always go to a temporary file. The output folder that yu can set in
> the settings  is the folder to be used when you enter a filename
> without a path (like "myfile.shp", but not "/home/myuser/myfile.shp"),
> not the folder to use for temporary files.
>
> Thanks again!
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Vienna hackfest: QGIS Legend discussion

[Nyall Dawson]

On 26 March 2014 11:07, Olivier Dalang  wrote:

>
> 3. Display labels/diagrams/actions/etc. as icons just after the layer's
> name
> Allow for quick toggling of those frequently useful features.
> Right clicking allow for related actions (edit, set label expression, etc.)
>
>
A big +1 to this from me. Having an immediately accessible label toggle
checkbox in the layer control is one of the few things MapInfo does right...

Nyall
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Labeling unit tests and bug fixing

[Larry Shaffer]

Hi,

I've mostly finished the labeling unit test framework (at least for *nix).
You can read about the setup in its README [0] and about the local test
server [1], if interested.

I'm happy to report that there is considerably good parity between all
outputs now (at least from my testing on Mac), especially since the advent
of mulit-threaded rendering. However, there are only a few actual unit
tests right now. The only outstanding issues revolve around too much
antialiasing in test renders.

There is still some CMake coding to do, adding modules for the now-required
utilities when ENABLE_TESTS=ON. These are currently:

* lighttpd (local HTTP server)
* spawn-fcgi (qgis_mapserv.fcgi spawner)
* poppler, with cairo support (composer PDF -> image rasterizer, for
comparisons)

There still needs config files set up for lighttpd/spawn-fcgi on Windows.

I'll be triaging labeling bugs and trying to fix them, while ensuring
regression tests are in place and passing. Then I'll add more unit tests
for general API stuff.

Lastly, I will be putting forth a proposal concerning a redesign of the
labeling classes, which includes rule- and style-based labeling. That will
come after an effort on the previously noted bug fixing.

Any help with the bug squashing (or anything related to this stuff) would
be greatly appreciated. :-)

[0] https://github.com/qgis/QGIS/tree/master/tests/testdata/labeling
[1]
http://osgeo-org.1560.x6.nabble.com/New-Python-QgsLocalServer-and-standard-fonts-for-testing-td5104818.html

Regards,

Larry Shaffer
Dakota Cartography
Black Hills, South Dakota
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Vienna hackfest: QGIS Legend discussion

[Olivier Dalang]

Hi !

Great news !
(by the way, I'm crossposting to qgis-ux as well)

I've been thinking about this a little and have some more long term ideas I
want to share (see the mockup below, it may be more clear):

1. Trivalent checkboxes (neutral, visible, hidden) (#5787)
When neutral, a layer/group is shown if it's parent is shown (top level
neutral elements are shown).
When hidden, a layer/group is hidden regardless of its parents.
When visible, a layer/group is shown regardless of its parents.

This would make big groups much more usable. Cinema4D's object manager
features this, and its very efficient. Also Qt's checkboxes can be tristate
out of the box, so it may not be that hard to implement.


2. Display the data source as an icon in a column on the left (not
indented, separated by a thin vertical line)
The icon would help to know whether it's a WMS, Shapefile, postgis, etc.
layer.
The icon would be grayed out when the source is unavailable.
Right clicking on the source would only source related actions (subset,
relink source, save data as, open in db manager, ...).

It would make the distinction between layer and source much clearer. This
distinction is obvious for experienced GIS users, but I've seen a lot of
beginners being very confused because of that (software working with
external links/files are quite rare for common usage). We could also
display the "edit pencil" on the source, giving one more hint to the user
that he's not modifying the QGIS file, but the datasource behind.


3. Display labels/diagrams/actions/etc. as icons just after the layer's name
Allow for quick toggling of those frequently useful features.
Right clicking allow for related actions (edit, set label expression, etc.)
Imagine what plugins could add there !! Any per-layer setting could be
shown as a tag. Think of Anita's TimeManager or Minoru's qgis2ThreeJs... We
could put the snap settings there too, or new features like alternative
styles, saved feature selections, snap settings,  !!!

Again, this is how Cinema4D's manager works (they call those icons "tags",
and C4D makes extensive use of them). In their version, they align all the
tags after another vertical bar, which may be the most elegant way to solve
the layout with long layer's names.


Conclusion.

I think it's worth making the Legend GUI a bit more rich than it is now.
This is the part of the GUI where the users spend most of their clicks, and
IMO the potential efficiency bonus is by far worth the small
complexification. And it's not only complexification, it's also
clarification.
The segregation of source/layer/addendi in the legend GUI will be very
powerful in combination with contextual menus. By having three different
contextual menus, we'll be able to have much more actions at finger tips,
without over-crowding the menus.




About Nyall's idea :
+1 !
It could be implemented as it is in Photoshop, i.e. group's blending mode
is "transfer" by default, which simply draws the children as if there was
no group. But setting the group's blending mode to anything else
("normal","multiply",...) composites the group as one layer, and blends
that composition.
Having this in the legend UI rather than in the layer style would also make
the UI much clearer. Right now, the UI is a bit confusing because the
transparency at layer level, feature level, and at color alpha level are
almost at the same place.
In general, using Photoshop as a reference for the UI in regard of
layers/blending modes makes much sense since PS is the leader in that
matter (as a remainder, they just have a drop-down menu at the top of the
layer panel which applies to the selected layer).


What do you think ?
If the community likes the principles, I'd be happy to work on the mockup a
bit more (by integrating Nydall's suggestion), so we can have a common
reference for further development.

Anyways, I'm very happy to know some improvements are in preparation !


Kind regards,

Olivier


[image: Images intégrées 1]
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] processing.runalg() issue with grass algorithms

[Victor Olaya]

The runalg method does not return a layer. It returns a dict with the
outputs of the algorithms. Keys are output names, values are paths to
output objects

Hope this helps

2014-03-25 22:36 GMT+01:00 Eric Goddard :
> Alright, latest master (9e42c5b), with processing version
> 2.0-20131120. Running a custom tool that consists of the following
> code results in an error:
>
> ##dem=raster
> ##radius=number 4
> ##lpf=output raster
>
> lpf = processing.runalg("saga:simplefilter", dem, 0, 0, radius, None)
>
>
> An error has occured while executing Python code:
>
> Traceback (most recent call last):
>   File 
> "/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
> line 256, in accept
> self.finish()
>   File 
> "/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
> line 281, in finish
> Postprocessing.handleAlgorithmResults(self.alg, self, not keepOpen)
>   File "/home/eric/.qgis2/python/plugins/processing/gui/Postprocessing.py",
> line 77, in handleAlgorithmResults
> dlg = CouldNotLoadResultsDialog(wrongLayers, alg)
>   File 
> "/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
> line 41, in __init__
> self.setupUi()
>   File 
> "/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
> line 50, in setupUi
> html = self.alg.getPostProcessingErrorMessage(self.wrongLayers)
>   File "/home/eric/.qgis2/python/plugins/processing/core/GeoAlgorithm.py",
> line 520, in getPostProcessingErrorMessage
> + layer.value + '\n'
> TypeError: cannot concatenate 'str' and 'dict' objects
>
>
> Python version:
> 2.7.6 (default, Feb 26 2014, 12:07:17)
> [GCC 4.8.2 20140206 (prerelease)]
>
>
> QGIS version:
> 2.3.0-Master Master, 9e42c5b
>
> Python path: ['/home/eric/.qgis2/python/plugins/processing',
> '/usr/share/qgis/python', u'/home/eric/.qgis2/python',
> u'/home/eric/.qgis2/python/plugins', '/usr/share/qgis/python/plugins',
> '/usr/lib/python27.zip', '/usr/lib/python2.7',
> '/usr/lib/python2.7/plat-linux2', '/usr/lib/python2.7/lib-tk',
> '/usr/lib/python2.7/lib-old', '/usr/lib/python2.7/lib-dynload',
> '/usr/lib/python2.7/site-packages',
> '/usr/lib/python2.7/site-packages/gtk-2.0',
> '/usr/lib/python2.7/site-packages/wx-3.0-gtk2',
> '/usr/share/qgis/python/plugins/fTools/tools', '/home/eric/Documents']
>
>
> Running the code in the console completes successfully. However, I'm
> still unable to run a grass algorithm when the grass_region_parameter
> and grass_region_cellsize_parameter are set to None.
>
> Thanks,
> Eric
>
> On Mon, Mar 24, 2014 at 10:13 PM, Eric Goddard  wrote:
>> Oops, my mistake. I have ##lrm=output raster in the header, but lpf = ... Is
>> the only line in my code.
>>
>> I don't get the error message after changing lrm to lpf in the header. I do
>> get another error message, but I think I broke my processing by trying to
>> force an update. I'll test it tomorrow. Sorry about the confusion!
>>
>> Eric
>>
>> On Mar 22, 2014 5:06 PM, "Eric Goddard"  wrote:
>>>
>>> Ah, I didn't know a file could he specified without a path. Thanks for the
>>> clarification!
>>>
>>> On Mar 22, 2014 4:37 PM, "Victor Olaya"  wrote:

 > If I run the same line in the python console, it works correctly:
  radius = 4
  lpf = processing.runalg("saga:simplefilter", "SR_50M_alaska_nad", 0,
  0, radius, None)
  processing.load(lpf['RESULT'])
 >
 > Results in the map being loaded.

 Not sure why this is happening. I will try to reproduce the error and
 see if it can be fixed
 Thanks for the report

 >
 > Also, if I change the output folder in the processing settings to a
 > different location (in this case, tmp/processing in my home folder)
 > and run another algorithm in the python console, the RESULT key still
 > points to /tmp/processing/...

 This is not a bug. If you do not enter a filename, the output will
 always go to a temporary file. The output folder that yu can set in
 the settings  is the folder to be used when you enter a filename
 without a path (like "myfile.shp", but not "/home/myuser/myfile.shp"),
 not the folder to use for temporary files.

 Thanks again!
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] processing.runalg() issue with grass algorithms

[Eric Goddard]

Alright, latest master (9e42c5b), with processing version
2.0-20131120. Running a custom tool that consists of the following
code results in an error:

##dem=raster
##radius=number 4
##lpf=output raster

lpf = processing.runalg("saga:simplefilter", dem, 0, 0, radius, None)


An error has occured while executing Python code:

Traceback (most recent call last):
  File 
"/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
line 256, in accept
self.finish()
  File 
"/home/eric/.qgis2/python/plugins/processing/gui/AlgorithmExecutionDialog.py",
line 281, in finish
Postprocessing.handleAlgorithmResults(self.alg, self, not keepOpen)
  File "/home/eric/.qgis2/python/plugins/processing/gui/Postprocessing.py",
line 77, in handleAlgorithmResults
dlg = CouldNotLoadResultsDialog(wrongLayers, alg)
  File 
"/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
line 41, in __init__
self.setupUi()
  File 
"/home/eric/.qgis2/python/plugins/processing/gui/CouldNotLoadResultsDialog.py",
line 50, in setupUi
html = self.alg.getPostProcessingErrorMessage(self.wrongLayers)
  File "/home/eric/.qgis2/python/plugins/processing/core/GeoAlgorithm.py",
line 520, in getPostProcessingErrorMessage
+ layer.value + '\n'
TypeError: cannot concatenate 'str' and 'dict' objects


Python version:
2.7.6 (default, Feb 26 2014, 12:07:17)
[GCC 4.8.2 20140206 (prerelease)]


QGIS version:
2.3.0-Master Master, 9e42c5b

Python path: ['/home/eric/.qgis2/python/plugins/processing',
'/usr/share/qgis/python', u'/home/eric/.qgis2/python',
u'/home/eric/.qgis2/python/plugins', '/usr/share/qgis/python/plugins',
'/usr/lib/python27.zip', '/usr/lib/python2.7',
'/usr/lib/python2.7/plat-linux2', '/usr/lib/python2.7/lib-tk',
'/usr/lib/python2.7/lib-old', '/usr/lib/python2.7/lib-dynload',
'/usr/lib/python2.7/site-packages',
'/usr/lib/python2.7/site-packages/gtk-2.0',
'/usr/lib/python2.7/site-packages/wx-3.0-gtk2',
'/usr/share/qgis/python/plugins/fTools/tools', '/home/eric/Documents']


Running the code in the console completes successfully. However, I'm
still unable to run a grass algorithm when the grass_region_parameter
and grass_region_cellsize_parameter are set to None.

Thanks,
Eric

On Mon, Mar 24, 2014 at 10:13 PM, Eric Goddard  wrote:
> Oops, my mistake. I have ##lrm=output raster in the header, but lpf = ... Is
> the only line in my code.
>
> I don't get the error message after changing lrm to lpf in the header. I do
> get another error message, but I think I broke my processing by trying to
> force an update. I'll test it tomorrow. Sorry about the confusion!
>
> Eric
>
> On Mar 22, 2014 5:06 PM, "Eric Goddard"  wrote:
>>
>> Ah, I didn't know a file could he specified without a path. Thanks for the
>> clarification!
>>
>> On Mar 22, 2014 4:37 PM, "Victor Olaya"  wrote:
>>>
>>> > If I run the same line in the python console, it works correctly:
>>>  radius = 4
>>>  lpf = processing.runalg("saga:simplefilter", "SR_50M_alaska_nad", 0,
>>>  0, radius, None)
>>>  processing.load(lpf['RESULT'])
>>> >
>>> > Results in the map being loaded.
>>>
>>> Not sure why this is happening. I will try to reproduce the error and
>>> see if it can be fixed
>>> Thanks for the report
>>>
>>> >
>>> > Also, if I change the output folder in the processing settings to a
>>> > different location (in this case, tmp/processing in my home folder)
>>> > and run another algorithm in the python console, the RESULT key still
>>> > points to /tmp/processing/...
>>>
>>> This is not a bug. If you do not enter a filename, the output will
>>> always go to a temporary file. The output folder that yu can set in
>>> the settings  is the folder to be used when you enter a filename
>>> without a path (like "myfile.shp", but not "/home/myuser/myfile.shp"),
>>> not the folder to use for temporary files.
>>>
>>> Thanks again!
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Duplication of functionality: Data-defined size and Size scale field

[Nyall Dawson]

>
> If we do that, we need to find a more readable interface interface for data
> defined properties. It is unclear to the user if some options are activated
> in there or not. A visual flag could do the trick, or a complete gui review,
> to avoid another subdialog. (hierarchical view is the current ergonomy
> option, and I like it).
>

I think the ultimate solution would be to port the data defined gui
control which is used in the labelling settings over to symbology.
Larry's done a great job with this control and it's behaves very
nicely - it's much easier to set fields, copy expressions, temporarily
deactivate expressions, and see immediately if a setting is data
defined.

I'm hoping during 2.4 to make some composer item settings data
definable, and I plan on using this same control. It would be nice to
add it to symbology too so that there's consistency across the entire
QGIS gui.

Nyall
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Vienna hackfest: QGIS Legend discussion

[Nyall Dawson]

On 26 March 2014 07:16, Martin Dobias  wrote:
> Hi Vienna hackfest participants
>
> tomorrow (Wednesday) I would like to have a session on QGIS Legend
> (the thing on the left side of map canvas) and its future - starting
> from 10:00. I am planning some refactoring of the legend widget and it
> would be great to discuss various possible future use cases - the
> refactoring may help to make our dreams come true :-)
>

Martin - I'm not sure if it falls outside of the scope of this but an
idea which has been bouncing around my head is the ability to set
transparency and blending options for a layer group. Here's how I
think it should work:

- Have a setting for layer groups for something like "render child
layers together"
- If this is set, then all child layers from that group are first
composited together onto a blank image using their individual layer
blending/transparency settings
- This entire group image is then composited onto the map using the
group's transparency/blending setting

There's two main use cases I can see which would benefit from this:
1. I've played with exposing the various QPainter masking blend modes
to QGIS (see DestinationOver, SourceIn/DestinationIn and
SourceOut/DestinationOut at
http://doc.qt.digia.com/qq/qq17-compositionmodes.html ). However,
since these modes currently apply to all underlying layers in the
stack (including the canvas background color) the end result is
confusing and not very useful. If the scope of the masking operation
was restricted by the process outlined above then it would be possible
for these masking modes to only apply to underlying child layers from
the same layer group.

2. Sometimes it's cartographically desirable to have several layers
completely obscure each other (100% opacity), but be rendered as a
group with some level of transparency.

I'm happy to code this up myself sometime, but if changes are being
made to how layer groups operate then it may be beneficial to consider
the possibility of layer group settings like this for the future.

Nyall
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [360] pyArchInit approval notification.

[noreply]

Plugin pyArchInit approval by pcav.
The plugin version "[360] pyArchInit 0.1.1 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/pyarchinit/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [360] pyArchInit approval notification.

[noreply]

Plugin pyArchInit approval by pcav.
The plugin version "[360] pyArchInit 0.1 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/pyarchinit/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Vienna hackfest: QGIS Legend discussion

[Marco Lechner - FOSSGIS e.V.]

hi,

just a little input from my last QGIS beginners course. Many (!)
participants did not understand that toggling a layer group swutches ALL
layers on/off instead of the visible layers. Infact if some layers of a
group are visible toggling the group activates ALL layers (and may
result in QGIS trying to render a big bunch of layers).

May be there are alternative (or customizable) ideas.

Marco

Am 25.03.2014 21:16, schrieb Martin Dobias:
> Hi Vienna hackfest participants
>
> tomorrow (Wednesday) I would like to have a session on QGIS Legend
> (the thing on the left side of map canvas) and its future - starting
> from 10:00. I am planning some refactoring of the legend widget and it
> would be great to discuss various possible future use cases - the
> refactoring may help to make our dreams come true :-)
>
> Some items from my wishlist:
> - make legend available in GUI library
> - split legend into model/view architecture
> - allow more flexible display of symbology in legend
> - make legend easily customizable
> - inline editing of symbols
>
> Cheers
> Martin
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer

___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Vienna hackfest: QGIS Legend discussion

[Martin Dobias]

Hi Vienna hackfest participants

tomorrow (Wednesday) I would like to have a session on QGIS Legend
(the thing on the left side of map canvas) and its future - starting
from 10:00. I am planning some refactoring of the legend widget and it
would be great to discuss various possible future use cases - the
refactoring may help to make our dreams come true :-)

Some items from my wishlist:
- make legend available in GUI library
- split legend into model/view architecture
- allow more flexible display of symbology in legend
- make legend easily customizable
- inline editing of symbols

Cheers
Martin
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Duplication of functionality: Data-defined size and Size scale field

[Régis Haubourg]

agreed, current situation is confusing and prone to bugs.
first drawback of proposed solution: advanced fields were a bit more user
friendly (user interface> no formula), and data defined settings are hidden
one step deeper in dialogs. I also remember how trainees reacted last week
when I explained data defined properties: lost..

If we do that, we need to find a more readable interface interface for data
defined properties. It is unclear to the user if some options are activated
in there or not. A visual flag could do the trick, or a complete gui review,
to avoid another subdialog. (hierarchical view is the current ergonomy
option, and I like it).

I'm really wondering if the right solution could not be an assistant asking
for min size, proportional rules to apply( linear, square tail, improved
square tail (exp 0.57) .. ) and showing and generating a legend preview
based on a data sample. Assistant will then fill the right expression in
data defined expression, and tell the user where to find it. 

I have been sharing thoose thoughts with Martin Dobias who has legend
refactoring in his roadmap. Martin, any opinion?

Cheers
Régis



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/Duplication-of-functionality-Data-defined-size-and-Size-scale-field-tp5131007p5131055.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] New website and symbols/styles GSOC PR

[Alessandro Pasotti]

Hi,

I'm giving a quick look into the gsoc PR
https://github.com/qgis/QGIS-Django/pull/11

Frankly speaking it's a very good student exercise, but I dont' think it's
production ready.

It's missing basic validation, unique constraints, i18n strings,  list
views, integration with the main app and there are a few cosmetic issues.

I'm busy on other tasks and I don't want to work on this. Is there anybody
willing to complete the job?


-- 
Alessandro Pasotti
w3:   www.itopen.it
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] public key on hub.qgis.org problem

[Gino Pirelli]

ha sorry I did't understand you have your code in the hub.qgis.org git


On 25 March 2014 18:48, Giuseppe De Marco wrote:

> I guess so since you're telling me so let's say that I choose SCM ->
> Git inside my hub.qgis.org
> account and name it "my_plugin", then I would have to add this
> repository with git inside my local folder:
>
> git init
> git add .
> git commit -m 'init repo'
> git remote add origin gito...@qgis.org:my_plugin
> git push origin master
>
> and it should be working
> right or am I missing something?
>
>
>
>
> 2014-03-25 18:33 GMT+01:00 Gino Pirelli :
> > I don't remember there is need of a key to upload plugins... am I wrong?
> >
> > "git push origin master"
> > could be a problem related to you remote git repository configuration?
> >
> > ciao Luigi Pirelli
> >
> >
> >
> >
> > On 25 March 2014 18:23, Giuseppe De Marco 
> > wrote:
> >>
> >> I was trying to upload a plugin version on hub.qgis.org
> >> after updating my rss key on laptop, and I was asked for a password
> >> after entering  $ git push origin master
> >>
> >> opening a terminal and issuing "ssh -v gito...@qgis.org "
> >> I get
> >>
> >> OpenSSH_6.2p2 Ubuntu-6ubuntu0.1, OpenSSL 1.0.1e 11 Feb 2013
> >> debug1: Reading configuration data /etc/ssh/ssh_config
> >> debug1: /etc/ssh/ssh_config line 19: Applying options for *
> >> debug1: Connecting to qgis.org [140.211.15.67] port 22.
> >> debug1: Connection established.
> >> debug1: identity file /home/joe/.ssh/id_rsa type 1
> >> debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
> >> debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
> >> debug1: identity file /home/joe/.ssh/id_rsa-cert type -1
> >> debug1: identity file /home/joe/.ssh/id_dsa type -1
> >> debug1: identity file /home/joe/.ssh/id_dsa-cert type -1
> >> debug1: identity file /home/joe/.ssh/id_ecdsa type -1
> >> debug1: identity file /home/joe/.ssh/id_ecdsa-cert type -1
> >> debug1: Enabling compatibility mode for protocol 2.0
> >> debug1: Local version string SSH-2.0-OpenSSH_6.2p2 Ubuntu-6ubuntu0.1
> >> debug1: Remote protocol version 2.0, remote software version
> >> OpenSSH_5.5p1 Debian-6+squeeze4
> >> debug1: match: OpenSSH_5.5p1 Debian-6+squeeze4 pat OpenSSH_5*
> >> debug1: SSH2_MSG_KEXINIT sent
> >> debug1: SSH2_MSG_KEXINIT received
> >> debug1: kex: server->client aes128-ctr hmac-md5 none
> >> debug1: kex: client->server aes128-ctr hmac-md5 none
> >> debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(1024<1024<8192) sent
> >> debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
> >> debug1: SSH2_MSG_KEX_DH_GEX_INIT sent
> >> debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY
> >> debug1: Server host key: RSA
> >> a7:05:f6:a5:02:ec:92:73:4a:0a:12:16:4c:11:e1:1c
> >> debug1: Host 'qgis.org' is known and matches the RSA host key.
> >> debug1: Found key in /home/joe/.ssh/known_hosts:1
> >> debug1: ssh_rsa_verify: signature correct
> >> debug1: SSH2_MSG_NEWKEYS sent
> >> debug1: expecting SSH2_MSG_NEWKEYS
> >> debug1: SSH2_MSG_NEWKEYS received
> >> debug1: Roaming not allowed by server
> >> debug1: SSH2_MSG_SERVICE_REQUEST sent
> >> debug1: SSH2_MSG_SERVICE_ACCEPT received
> >> debug1: Authentications that can continue: publickey,password
> >> debug1: Next authentication method: publickey
> >> debug1: Offering RSA public key: /home/joe/.ssh/id_rsa
> >> debug1: Authentications that can continue: publickey,password
> >> debug1: Trying private key: /home/joe/.ssh/id_dsa
> >> debug1: Trying private key: /home/joe/.ssh/id_ecdsa
> >> debug1: Next authentication method: password
> >> gito...@qgis.org's password:
> >> debug1: Authentications that can continue: publickey,password
> >> Permission denied, please try again.
> >> gito...@qgis.org's password:
> >> debug1: Authentications that can continue: publickey,password
> >> Permission denied, please try again.
> >> gito...@qgis.org's password:
> >> debug1: Authentications that can continue: publickey,password
> >> debug1: No more authentication methods to try.
> >> Permission denied (publickey,password).
> >>
> >>
> >> So I went to my account looking for my public keys at:
> >> http://hub.qgis.org/my/public_keys
> >> but my browser tells me page not found error
> >>
> >> anyone could help?
> >>
> >> thank you
> >>
> >> --
> >> Dott. Agr. Giuseppe De Marco
> >> RSPP settore ATECO 1
> >> Cell./Mobile: +39 3935041115
> >> Fax.: +39 0503869174
> >> PEC: giuseppe.de_ma...@epap.conafpec.it
> >> ___
> >> Qgis-developer mailing list
> >> Qgis-developer@lists.osgeo.org
> >> http://lists.osgeo.org/mailman/listinfo/qgis-developer
> >
> >
>
>
>
> --
> Dott. Agr. Giuseppe De Marco
> RSPP settore ATECO 1
> Cell./Mobile: +39 3935041115
> Fax.: +39 0503869174
> PEC: giuseppe.de_ma...@epap.conafpec.it
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] public key on hub.qgis.org problem

[Giuseppe De Marco]

I guess so since you're telling me so let's say that I choose SCM ->
Git inside my hub.qgis.org
account and name it "my_plugin", then I would have to add this
repository with git inside my local folder:

git init
git add .
git commit -m 'init repo'
git remote add origin gito...@qgis.org:my_plugin
git push origin master

and it should be working
right or am I missing something?




2014-03-25 18:33 GMT+01:00 Gino Pirelli :
> I don't remember there is need of a key to upload plugins... am I wrong?
>
> "git push origin master"
> could be a problem related to you remote git repository configuration?
>
> ciao Luigi Pirelli
>
>
>
>
> On 25 March 2014 18:23, Giuseppe De Marco 
> wrote:
>>
>> I was trying to upload a plugin version on hub.qgis.org
>> after updating my rss key on laptop, and I was asked for a password
>> after entering  $ git push origin master
>>
>> opening a terminal and issuing "ssh -v gito...@qgis.org "
>> I get
>>
>> OpenSSH_6.2p2 Ubuntu-6ubuntu0.1, OpenSSL 1.0.1e 11 Feb 2013
>> debug1: Reading configuration data /etc/ssh/ssh_config
>> debug1: /etc/ssh/ssh_config line 19: Applying options for *
>> debug1: Connecting to qgis.org [140.211.15.67] port 22.
>> debug1: Connection established.
>> debug1: identity file /home/joe/.ssh/id_rsa type 1
>> debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
>> debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
>> debug1: identity file /home/joe/.ssh/id_rsa-cert type -1
>> debug1: identity file /home/joe/.ssh/id_dsa type -1
>> debug1: identity file /home/joe/.ssh/id_dsa-cert type -1
>> debug1: identity file /home/joe/.ssh/id_ecdsa type -1
>> debug1: identity file /home/joe/.ssh/id_ecdsa-cert type -1
>> debug1: Enabling compatibility mode for protocol 2.0
>> debug1: Local version string SSH-2.0-OpenSSH_6.2p2 Ubuntu-6ubuntu0.1
>> debug1: Remote protocol version 2.0, remote software version
>> OpenSSH_5.5p1 Debian-6+squeeze4
>> debug1: match: OpenSSH_5.5p1 Debian-6+squeeze4 pat OpenSSH_5*
>> debug1: SSH2_MSG_KEXINIT sent
>> debug1: SSH2_MSG_KEXINIT received
>> debug1: kex: server->client aes128-ctr hmac-md5 none
>> debug1: kex: client->server aes128-ctr hmac-md5 none
>> debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(1024<1024<8192) sent
>> debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
>> debug1: SSH2_MSG_KEX_DH_GEX_INIT sent
>> debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY
>> debug1: Server host key: RSA
>> a7:05:f6:a5:02:ec:92:73:4a:0a:12:16:4c:11:e1:1c
>> debug1: Host 'qgis.org' is known and matches the RSA host key.
>> debug1: Found key in /home/joe/.ssh/known_hosts:1
>> debug1: ssh_rsa_verify: signature correct
>> debug1: SSH2_MSG_NEWKEYS sent
>> debug1: expecting SSH2_MSG_NEWKEYS
>> debug1: SSH2_MSG_NEWKEYS received
>> debug1: Roaming not allowed by server
>> debug1: SSH2_MSG_SERVICE_REQUEST sent
>> debug1: SSH2_MSG_SERVICE_ACCEPT received
>> debug1: Authentications that can continue: publickey,password
>> debug1: Next authentication method: publickey
>> debug1: Offering RSA public key: /home/joe/.ssh/id_rsa
>> debug1: Authentications that can continue: publickey,password
>> debug1: Trying private key: /home/joe/.ssh/id_dsa
>> debug1: Trying private key: /home/joe/.ssh/id_ecdsa
>> debug1: Next authentication method: password
>> gito...@qgis.org's password:
>> debug1: Authentications that can continue: publickey,password
>> Permission denied, please try again.
>> gito...@qgis.org's password:
>> debug1: Authentications that can continue: publickey,password
>> Permission denied, please try again.
>> gito...@qgis.org's password:
>> debug1: Authentications that can continue: publickey,password
>> debug1: No more authentication methods to try.
>> Permission denied (publickey,password).
>>
>>
>> So I went to my account looking for my public keys at:
>> http://hub.qgis.org/my/public_keys
>> but my browser tells me page not found error
>>
>> anyone could help?
>>
>> thank you
>>
>> --
>> Dott. Agr. Giuseppe De Marco
>> RSPP settore ATECO 1
>> Cell./Mobile: +39 3935041115
>> Fax.: +39 0503869174
>> PEC: giuseppe.de_ma...@epap.conafpec.it
>> ___
>> Qgis-developer mailing list
>> Qgis-developer@lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>



-- 
Dott. Agr. Giuseppe De Marco
RSPP settore ATECO 1
Cell./Mobile: +39 3935041115
Fax.: +39 0503869174
PEC: giuseppe.de_ma...@epap.conafpec.it
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] public key on hub.qgis.org problem

[Gino Pirelli]

I don't remember there is need of a key to upload plugins... am I wrong?

"git push origin master"
could be a problem related to you remote git repository configuration?

ciao Luigi Pirelli




On 25 March 2014 18:23, Giuseppe De Marco wrote:

> I was trying to upload a plugin version on hub.qgis.org
> after updating my rss key on laptop, and I was asked for a password
> after entering  $ git push origin master
>
> opening a terminal and issuing "ssh -v gito...@qgis.org "
> I get
>
> OpenSSH_6.2p2 Ubuntu-6ubuntu0.1, OpenSSL 1.0.1e 11 Feb 2013
> debug1: Reading configuration data /etc/ssh/ssh_config
> debug1: /etc/ssh/ssh_config line 19: Applying options for *
> debug1: Connecting to qgis.org [140.211.15.67] port 22.
> debug1: Connection established.
> debug1: identity file /home/joe/.ssh/id_rsa type 1
> debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
> debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
> debug1: identity file /home/joe/.ssh/id_rsa-cert type -1
> debug1: identity file /home/joe/.ssh/id_dsa type -1
> debug1: identity file /home/joe/.ssh/id_dsa-cert type -1
> debug1: identity file /home/joe/.ssh/id_ecdsa type -1
> debug1: identity file /home/joe/.ssh/id_ecdsa-cert type -1
> debug1: Enabling compatibility mode for protocol 2.0
> debug1: Local version string SSH-2.0-OpenSSH_6.2p2 Ubuntu-6ubuntu0.1
> debug1: Remote protocol version 2.0, remote software version
> OpenSSH_5.5p1 Debian-6+squeeze4
> debug1: match: OpenSSH_5.5p1 Debian-6+squeeze4 pat OpenSSH_5*
> debug1: SSH2_MSG_KEXINIT sent
> debug1: SSH2_MSG_KEXINIT received
> debug1: kex: server->client aes128-ctr hmac-md5 none
> debug1: kex: client->server aes128-ctr hmac-md5 none
> debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(1024<1024<8192) sent
> debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
> debug1: SSH2_MSG_KEX_DH_GEX_INIT sent
> debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY
> debug1: Server host key: RSA
> a7:05:f6:a5:02:ec:92:73:4a:0a:12:16:4c:11:e1:1c
> debug1: Host 'qgis.org' is known and matches the RSA host key.
> debug1: Found key in /home/joe/.ssh/known_hosts:1
> debug1: ssh_rsa_verify: signature correct
> debug1: SSH2_MSG_NEWKEYS sent
> debug1: expecting SSH2_MSG_NEWKEYS
> debug1: SSH2_MSG_NEWKEYS received
> debug1: Roaming not allowed by server
> debug1: SSH2_MSG_SERVICE_REQUEST sent
> debug1: SSH2_MSG_SERVICE_ACCEPT received
> debug1: Authentications that can continue: publickey,password
> debug1: Next authentication method: publickey
> debug1: Offering RSA public key: /home/joe/.ssh/id_rsa
> debug1: Authentications that can continue: publickey,password
> debug1: Trying private key: /home/joe/.ssh/id_dsa
> debug1: Trying private key: /home/joe/.ssh/id_ecdsa
> debug1: Next authentication method: password
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> Permission denied, please try again.
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> Permission denied, please try again.
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> debug1: No more authentication methods to try.
> Permission denied (publickey,password).
>
>
> So I went to my account looking for my public keys at:
> http://hub.qgis.org/my/public_keys
> but my browser tells me page not found error
>
> anyone could help?
>
> thank you
>
> --
> Dott. Agr. Giuseppe De Marco
> RSPP settore ATECO 1
> Cell./Mobile: +39 3935041115
> Fax.: +39 0503869174
> PEC: giuseppe.de_ma...@epap.conafpec.it
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] public key on hub.qgis.org problem

[Alex Mandel]

Guess no one has looked into this since you last emailed about the
issue. I'm once again forwarding to the people who I think will know how
to fix.

Thanks,
Alex

On 03/25/2014 10:23 AM, Giuseppe De Marco wrote:
> I was trying to upload a plugin version on hub.qgis.org
> after updating my rss key on laptop, and I was asked for a password
> after entering  $ git push origin master
> 
> opening a terminal and issuing "ssh -v gito...@qgis.org "
> I get
> 
> OpenSSH_6.2p2 Ubuntu-6ubuntu0.1, OpenSSL 1.0.1e 11 Feb 2013
> debug1: Reading configuration data /etc/ssh/ssh_config
> debug1: /etc/ssh/ssh_config line 19: Applying options for *
> debug1: Connecting to qgis.org [140.211.15.67] port 22.
> debug1: Connection established.
> debug1: identity file /home/joe/.ssh/id_rsa type 1
> debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
> debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
> debug1: identity file /home/joe/.ssh/id_rsa-cert type -1
> debug1: identity file /home/joe/.ssh/id_dsa type -1
> debug1: identity file /home/joe/.ssh/id_dsa-cert type -1
> debug1: identity file /home/joe/.ssh/id_ecdsa type -1
> debug1: identity file /home/joe/.ssh/id_ecdsa-cert type -1
> debug1: Enabling compatibility mode for protocol 2.0
> debug1: Local version string SSH-2.0-OpenSSH_6.2p2 Ubuntu-6ubuntu0.1
> debug1: Remote protocol version 2.0, remote software version
> OpenSSH_5.5p1 Debian-6+squeeze4
> debug1: match: OpenSSH_5.5p1 Debian-6+squeeze4 pat OpenSSH_5*
> debug1: SSH2_MSG_KEXINIT sent
> debug1: SSH2_MSG_KEXINIT received
> debug1: kex: server->client aes128-ctr hmac-md5 none
> debug1: kex: client->server aes128-ctr hmac-md5 none
> debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(1024<1024<8192) sent
> debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
> debug1: SSH2_MSG_KEX_DH_GEX_INIT sent
> debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY
> debug1: Server host key: RSA a7:05:f6:a5:02:ec:92:73:4a:0a:12:16:4c:11:e1:1c
> debug1: Host 'qgis.org' is known and matches the RSA host key.
> debug1: Found key in /home/joe/.ssh/known_hosts:1
> debug1: ssh_rsa_verify: signature correct
> debug1: SSH2_MSG_NEWKEYS sent
> debug1: expecting SSH2_MSG_NEWKEYS
> debug1: SSH2_MSG_NEWKEYS received
> debug1: Roaming not allowed by server
> debug1: SSH2_MSG_SERVICE_REQUEST sent
> debug1: SSH2_MSG_SERVICE_ACCEPT received
> debug1: Authentications that can continue: publickey,password
> debug1: Next authentication method: publickey
> debug1: Offering RSA public key: /home/joe/.ssh/id_rsa
> debug1: Authentications that can continue: publickey,password
> debug1: Trying private key: /home/joe/.ssh/id_dsa
> debug1: Trying private key: /home/joe/.ssh/id_ecdsa
> debug1: Next authentication method: password
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> Permission denied, please try again.
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> Permission denied, please try again.
> gito...@qgis.org's password:
> debug1: Authentications that can continue: publickey,password
> debug1: No more authentication methods to try.
> Permission denied (publickey,password).
> 
> 
> So I went to my account looking for my public keys at:
> http://hub.qgis.org/my/public_keys
> but my browser tells me page not found error
> 
> anyone could help?
> 
> thank you
> 


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] public key on hub.qgis.org problem

[Giuseppe De Marco]

I was trying to upload a plugin version on hub.qgis.org
after updating my rss key on laptop, and I was asked for a password
after entering  $ git push origin master

opening a terminal and issuing "ssh -v gito...@qgis.org "
I get

OpenSSH_6.2p2 Ubuntu-6ubuntu0.1, OpenSSL 1.0.1e 11 Feb 2013
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: /etc/ssh/ssh_config line 19: Applying options for *
debug1: Connecting to qgis.org [140.211.15.67] port 22.
debug1: Connection established.
debug1: identity file /home/joe/.ssh/id_rsa type 1
debug1: Checking blacklist file /usr/share/ssh/blacklist.RSA-2048
debug1: Checking blacklist file /etc/ssh/blacklist.RSA-2048
debug1: identity file /home/joe/.ssh/id_rsa-cert type -1
debug1: identity file /home/joe/.ssh/id_dsa type -1
debug1: identity file /home/joe/.ssh/id_dsa-cert type -1
debug1: identity file /home/joe/.ssh/id_ecdsa type -1
debug1: identity file /home/joe/.ssh/id_ecdsa-cert type -1
debug1: Enabling compatibility mode for protocol 2.0
debug1: Local version string SSH-2.0-OpenSSH_6.2p2 Ubuntu-6ubuntu0.1
debug1: Remote protocol version 2.0, remote software version
OpenSSH_5.5p1 Debian-6+squeeze4
debug1: match: OpenSSH_5.5p1 Debian-6+squeeze4 pat OpenSSH_5*
debug1: SSH2_MSG_KEXINIT sent
debug1: SSH2_MSG_KEXINIT received
debug1: kex: server->client aes128-ctr hmac-md5 none
debug1: kex: client->server aes128-ctr hmac-md5 none
debug1: SSH2_MSG_KEX_DH_GEX_REQUEST(1024<1024<8192) sent
debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
debug1: SSH2_MSG_KEX_DH_GEX_INIT sent
debug1: expecting SSH2_MSG_KEX_DH_GEX_REPLY
debug1: Server host key: RSA a7:05:f6:a5:02:ec:92:73:4a:0a:12:16:4c:11:e1:1c
debug1: Host 'qgis.org' is known and matches the RSA host key.
debug1: Found key in /home/joe/.ssh/known_hosts:1
debug1: ssh_rsa_verify: signature correct
debug1: SSH2_MSG_NEWKEYS sent
debug1: expecting SSH2_MSG_NEWKEYS
debug1: SSH2_MSG_NEWKEYS received
debug1: Roaming not allowed by server
debug1: SSH2_MSG_SERVICE_REQUEST sent
debug1: SSH2_MSG_SERVICE_ACCEPT received
debug1: Authentications that can continue: publickey,password
debug1: Next authentication method: publickey
debug1: Offering RSA public key: /home/joe/.ssh/id_rsa
debug1: Authentications that can continue: publickey,password
debug1: Trying private key: /home/joe/.ssh/id_dsa
debug1: Trying private key: /home/joe/.ssh/id_ecdsa
debug1: Next authentication method: password
gito...@qgis.org's password:
debug1: Authentications that can continue: publickey,password
Permission denied, please try again.
gito...@qgis.org's password:
debug1: Authentications that can continue: publickey,password
Permission denied, please try again.
gito...@qgis.org's password:
debug1: Authentications that can continue: publickey,password
debug1: No more authentication methods to try.
Permission denied (publickey,password).


So I went to my account looking for my public keys at:
http://hub.qgis.org/my/public_keys
but my browser tells me page not found error

anyone could help?

thank you

-- 
Dott. Agr. Giuseppe De Marco
RSPP settore ATECO 1
Cell./Mobile: +39 3935041115
Fax.: +39 0503869174
PEC: giuseppe.de_ma...@epap.conafpec.it
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Code Sprint agenda

[Gino Pirelli]

my agenda is focused on add functions to rt_mapserver_exporter plugin to
support missing parameter passed to map file

first goal was reached negotiating a solution with mapserver team to allow
managing of external graphic  (svg too) thanks to introduce a validation
block check in applySld mapscript api (
https://github.com/mapserver/mapserver/commit/e18424940d5c48391c61cf62b0bda883805bc871
)

next steps are to investigate how to integrate current
rt_mapserver_exporter code (that use SLD interface) with some missing
feature inspiring me from the Hugo Mercier fork of the same plugin.
The goal is... programmatically add new symbols and manage some not managed
parameters from qgis sld as GAP in point pattern.

bye from Vienna OSGeo Code Sprint

Luigi Pirelli


On 24 March 2014 18:49, Anita Graser  wrote:

> Hi,
>
> Thanks for sharing Paolo!
>
> My tasklist for this week is:
>
> * Processing bughunt
> * Expand Human Interface Guidelines
> * Write user group guidelines
> * UX bug/issue hunting, e.g. layer properties style dialog
>
> Just come by if you want to chime in.
>
> Best wishes,
> Anita
>
>
>
> Am 24.03.2014, 11:32 Uhr, schrieb Paolo Cavallini :
>
>
>  Hi all.
>> This is my agenda for the just started Code Sprint (AKA HackFest):
>>
>> * better management of sponsor and donor lists
>> * translation of lizmap
>> * setup a procedure for a more reliable and timely plugin approval
>> * help implementing the repository for symbols
>> * help improving course material: manual, data, Python examples.
>>
>> Anyone willing to collaborate on this, either in presence or remotely,
>> is welcome.
>> All the best.
>>
>
>
> --
> anitagraser.com
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Gino Pirelli]

just a note... there's no alternative RemoteDebug plugin... my changes
where merged in the main project

https://github.com/sourcepole/qgis-remote-debug

In my case it works... now I can't rememnber If I set something particular
(and I've no time to check now)... what I can say is that remote debug
plugin connect pydev server only if it is strictly clearly started...
otherwise sometimes it does nothing.

so my procedure is... running pydev debug server... then start qgis... then
start remotedebug plugin then I start to debug

see you Luigi Pirelli


On 25 March 2014 14:34, Franz1965  wrote:

> I've also taken a try with an alternate RemoteDebug plugin
> (https://github.com/luipir/qgis-remote-debug) but with no luck.
>
> It's strange... RemoteDebug is not in the list of installable plugins and
> you have to manually download and install it. But on downloaded version,
> inside metadata.txt, I find qgisMaximumVersion=2.99; so I guess that it
> should be compatible.
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130945.html
> Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Duplication of functionality: Data-defined size and Size scale field

[Anita Graser]

Hi,

I'd like to discuss the future of symbol scaling :)

We currently have two ways of controlling symbol size using attribute  
values: using data-defined properties (newer) and size scale field (older).


Theoretically, the user could enable both but to me it seems that size  
scale field does not take effect if data-defined size is activated.


Additionally, size scale field had some issues, e.g.
"Size scale field does not work with symbol levels"  
http://hub.qgis.org/issues/3659


For the sake of simplicity, I think it would be preferable to only have  
one control for controlling symbol size using attribute values. And for  
me, data-defined properties are the more refined and better working  
solution.


Can we move to git rid of the size scale field option?

Best wishes,
Anita



--
anitagraser.com
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [291] Tile Map Scale Plugin unapproval notification.

[noreply]

Plugin Tile Map Scale Plugin unapproval by pcav.
The plugin version "[291] Tile Map Scale Plugin 0.2" is now unapproved
Link: http://plugins.qgis.org/plugins/TileMapScaleLevels/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [291] Tile Map Scale Plugin approval notification.

[noreply]

Plugin Tile Map Scale Plugin approval by pcav.
The plugin version "[291] Tile Map Scale Plugin 0.3" is now unapproved
Link: http://plugins.qgis.org/plugins/TileMapScaleLevels/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [291] Tile Map Scale Plugin approval notification.

[noreply]

Plugin Tile Map Scale Plugin approval by pcav.
The plugin version "[291] Tile Map Scale Plugin 0.5" is now unapproved
Link: http://plugins.qgis.org/plugins/TileMapScaleLevels/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[G. Allegri]

I'm sorry, I can't help you right now because I'm on a Linux machine. I
hope to reply as soon as I will be on a Windows system.

giovanni


2014-03-25 16:14 GMT+01:00 Franz1965 :

> ah yes... I confirm I see this message in QGis status bar
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130979.html
> Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>



-- 
Giovanni Allegri
http://about.me/giovanniallegri
Twitter: https://twitter.com/_giohappy_
blog: http://blog.spaziogis.it
GEO+ geomatica in Italia http://bit.ly/GEOplus
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Franz1965]

ah yes... I confirm I see this message in QGis status bar



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130979.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Franz1965]

If you refer to option "RemoteDebug", that's what I do:
- On Eclipse I execute PyDev->Start Debug Server
- On Eclipse, debug panel, I see a root named “Debug Server [Python Server]”
and a node named “Debug Server”
- On Eclipse, console panel, I  read “Debug Server at port: 5678”
- On QGis I press “StartDebug” button
- On Eclipse, debug panel, I see a new node under “Debug Server [Python
Server]” named “unknown”

I don’t see anywhere the message “PyDev debugging active”. Where would I see
this message, in the console?... 



giohappy wrote
> Does the QGIS debugger starts correctly? I mean, does the message "PyDev
> debugging active" appear?
> 
> giovanni





--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130976.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Luca Manganelli]

On Tue, Mar 25, 2014 at 3:46 PM, G. Allegri  wrote:

> Does the QGIS debugger starts correctly? I mean, does the message "PyDev
> debugging active" appear?
>

To me: yes, I see that in qgis status bar.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Qgis-Processing (former Sextante) has two bug trackers

[Alexander Bruy]

+1 from me to make Processing/SEXTANTE subproject if
this keep all tickets categories and allow to see Processing
bugs among QGIS ones

2014-03-24 21:32 GMT+02:00 Alex Mandel :
> As a sub-project, it would show up in queries of the main QGIS ticket
> system, but at the same time be able to have it's own internal categories.
>
> This to me seems the best compromise. It's really a big plugin with a
> myriad of backends. That complexity is what makes it a little different
> from the others mentioned. The others mentioned could be sub-projects of
> the main QGIS too. In reality all of them are core plugins, but they are
> still plugins. None of them are required to make QGIS work (boot). Hence
> they each have their own category. The challenge here is
> Processing/Sextante(I always call it by both since I do the the
> possibility of Processing stuff that isn't sextante) needs it's own
> categories/sub-categories.
>
> Also it's really easy for use to try moving it to be a subproject and
> then reassess (Takes about 30 seconds to move it). If it's still not
> right then we can talk about how to move tickets.
>
> Thanks,
> Alex
>
> On 03/24/2014 12:05 PM, Filipe Dias wrote:
>> Hi
>> I think Processing tickets should be in the main bug tracker, since GDAL
>> tools, fTools and Grass plugin tickets are also there.
>>
>> Also, Qgis-Processing is a really big (and great) feature, that interests a
>> lot of users, so it makes sense to keep its tickets in the main tracker.
>>
>> Regards
>> F.
>>
>>
>> On Mon, Mar 24, 2014 at 6:02 PM, Alex Mandel 
>> wrote:
>>
>>> On 03/24/2014 05:10 AM, Filipe Dias wrote:
 Hi

 QGIS-Processing (former Sextante) has open tickets on two bugtrackers:

 - the plugin bug tracker, built when it was a plugin, but with a lot of
 active tickets -> http://hub.qgis.org/projects/sextante/activity

 - the main bug tracker with a few open tickets ->
 http://hub.qgis.org/projects/quantum-gis/issues

 I understand those two can't be merged in one go, because the
 plugin-bugtracker has categories (QGIS-Processing SAGA GIS,
>>> QGIS-Processing
 GRASS GIS etc) that don't fit in the main bug tracker. But this should be
 solved ASAP, because the current situation is quite confusing

 Regards
 F.


>>>
>>> I'm of the opinion that the tracker should continue to be the Plugin
>>> tracker, not the main QGIS tracker. So I'd propse moving any
>>> Sextante/Processing tickets from the main or linking them into the
>>> Plugin tracker.
>>>
>>> To me it's still a plugin, just a core plugin but big enough that it's
>>> really a subproject. If we wanted to move the Plugin tracker as a
>>> sub-project of QGIS that would also make sense.
>>>
>>> Also Processing bugs aside from Processing not loading aren't
>>> showstoppers for release.
>>>
>>> Thanks,
>>> Alex
>>>
>>
>>
>>
>> ___
>> Qgis-developer mailing list
>> Qgis-developer@lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>>
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer



-- 
Alexander Bruy
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[G. Allegri]

Does the QGIS debugger starts correctly? I mean, does the message "PyDev
debugging active" appear?

giovanni


2014-03-25 15:36 GMT+01:00 Franz1965 :

> giohappy wrote
> > I usually do it without using the QGIS remote debugger.
> > http://pydev.org/manual_adv_remote_debugger.html
> > giovanni
>
> Hi Giovanni,
>
> Many thanks for help, it works :)
>
> But... it seems that, with this approach, dynamic breakpoints are replaced
> by static calls of pydevd.settrace() written into code. I'm right?... This
> can be good for simple scenario, but with complex code it could be tricky
> to
> statically write pydevd.settrace() wherever you suppose to need a break.
>
> So... that's a very good starting point for me, but I still need to fix
> RemoteDebug (and use breakpoints) for developing scenario I would like to
> open in the future.
>
> Best regards
>
> Francesco
>
>
>
>
> --
> View this message in context:
> http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130963.html
> Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>



-- 
Giovanni Allegri
http://about.me/giovanniallegri
Twitter: https://twitter.com/_giohappy_
blog: http://blog.spaziogis.it
GEO+ geomatica in Italia http://bit.ly/GEOplus
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Franz1965]

giohappy wrote
> I usually do it without using the QGIS remote debugger.
> http://pydev.org/manual_adv_remote_debugger.html
> giovanni

Hi Giovanni,

Many thanks for help, it works :)

But... it seems that, with this approach, dynamic breakpoints are replaced
by static calls of pydevd.settrace() written into code. I'm right?... This
can be good for simple scenario, but with complex code it could be tricky to
statically write pydevd.settrace() wherever you suppose to need a break.

So... that's a very good starting point for me, but I still need to fix
RemoteDebug (and use breakpoints) for developing scenario I would like to
open in the future.

Best regards

Francesco




--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130963.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] WMS load regression in 2.2

[Luca Manganelli]

On Tue, Mar 18, 2014 at 10:58 PM, Nyall Dawson wrote:
> [...]

After some investigations with Process Monitor, we discovered that QGIS
2.2, during the operation of adding a WMS Layer, is querying continuously
the file "srs.db".

How did we discover it?

 - with QGIS installed on local hard drive, adding a WMS layer is
instantaneous;
 - with QGIS installed on a network samba folder, adding a WMS is very slow
because it is repeatedly accessing the srs.db file (hundred times!).

The problem doesn't occur with QGIS 2.0.

Maybe the SRS caching code is changed?
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Want a stable, bug free release?

[Paolo Cavallini]

Hi all.
QGIS core team is committed to provide everyone the best possible free,
libre, open source GIS. In this complex process, there are areas that
are more difficult to cover with either volunteer or contract work,
especially:
* bugs fixing
* backporting bugfixes, and
* packaging and releasing minor, bugfixed versions.
Our first experiments, drawing from our funds from sponsors and donors
to support ad hoc work from core developers, was a clear success (tens
of bugs fixed immediately before the release of QGIS 2.2).
In order to make this effort sustainable over the long term, we need
more resources. Please consider supporting us, by directly hiring a
developer or by sponsoring the project. We believe this is in your best
interest, especially for corporate and institutional users, as it will
allow us to release cleaner versions, with far lesser bugs.
Please do not hesitate writing the list, or to me directly, in case you
are interested.
All the best.
-- 
Paolo Cavallini - www.faunalia.eu

on behalf of QGIS Project Steering Committee
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[G. Allegri]

I usually do it without using the QGIS remote debugger.
http://pydev.org/manual_adv_remote_debugger.html

giovanni
Il 25/mar/2014 14:02 "Luca Manganelli"  ha scritto:

> On Tue, Mar 25, 2014 at 9:39 AM, Francesco Sozzi wrote:
>
>> Hi Everybody,
>>
>> I'm new on QGis developing and Python and I'm trying to setup a developing
>> environment for building plugins on QGis.
>> My problem is that breakpoints on Eclipse do not work, process does not
>> stop
>> on them.
>> [...]
>> Any suggestion?...
>>
>> Best regards
>>
>
> Incredible,  I am trying to do it in this moment, and I have the same
> problem. I need help, too :-)
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Franz1965]

I've also taken a try with an alternate RemoteDebug plugin
(https://github.com/luipir/qgis-remote-debug) but with no luck.

It's strange... RemoteDebug is not in the list of installable plugins and
you have to manually download and install it. But on downloaded version,
inside metadata.txt, I find qgisMaximumVersion=2.99; so I guess that it
should be compatible.



--
View this message in context: 
http://osgeo-org.1560.x6.nabble.com/Debugging-Python-plugin-with-Eclipse-breakpoints-do-not-wotk-tp5130877p5130945.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Luca Manganelli]

On Tue, Mar 25, 2014 at 9:39 AM, Francesco Sozzi  wrote:

> Hi Everybody,
>
> I'm new on QGis developing and Python and I'm trying to setup a developing
> environment for building plugins on QGis.
> My problem is that breakpoints on Eclipse do not work, process does not
> stop
> on them.
> [...]
> Any suggestion?...
>
> Best regards
>

Incredible,  I am trying to do it in this moment, and I have the same
problem. I need help, too :-)
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [250] IDECANARIAS approval notification.

[noreply]

Plugin IDECANARIAS approval by pcav.
The plugin version "[250] IDECANARIAS 1.0 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/IDESearch/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [250] IDECANARIAS approval notification.

[noreply]

Plugin IDECANARIAS approval by pcav.
The plugin version "[250] IDECANARIAS 1.1 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/IDESearch/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [250] IDECANARIAS approval notification.

[noreply]

Plugin IDECANARIAS approval by pcav.
The plugin version "[250] IDECANARIAS 1.4 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/IDESearch/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [250] IDECANARIAS approval notification.

[noreply]

Plugin IDECANARIAS approval by pcav.
The plugin version "[250] IDECANARIAS 1.3 Experimental" is now approved
Link: http://plugins.qgis.org/plugins/IDESearch/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Plugin [479] Arrows approval notification.

[noreply]

Plugin Arrows approval by pcav.
The plugin version "[479] Arrows 0.2.1" is now approved
Link: http://plugins.qgis.org/plugins/arrows/
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Debugging Python plugin with Eclipse: breakpoints do not wotk

[Francesco Sozzi]

Hi Everybody,

I'm new on QGis developing and Python and I'm trying to setup a developing
environment for building plugins on QGis.
My problem is that breakpoints on Eclipse do not work, process does not stop
on them.

This is what I do to debug:
1) On Eclipse I execute PyDev->Start Debug Server
2) On QGis I press "Remote Debug" button
3) On Eclipse I then see, under "Debug Server [Python Server]", a new node
named "unknown"

This is what I do to setup plugin and Eclipse project:
1) on C:\Program Files\QGIS Valmiera\apps\qgis\python\plugins I've created a
folder named "MyFisrtPlugin" with all files I need for a basic test. 
   This test work fine in QGis
   Code is nested by use of TAB
2) on Eclispe I've created a new PyDev project with 
- Directory = c:\app\Eclipse\Workspace\MyFirstPlugin
- grammar version = 2.7
- Interpreter = Python 2.7
- Create links to existing sources
- Project External Source Folders = C:\Program Files\QGIS
Valmiera\apps\qgis\python\plugins\MyFirstPlugin
- Referenced projects = none


Here you are in detail what I've done for installing and setup:
- I use Windows 7 64 bit

- I use QGis 2.2

- I've installed PyDev on Eclipse 
- Help->"Install new software" set "http://pydev.org/updates"; and
install PyDev

- I've configured PyDev on Eclipse 
- Windows->Preferences choose page PyDev->Interpreters->Python
Interpreter, press "new" and set
- Python 2.7
- C:\Program Files\QGIS Valmiera\bin\python.exe

- I've installed on QGis "Plugin reloader" and "Plugin builder"
- Plugin->"Manage and install plugin"
- Settings: check "show also experimental plugin"
- search for "remote debug" and install (I did not find, check the
following section on how to install)
- search for "plugin reloader" and install
- search for "plugin builder" and install

- I've installed on QGis "Remote Debug" (it is not reachable by
Plugin->"Manage and install plugin")
- download from https://github.com/sourcepole/qgis-remote-debug
- copy RemoteDebug folder to C:\Program Files\QGIS
Valmiera\apps\qgis\python\plugins

- I've enabled Eclipse debugging:
- create a general project
- go to debug perspective
- PyDev->Start debug Server
- Open QGis and start Debug Remote

- I've made Eclipse understanding QGis API
- Click Window=>Preferences=>PyDev=>Interpreter - Python
- press "New folder" on Library tab
- add C:\Program Files\QGIS Valmiera\apps\qgis\python   

- to enable pyrcc4 and pyuic4 I've set Windows system variables
- PYTHONHOME = C:\Program Files\QGIS Valmiera\apps\Python27
- PYTHONPATH  = C:\Program Files\QGIS Valmiera\apps\Python27\Lib


Any suggestion?...

Best regards

Franz

___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Zoom to full extent in composer (Python)

[Vincent Damoy]

Hello,

I open a new composer in QGIS with Python, and I'd like to zoom to the 
QgsComposition extent (like the "zoom to full extent" button in the 
composer).


I just can't find the way to do so.

I sought in PYQGIS cookbook, in Qgis API documentation, I tried to find 
the classes and methods used for this button in QGIS sources... (I 
basically want to include this already made code into mine, in Python)


Have you any idea?
Thanks in advance for any reply.
All the best.
Vincent Damoy
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer