Re: [QGIS-Developer] QSqlTableModel not working with PostgreSQL 12

[Arnaud Morvan]

Thanks Jürgen for you reply and for such great work !

Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 10/12/2020 à 16:06, Jürgen E. Fischer a écrit :

Hi Arnaud,

On Thu, 10. Dec 2020 at 15:40:31 +0100, Arnaud Morvan wrote:

Is it possible we get Qt >= 5.14 in next QGIS version for Windows ?

I'm working on (well - actually working on getting back to work on) the OSGeo4W
reboot.  It'll have Qt 5.15 along with a lot of other updates - and less if any
of the cruft.  Essentially everything is inplace except for QGIS itself (but
that didn't change in the last few weeks…).

Jürgen


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] QSqlTableModel not working with PostgreSQL 12

[Arnaud Morvan]

Hello,

Seems support for PostgreSQL 12 has been added to QtSQL 5.14 : 
https://codereview.qt-project.org/c/qt/qtbase/+/277205


Concerning Ubuntu, seems Qt 5.14 is available in Ubuntu groovy 20.10, 
but for dev purpose we could always find a workaround.


On the other hand, final users mostly use QGIS on Windows, some of them 
use plugins base on QtSQL and this is blocking them from upgrading to 
PostgreSQL 12.


I have no visibility on the available Qt version in next QGIS version 
for Windows.


Is it possible we get Qt >= 5.14 in next QGIS version for Windows ?

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 18/06/2020 à 00:48, Nyall Dawson a écrit :

On Thu, 18 Jun 2020 at 02:20, David Erill  wrote:


Developing with QGIS 3.12 in Ubuntu Bionic

Any hints how could I solve this issue? Will Qt version updated in next 
releases?

That's up to Ubuntu to decide -- you're getting the Qt version pushed
out by Ubuntu, and the QGIS project doesn't influence this decision in
any way...

Nyall
___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] Processing: sort algorithm not working?

[Arnaud Morvan]

Hi Andreas.

Note that the order of features in attributes table does not seems to 
reflect the source order.


After sorting your dataset with sortbyexpression, try to add a 
auto-incremental field and open this new output in attribute table.


Sort this attribute table by the new auto-incremental field. If the 
order is good, it is not a sortbyexpression issue, but an attribute 
table issue.


Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 21/11/2020 à 17:23, Bernd Vogelgesang a écrit :


Hi Andreas,

have a look at my last years issue on this

https://github.com/qgis/QGIS/issues/29794

maybe it's the same

Cheers,

Bernd

On 19.11.20 14:16, Andreas Neumann wrote:

Hi,

I have a model where at the very end before adding it to the project I was
trying to sort the result with an expression.

But: the sorting doesn't work. In the help there is the cryptic hint at
https://docs.qgis.org/3.10/de/docs/user_manual/processing_algs/qgis/vectorgeneral.html#qgisorderbyexpression
that "Be careful, it might not work as expected with some providers, the
order might not be kept every time."

Does someone know more about it? Which providers specifically do not
support sorting?

Does this statement apply to the reading provider or the writing provider?

In my case, I never got any correct sorting ...

Thanks,
Andreas

--
Andreas Neumann
QGIS.ORG board member (treasurer)


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info:https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe:https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] EditInPlace Processing script

[Arnaud Morvan]

Note that when heriting form QgsProcessingFeatureBasedAlgorithm,

you do not have to implement processAlgorithm method which is already 
implemented in QgsProcessingFeatureBasedAlgorithm


but should implement the processFeature method.

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 09/04/2020 à 13:35, Arnaud Morvan a écrit :

Hello Matteo,

As far as I know, if your algorithm :

    does not change the table structure (geometry type, fields list 
and their types)


    is an instance of QgsProcessingFeatureBasedAlgorithm (output one 
feature for each input feature)


In brief:

    - If your algorithm output one feature for each input feature:

       => your should inherit from QgsProcessingFeatureBasedAlgorithm.

    - If your outputs features have same geometry type and same fields 
with same types:


       => you can put the flag on your algorithm as it can be used for 
in place editing.


When in place editing is checked in dialog, output features replace 
input features in source instead of being appended to an output.


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 09/04/2020 à 10:53, matteo a écrit :

Hi all,

is it possible to write a custom Processing script that can edit the
input layer in place (with the core functionality of course)?

In the documentation I found the method to check if the algorithm
supports the in place editing [0] but not the method to set the script
to support the functionality.

There are a few python core algorithms with

Thanks for any hint!

Cheers

Matteo


[0]
https://qgis.org/pyqgis/3.10/core/QgsProcessingAlgorithm.html#qgis.core.QgsProcessingAlgorithm.supportInPlaceEdit 


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] EditInPlace Processing script

[Arnaud Morvan]

Hello Matteo,

As far as I know, if your algorithm :

    does not change the table structure (geometry type, fields list and 
their types)


    is an instance of QgsProcessingFeatureBasedAlgorithm (output one 
feature for each input feature)


In brief:

    - If your algorithm output one feature for each input feature:

       => your should inherit from QgsProcessingFeatureBasedAlgorithm.

    - If your outputs features have same geometry type and same fields 
with same types:


       => you can put the flag on your algorithm as it can be used for 
in place editing.


When in place editing is checked in dialog, output features replace 
input features in source instead of being appended to an output.


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 09/04/2020 à 10:53, matteo a écrit :

Hi all,

is it possible to write a custom Processing script that can edit the
input layer in place (with the core functionality of course)?

In the documentation I found the method to check if the algorithm
supports the in place editing [0] but not the method to set the script
to support the functionality.

There are a few python core algorithms with

Thanks for any hint!

Cheers

Matteo


[0]
https://qgis.org/pyqgis/3.10/core/QgsProcessingAlgorithm.html#qgis.core.QgsProcessingAlgorithm.supportInPlaceEdit
___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] Documentation: pyqgis cookbook python code tests

[Arnaud Morvan]

Same is coming for the Python API docstrings: 
https://github.com/qgis/QGIS/pull/9615


Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 24/03/2019 à 22:44, Tim Sutton a écrit :

Yeah this is a really great step forward - thanks Arnaud!

Regards

Tim

On 24 Mar 2019, at 20:52, Luigi Pirelli <mailto:lui...@gmail.com>> wrote:


impressive... tnx Arnoud

Luigi Pirelli

**
* LinkedIn: https://www.linkedin.com/in/luigipirelli
* Stackexchange: http://gis.stackexchange.com/users/19667/luigi-pirelli
* GitHub: https://github.com/luipir
* Mastering QGIS 2nd Edition:
* 
https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition

* Hire me: http://goo.gl/BYRQKg
**


On Sat, 16 Mar 2019 at 14:02, Richard Duivenvoorde 
mailto:rdmaili...@duif.net>> wrote:


See also:

https://docs.qgis.org/testing/en/docs/documentation_guidelines/cookbook_guidelines.html

R

On 16/03/2019 11.05, Richard Duivenvoorde wrote:
> Hi,
>
> For those not aware of Arnaud's work during the Hackfest:
>
> His work makes it relatively easy to test the python snippets
in the
> pyqgisbook. In this way keeping the python snippets up to date with
> code/api changes of QGIS itself is much easier.
>
> We use a Sphinx extension:
> https://www.sphinx-doc.org/en/master/usage/extensions/doctest.html
>
> In short:
> - you run the snippets from within the cookbook rst using
QGIS(libs)
> either from your build/install dir, OR from within a
QGIS-Desktop docker
> instance (same as is being used for automatic tests)
> - so we (have to!) adapt the code in the pycookbook a little to
make it
> testable (adding testsetup:: and testcode:: directives)
> - he slightly changed the way you can build the documentation
(on Linux)
> using one or two extra make (.mk) files.
>
> See:
> https://github.com/qgis/qgis-documentation#testing-python-snippets
>
> I've just tested and this 1 test :-) works flawlessly if you
create a
> 'user.mk <http://user.mk/>' pointing to your QGIS_PREFIX_PATH
>
> So: now we can make all code snippets testable!
> If I am correct he was also busy to get the code examples in the
> doxygen/pyqgiscook work with this too @arnaud ?)
>
> Please try!
>
> Regards,
>
> Richard Duivenvoorde
>
> PS this is Sphinx/Python only, but if somebody jumps in to make
it work
> in a Windows way this would be even nicer. Or we should update
the paver
> way again?
> ___
> QGIS-Developer mailing list
> QGIS-Developer@lists.osgeo.org
<mailto:QGIS-Developer@lists.osgeo.org>
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe:
https://lists.osgeo.org/mailman/listinfo/qgis-developer
>

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
<mailto:QGIS-Developer@lists.osgeo.org>
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org <mailto:QGIS-Developer@lists.osgeo.org>
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


—







*Tim Sutton*

*Co-founder:*Kartoza
*Ex Project chair:*QGIS.org <http://QGIS.org>

Visit http://kartoza.com <http://kartoza.com/> to find out about open 
source:


Desktop GIS programming services
Geospatial web development
GIS Training
Consulting Services

*Skype*: timlinux
*IRC:*timlinux on #qgis at freenode.net <http://freenode.net>

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] Can we remove a confirmation dialog in Refactor Fields algorithm?

[Arnaud Morvan]

Hello Germán

Note that I've already fixed some related problems with 
https://github.com/qgis/QGIS/pull/4964


Calling deleteLater should delete the dialog on next events processing.
Do dialog and child widgets signals should be correctly disconnected.

Arnaud


Le 25/04/2018 à 16:55, Germán Carrillo a écrit :

Hi All,


In the Refactor Fields algorithm dialog there is an "Input layer" 
combo box. Each time the user selects one layer a modal dialog 
(message box) appears, letting her select whether the fields mapping 
should be reset or not. See image: [1].



I'd like to ask you if we can get rid of such modal dialog for these 
reasons:



1) In general, it's good to avoid modal dialogs as they block 
workflows and hamper interaction.


2) Other algorithms don't use such dialog to warn users about 
potential changes in other widgets (e.g., field combo boxes dependng 
on the selected layer).


3) The dialog is only *useful* if the user clicks on the combo box by 
accident and wants to avoid resetting the mapping (a very specific use 
case). However, in that case the combo box shows that a new layer is 
selected, but the mapping won't correspond to it, which is confusing.


4) Follow these steps:

  a. Open QGIS and load a vector layer.
  b. Open the Refactor Fields dialog and press ESC to close it.
  c. Add another vector layer to QGIS.
  d. Close the QGIS project.

As soon as you close the project (e.g., when you want to start a new 
one) the modal  dialog shows up :(  (see this short video [2]).


5) The dialog is triggered too many times when using the "Fields 
Mapper" widget in a custom Processing model (and triggering it from a 
plugin) (see this short video [3]).




So, the question for you is: Can we get rid of it? I've created a PR 
for that :D [4].



Regards,

Germán
---
[1] http://downloads.tuxfamily.org/tuxgis/tmp/ai/reset_field_mapping.png
[2] 
http://downloads.tuxfamily.org/tuxgis/tmp/ai/gif_refactor_fields_qgis.mp4
[3] 
http://downloads.tuxfamily.org/tuxgis/tmp/ai/gif_refactor_fields_plugin.mp4

[4] https://github.com/qgis/QGIS/pull/6857
--
---
   |\__
(:>__)(
   |/
Soluciones Geoinformáticas Libres
http://geotux.tuxfamily.org/
http://twitter.com/GeoTux2
http://about.me/germancarrillo




___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] QGIS Server request extra parameters as variables for expressions

[Arnaud Morvan]

@Patrick :

Yes, I know that it is possible to use FILTER in request, but I'm not 
sure all layers will have the same field name for "floor".


So I would prefer to define a filter in each layer definition in the 
project file using a common variable name.


And I would like to have the same behavior on desktop side (using a UI 
plugin that set the variable value and refresh the canvas).


@Mathias :

Yes after taking a look, WMS dimensions is exactly the idea I have in mind.

In mapserver documentation about WMS dimension I see that the dimension 
field is defined in each layer.


LAYER
  NAME  "lakes_elev"
  METADATA
"wms_title"  "Lakes"
"wms_description""Lakes"
"wms_dimensionlist"  "elevation, text_dimension"
"wms_elevation_item"  "ELEV"
"wms_elevation_extent"  "500, 490, 480"
"wms_elevation_units"  "meters"
"wms_elevation_default"  "500"
"wms_text_dimension_item"  "text_dimen"
"wms_text_dimension_extent"  "first, second, third"
 "wms_text_dimension_units"  "my_units"
 "wms_enable_request"  "*"
  END
  TYPE  POLYGON
  ..

END


Is there something about WMS dimensions in QGIS server roadmap ?


For now we do it in mapserver using normal template variables, for 
example :


LAYER

VALIDATION

        "floor" "[+-][0-9A]+"'

    END

    DATA "geom FROM (SELECT * FROM mytable WHERE etage = '%floor%' OR 
etage IS NULL OR '%floor%' = 'NULL') AS foo USING UNIQUE id USING srid=3945"


END

Note that here the default is no filter at all (all features displayed 
when the parameter is missing), and features with etage == NULL are 
displayed on all floors.


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 23/11/2017 à 14:49, Patrick Valsecchi a écrit :

You could do that using the filter parameter with something like that:
FILTER=layer_name:"floor" = 3

On Thu, Nov 23, 2017 at 2:38 PM, Matthias Kuhn <mailto:matth...@opengis.ch>> wrote:


Like WMS dimension?

http://mapserver.org/ogc/wms_dimension.html
<http://mapserver.org/ogc/wms_dimension.html>

Matthias


On 11/23/17 2:11 PM, Arnaud Morvan wrote:
> Hello,
>
> I would like GetMap and GetFeature request extra parameters to be
> accessible as variables in server.
>
> So we could define, in QGIS project, a variable "floor", used to
> filter layers.
>
> On desktop side, we could have a desktop plugin named FloorSlider to
> change the "floor" value.
>
> And on the server side, the floor value could be passed as extra
> parameter.
>
> So it would be easy to handle multi-floor layers in a project file.
>
>
> Arnaud Morvan
> Ingénieur logiciel
> Tél: +33 (0)4 58 48 20 32 
    >
> Camptocamp France SAS
> 18 rue du Lac Saint André
> Savoie Technolac - Bâtiment Le Dauphin
> F-73370 Le Bourget du Lac
> http://www.camptocamp.com
>
> Le 22/11/2017 à 18:05, René-Luc Dhont a écrit :
>> Hi Arnaud,
>>
>> Is it like updated a QGIS project variables ?
>>
>> Regards,
>>
>> René-Luc
>>
>>
>> Le 22/11/2017 à 15:31, Arnaud Morvan a écrit :
>>> Hello,
>>>
>>> With mapserver, request extra parameters are accessible as
template
>>> variables in mapfile.
>>>
>>> I would like to implement the same possibility in QGIS Server,
extra
>>> parameters may be accessible in project as expression variables.
>>>
>>> For example : I would like to pass in a GetMap request an extra
>>> parameter FLOOR.
>>>
>>> This parameter could be handled by QGIS Server as a variable
value,
>>> so this could be used in the project to filter some layers
using an
>>> expression.
>>>
>>> Do you think this could be acceptable directly in QGIS Server, or
>>> may I have to wrote a plugin.
>>>
>>> Maybe this type of plugin already exists ?
>>>
>>> Note that I'm not familiar with server part.
>>>
>>> Best regards
>>>
>>
>> ___
>> QGIS-Developer mailing list
   

Re: [QGIS-Developer] QGIS Server request extra parameters as variables for expressions

[Arnaud Morvan]

Hello,

I would like GetMap and GetFeature request extra parameters to be 
accessible as variables in server.


So we could define, in QGIS project, a variable "floor", used to filter 
layers.


On desktop side, we could have a desktop plugin named FloorSlider to 
change the "floor" value.


And on the server side, the floor value could be passed as extra parameter.

So it would be easy to handle multi-floor layers in a project file.


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

Le 22/11/2017 à 18:05, René-Luc Dhont a écrit :

Hi Arnaud,

Is it like updated a QGIS project variables ?

Regards,

René-Luc


Le 22/11/2017 à 15:31, Arnaud Morvan a écrit :

Hello,

With mapserver, request extra parameters are accessible as template 
variables in mapfile.


I would like to implement the same possibility in QGIS Server, extra 
parameters may be accessible in project as expression variables.


For example : I would like to pass in a GetMap request an extra 
parameter FLOOR.


This parameter could be handled by QGIS Server as a variable value, 
so this could be used in the project to filter some layers using an 
expression.


Do you think this could be acceptable directly in QGIS Server, or may 
I have to wrote a plugin.


Maybe this type of plugin already exists ?

Note that I'm not familiar with server part.

Best regards



___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

[QGIS-Developer] QGIS Server request extra parameters as variables for expressions

[Arnaud Morvan]

Hello,

With mapserver, request extra parameters are accessible as template 
variables in mapfile.


I would like to implement the same possibility in QGIS Server, extra 
parameters may be accessible in project as expression variables.


For example : I would like to pass in a GetMap request an extra 
parameter FLOOR.


This parameter could be handled by QGIS Server as a variable value, so 
this could be used in the project to filter some layers using an expression.


Do you think this could be acceptable directly in QGIS Server, or may I 
have to wrote a plugin.


Maybe this type of plugin already exists ?

Note that I'm not familiar with server part.

Best regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

[QGIS-Developer] [processing] Allow to use expressions on géométries in RefactorFields

[Arnaud Morvan]

Hello,

A customer ask us to allow, in refactor fields, the use of an expression 
to calculate the output geometry.


I know that the orientation in QGIS 3.0 is to have simple algorithms 
instead of having complex ones,
it that direction, it should make sense to create another algorithm to 
do that.


But another side, having this functionnality in refactorfields allow 
doing complex things without need to create a model.


Does anyone has an opinion on this ?

Regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
18 rue du Lac Saint André
Savoie Technolac - Bâtiment Le Dauphin
F-73370 Le Bourget du Lac
http://www.camptocamp.com

___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] Creating Arrays and Maps from aggregates and relations

[Arnaud Morvan]

Hello,

Note that I start working on array_agg (aggregate as an array).

Regards

Arnaud


Le 04/08/2017 à 21:18, Andreas Neumann a écrit :


Hi,

At the moment an array or map data structure in an expression can only 
be created by using the array() and map() constructor function where 
one has to explicitly include all the values at the creation of the 
data structure. Arrays can also be created from strings with the 
string_to_array function.


Wouldn't it be useful if the arrays could be created by an aggregate 
function and in particular the relation_aggregate function (esp. 
useful for analyzing 1:n relations)?


Something like: aggregate(layer:='countries', aggregate:='array', 
expression:="area") --> array with all areas of the countries


or

relation_aggregate(relation:='cities_in_countries',aggregate:='array', 
calculation:="towns") --> array with all towns that are related to a 
country.


---

Similar for maps:

aggregate(layer:='countries', aggregate:='map', expression:="area", 
mapAttributes:="area, population") --> nested maps with all areas and 
population per country.


or

relation_aggregate(relation:='cities_in_countries',aggregate:='map', 
mapAttributes:="town_name,town_population") --> array with all town 
attributes that are related to a country.


Any thoughts?

Andreas



___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] Creating Arrays and Maps from aggregates and relations

[Arnaud Morvan]

Hi,

As "array" is already a key word in QGIS expressions, to aggregate 
values in an array, I would propose to use the same as postgis : "array_agg"



Le 04/08/2017 à 21:18, Andreas Neumann a écrit :


Hi,

At the moment an array or map data structure in an expression can only 
be created by using the array() and map() constructor function where 
one has to explicitly include all the values at the creation of the 
data structure. Arrays can also be created from strings with the 
string_to_array function.


Wouldn't it be useful if the arrays could be created by an aggregate 
function and in particular the relation_aggregate function (esp. 
useful for analyzing 1:n relations)?


Something like: aggregate(layer:='countries', aggregate:='array', 
expression:="area") --> array with all areas of the countries


or

relation_aggregate(relation:='cities_in_countries',aggregate:='array', 
calculation:="towns") --> array with all towns that are related to a 
country.


---

Similar for maps:

aggregate(layer:='countries', aggregate:='map', expression:="area", 
mapAttributes:="area, population") --> nested maps with all areas and 
population per country.


or

relation_aggregate(relation:='cities_in_countries',aggregate:='map', 
mapAttributes:="town_name,town_population") --> array with all town 
attributes that are related to a country.


Any thoughts?

Andreas



___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [QGIS-Developer] QGIS PyQt version conflict

[Arnaud Morvan]

Hello Michal,

I've also had this error some times ago, I do not remember if I succeed 
with pyuic4, nor the solution.


But IMHO it is simpler to load "ui" files at runtime.
Here is an example from "processing" plugin :

import os
from qgis.PyQt import uic

pluginPath = os.path.split(os.path.dirname(__file__))[0]
WIDGET, BASE = uic.loadUiType(
os.path.join(pluginPath, 'ui', 'DlgConfig.ui'))

class ConfigDialog(BASE, WIDGET):


=> no need to run make or similar to create python files each time you 
change something in your ui files.


load the plugin from the source folder (create a symbolic link from home 
QGIS plugins dir to plugin source dir)

change something in the ui file
reload the plugin using "plugin reloader" and you immediately see the 
result in QGIS.


Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 11/05/2017 à 12:44, Michal Zimmermann a écrit :

Hi,
I'm starting with qgis plugin development on ubuntu-based Mint OS (v 
18.1 Serena). I created a very simple form in Qt Designer 4.8.7 and 
tried to convert it to python file with


pyuic4 -d form.ui -o form.py

This results in

Traceback (most recent call last):
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/port_v2/invoke.py", 
line 34, in invoke

exit_status = driver.invoke()
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/driver.py", line 
59, in invoke

self._generate()
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/driver.py", line 
98, in _generate

self._opts.resource_suffix)
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/__init__.py", line 
173, in compileUi
winfo = compiler.UICompiler().compileUi(uifile, pyfile, 
from_imports, resource_suffix)
  File 
"/usr/lib/python2.7/dist-packages/PyQt4/uic/Compiler/compiler.py", 
line 55, in __init__

CompilerCreatorPolicy())
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/uiparser.py", line 
170, in __init__

self.factory = QObjectCreator(creatorPolicy)
  File "/usr/lib/python2.7/dist-packages/PyQt4/uic/objcreator.py", 
line 94, in __init__

modinfo = plugin_locals["moduleInformation"]()
  File "", line 52, in moduleInformation
  File "/usr/lib/python2.7/dist-packages/qgis/__init__.py", line 26, 
in 

from qgis.PyQt import QtCore
  File "/usr/lib/python2.7/dist-packages/qgis/PyQt/QtCore.py", line 
28, in 

sip.setapi(api, 2)
ValueError: API 'QDate' has already been set to version 1

If I change /usr/lib/python2.7/dist-packages/qgis/PyQt/QtCore.py line 
28 to sip.setapi(api, 1), it works, QGIS starts with a following 
message though:


Couldn't load PyQGIS.
Python support will be disabled.

Traceback (most recent call last):   File "", line 1, in   File 
"/usr/lib/python2.7/dist-packages/qgis/__init__.py", line 26, in 
from qgis.PyQt import QtCore   File 
"/usr/lib/python2.7/dist-packages/qgis/PyQt/QtCore.py", line 28, in   
  sip.setapi(api, 1) ValueError: API 'QDate' has already been set to 
version 2


Python version:
2.7.12 (default, Nov 19 2016, 06:48:10)
[GCC 5.4.0 20160609]

QGIS version:
2.18.7 'Las Palmas', 3b30e6a


What exactly do I have to do to make the both sides work?

--
Michal Zimmermann (zimmi)
WWW: https://www.zimmi.cz


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
QGIS-Developer mailing list
QGIS-Developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] overload closeEvent in AlgorithmDialog.py

[Arnaud Morvan]

Hello Enrico,

Do not change the common dialog closeEvent Method, you should subclass it.

class MyCustomDialog(AlgorithmDialog):

...

You can give a custom dialog to you algorithm using 
GeoAlgorithm.getCustomParametersDialog method.


class MyAlgorithm(GeoAlgorithm):

...

def getCustomParametersDialog(self):

return MyCustomDialog()


Do not forget to call the base AlgorithmDialog.closeEvent method in yours :

class MyCustomDialog(AlgorithmDialog):

def closeEvent(self):

super(MyCustomDialog, self).closeEvent()


This is the way to execute custom tasks in the closeEvent, I do not say 
that this the better way for your job.


Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 23/03/2017 à 18:14, enrico chiaradia a écrit :


Dear (processing) developers,

for my plugin (IO Geopaparazzi), I need to stop the execution of an 
external thread when the user presses the X button of the form.


I though to overload the closeEvent in this way:

from processing.gui.AlgorithmDialog import AlgorithmDialog

[omitted]

class ExportTilesAlgorithm(GeoAlgorithm):

[omitted]

# get the AlgorithmDialog dialog object and overload the closeEvent
AlgorithmDialog.closeEvent = self.closeEvent

[omitted]

def closeEvent(self, evt):
   # necessary to close the thread before it is finished
   self.workThread.stop()

it works perfectly but I'm not sure that it will not induce anomalous 
behaviours in other geoalgorithms as the original function is (from 
AlgorithmDialog.py) :


def closeEvent(self, evt):
QgsMapLayerRegistry.instance().layerWasAdded.disconnect(self.mainWidget.layerAdded)
QgsMapLayerRegistry.instance().layersWillBeRemoved.disconnect(self.mainWidget.layersWillBeRemoved)
super(AlgorithmDialog, self).closeEvent(evt)

Any suggestions?

Thank you

enrico



___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] [Processing] Modeler widget wrapper

[Arnaud Morvan]

Hello all,

I've proposed a refactoring of the modeler algorithm dialog :

https://github.com/qgis/QGIS-Enhancement-Proposals/issues/84

This would come with code simplification, more modularity and port all 
the functionnalities from the algorithm dialog to the modeler algorithm 
dialog.


But as this also has a side effect on the UI, I'd like to have more 
opinion before going deeper on this.


Best regards.

Arnaud Morvan

___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] New feature proposal for the Dissolve tool

[Arnaud Morvan]

Hello Marco,

Note that I'm currently working on a new Algorithm that supersede 
dissolve tool.


Here is how it should look like :

Name : "Aggregate"

Parameters :

* input vector layer

* group_by expression (field name or complex expression, possible 
to add conditions on geometry predicates)


* geometry aggregator : collect (multipart) or combine/union 
(single part)


* Fields aggregators, for each field :

aggregate function, input field, delimiter (for the concatenate 
aggregator)


output filed definition (type, length, precision)

with possibility to add/remove output fields, import the fields 
from an existing layer


The UI should looks like the Refactor Fields algorithm.

With this it should be possible to set a neighborhood condition in the 
group_by expression.


Cheers

Arnaud


Le 22/02/2017 à 10:32, Marco Grisolia a écrit :

Hi,
I recently answered to a question on GIS StackExchange [0]. In this 
question, the asker was looking for a way for dissolving features 
using an adjacency criterion instead of using common attribute field 
(please, follow the link below for a better understanding). This 
feature is already available in the analogous ArcGIS tool when 
the"Create multipart features" option is enabled and I think it could 
be of interest having an additional option like this in the "Dissolve" 
tool main dialog for next releases.
I never submitted a feature request on the QGIS Project site but, 
since there is already a (very rough) code and a general idea on how 
to approach the issue, I didn't know if this was the case of adding it 
there. Otherwise, let me know if I need to add it on the QGIS Project 
site (if this is the case, I'm sorry for wasting your time).

Regards,
Marco


[0] 
http://gis.stackexchange.com/questions/228267/merging-adjacent-lines-in-qgis



___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS 3 master available in OSGeo4W! Youhou....

[Arnaud Morvan]

Thanks Jürgen, for maintaining this

Arnaud

Le 22/02/2017 à 13:58, Jürgen E. Fischer a écrit :

Hi Arnaud,

On Wed, 22. Feb 2017 at 13:39:11 +0100, Jürgen E. Fischer wrote:

On Wed, 22. Feb 2017 at 11:15:46 +0100, Arnaud Morvan wrote:

Where can I found this daily osgeo4w installer ?
  

osgeo4w.osgeo.org.

For the dailies/nightlies.  It has:

qgis-ltr (2.14 release)
qgis-ltr-dev (nightlies of release-2_14)
qgis (2.18 release)
qgis-rel-dev (nightlies of release-2_18)
qgis-dev (nightlies of master)



I thought there would be NSIS installers too at the url you pointed at.
Probably a space issue.  I'll investigate.

There are only weekly snapshots of qgis-dev in OSGeo4W.  One so far - while
there were no master builds, it was switched to qgis-rel-dev (now removed to
make room).


Jürgen



___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] QGIS 3 master available in OSGeo4W! Youhou....

[Arnaud Morvan]

Hi all,

Where can I found this daily osgeo4w installer ?

Is it here : http://qgis.org/downloads/weekly/?C=M;O=D

Last build seems dated from : 14-Nov-2016

Is there a place where we can get more recent build for windows ?

Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 06/02/2017 à 12:20, DelazJ a écrit :

Hi all,

My daily update of osgeo4w installed today qgis-dev (2.99), meaning 
that qgis 3 dev can now be easily installed (and tested?) on Windows.

Wonderful!

Big thanks to you, Jüergen

Regards,
Harrissou


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] newbie build, run question

[Arnaud Morvan]

Hello Chris,

QgsMapLayerRegistry does not exist anymore (since some weeks).
It's methods have been moved to QgsProject to allow having multiple 
project open at the same time.


So the GdalTools plugin does not seems up to date or compatible with 
current master branch of QGIS.
I'm not really aware of that plugin, but I do not see it in QGIS core 
repository.

Maybe this is an obsolete thing, and you could just remove it.

Have you followed the install doc : 
https://github.com/qgis/QGIS/blob/master/INSTALL

For using custom build of QGIS, you should :
Change CMAKE_INSTALL_PREFIX to some folder in your home instead of 
default one in system folder

https://github.com/qgis/QGIS/blob/master/INSTALL#L256

And run QGIS using :
export QGIS_PREFIX_PATH=~/apps/qgis-master
export LD_LIBRARY_PATH=$QGIS_PREFIX_PATH/lib
$QGIS_PREFIX_PATH/bin/qgis

With a fresh build/install you should not get this error.

Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 10/02/2017 à 04:39, Chris Nicholas a écrit :
I successfully built (or so it appeared from cmake…)  QGIS from source 
on Linux according to the instructions of:

https://github.com/qgis/QGIS/blob/final-2_18_3/INSTALL

(….
[100%] Built target qgis_fieldcalculatortest_automoc
[100%] Built target qgis_fieldcalculatortest
[100%] Built target qgis_bench
*cgn@cgn-VirtualBox*:*~/src/QGIS/build-master*$
…)

after make install, now trying to run qgis:

*cgn@cgn-VirtualBox*:*~/src/QGIS/build-master*$ which qgis
/usr/local/bin/qgis


, but I get the following error message …. I’ve looked everywhere but 
can’t find this stuff … where might it be?


advise/RTFMs/etc welcomed…

thanks!!!
Chris

——

Couldn't load plugin GdalTools due to an error when calling its 
initGui() method


ImportError: cannot import name 'QgsMapLayerRegistry'
Traceback (most recent call last):
File "/usr/local/share/qgis/python/qgis/utils.py", line 347, in 
startPlugin

  plugins[packageName].initGui()
File "/usr/local/share/qgis/python/plugins/GdalTools/GdalTools.py", 
line 106, in initGui

  from .tools.GdalTools_utils import GdalConfig, LayerRegistry
File "/usr/local/share/qgis/python/qgis/utils.py", line 647, in _import
  mod = _builtin_import(name, globals, locals, fromlist, level)
File 
"/usr/local/share/qgis/python/plugins/GdalTools/tools/GdalTools_utils.py", 
line 40, in
  from qgis.core import QgsApplication, QgsMapLayerRegistry, 
QgsRectangle, QgsProviderRegistry, QgsLogger, QgsProject

ImportError: cannot import name 'QgsMapLayerRegistry'


Python version: 3.5.2 (default, Nov 17 2016, 17:05:23) [GCC 5.4.0 
20160609]

QGIS version: 2.99.0-Master Master, 2c188b2

Python Path:
/usr/local/share/qgis/python
/home/cgn/.qgis3/python
/home/cgn/.qgis3/python/plugins
/usr/local/share/qgis/python/plugins
/usr/local/share/qgis/python/qgis
/home/cgn/src/QGIS/build-master/python
/usr/lib/python35.zip
/usr/lib/python3.5
/usr/lib/python3.5/plat-x86_64-linux-gnu
/usr/lib/python3.5/lib-dynload
/usr/local/lib/python3.5/dist-packages
/usr/lib/python3/dist-packages
/home/cgn/.qgis3//python


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Plugin requirements

[Arnaud Morvan]

virtualenv has nothing to do with osgeo4w

virtualenv is a tool to create a site-package folder and install 
packages with pip.
you use it in your make file (easy with linux, but should not be much 
more difficult under windows)

you add it to your plugin package
if you add this folder in your python path (sys.path.append) at runtime, 
python will find them.


So virtualenv do not have to work in osgeo4w, you use it only at dev time.

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 08/12/2016 16:28, Régis Haubourg a écrit :

Hi Arnaud,
does virtualenv works in osgeo4w?
Régis

2016-12-08 16:11 GMT+01:00 Arnaud Morvan <mailto:arnaud.mor...@camptocamp.com>>:



Hello Xavier,

here we are used to create a virtualenv at build time (in make
build or make install target), in the plugin folder,
the virtual env site-package folder is added to the python sys
path in plugin __init__.py file before the class factory.

Note that it is necessary to add this virtualenv to the plugin
package (in make package target for example).

With this the plugin package can be big, but is self contained.

Cheers

    Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32 

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com


Le 07/12/2016 15:22, Luigi Pirelli a écrit :

Hi Xavier

1) pip can be installed on OSGeo4W (many references on the gis
stack
exchange and on the web)
2) paver do nothing other than creating python eggs or now you
can use
wheel format, then setting pythonpath starting the plugin
allow you to
load module from local modules... paver is a utility to simplify
packaging

cheers

Luigi Pirelli


**
* Boundless QGIS Support/Development: lpirelli AT boundlessgeo
DOT com
* LinkedIn: https://www.linkedin.com/in/luigipirelli
<https://www.linkedin.com/in/luigipirelli>
* Stackexchange:
http://gis.stackexchange.com/users/19667/luigi-pirelli
<http://gis.stackexchange.com/users/19667/luigi-pirelli>
* GitHub: https://github.com/luipir
* Mastering QGIS 2nd Edition:
*

https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition

<https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition>

**


On 7 December 2016 at 10:44, Xavier Barnada
mailto:xbarn...@gisce.net>> wrote:

Hi Alessandro,

Thank you for the example, I will study paver to see how
it works

Cheers

2016-12-07 9:49 GMT+01:00 Alessandro Pasotti
mailto:apaso...@gmail.com>>:

On Wed, Dec 7, 2016 at 9:43 AM, Xavier Barnada
mailto:xbarn...@gisce.net>> wrote:

Hi all,

I been working on a plugin that uses another
python requirements like
lxml.

The problem it's that I solved the requirements
problem on Linux calling
pip command but this don't work on QGIS with
Windows because Windows don't
incorporate python or pip.

In my opinion QGIS API should provide a way to
install packages but i
supose that it's not so easy as I imagine.

How do you use to solve this problem?  There is
any example of it?  I
searched it on the documentation and don't appear
any reference about how to
do it.

Best reggards
Xavier Barnada



Hi Xavier!

We've considered adding a mechanism for plugin
dependencies, (I've
recently added a metadata for that purpose) but this
seems a bit difficult
to work cross-platform, BTW I'd be very happy to see
some efforts in that
area.

What we normally do is to package and ship
dependencies with the plugin
itself, you can have a look to this example:

https://github.com/boundlessgeo/qgis-geoserver-plugin/blob/master/pavement.py#L48

<https://github.com

Re: [Qgis-developer] Plugin requirements

[Arnaud Morvan]

Note that owslib is distributed with QGIS so you can use :

from owslib.etree import etree

See here for more info : 
https://github.com/geopython/OWSLib/blob/840284bdd1d5e335e7a0d5d276afe4d0b34ac4cf/owslib/etree.py#L36


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 08/12/2016 16:11, Arnaud Morvan a écrit :


Hello Xavier,

here we are used to create a virtualenv at build time (in make build 
or make install target), in the plugin folder,
the virtual env site-package folder is added to the python sys path in 
plugin __init__.py file before the class factory.


Note that it is necessary to add this virtualenv to the plugin package 
(in make package target for example).


With this the plugin package can be big, but is self contained.

Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 07/12/2016 15:22, Luigi Pirelli a écrit :

Hi Xavier

1) pip can be installed on OSGeo4W (many references on the gis stack
exchange and on the web)
2) paver do nothing other than creating python eggs or now you can use
wheel format, then setting pythonpath starting the plugin allow you to
load module from local modules... paver is a utility to simplify
packaging

cheers

Luigi Pirelli

** 


* Boundless QGIS Support/Development: lpirelli AT boundlessgeo DOT com
* LinkedIn: https://www.linkedin.com/in/luigipirelli
* Stackexchange: http://gis.stackexchange.com/users/19667/luigi-pirelli
* GitHub: https://github.com/luipir
* Mastering QGIS 2nd Edition:
* 
https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition
** 




On 7 December 2016 at 10:44, Xavier Barnada  wrote:

Hi Alessandro,

Thank you for the example, I will study paver to see how it works

Cheers

2016-12-07 9:49 GMT+01:00 Alessandro Pasotti :
On Wed, Dec 7, 2016 at 9:43 AM, Xavier Barnada  
wrote:

Hi all,

I been working on a plugin that uses another python requirements like
lxml.

The problem it's that I solved the requirements problem on Linux 
calling
pip command but this don't work on QGIS with Windows because 
Windows don't

incorporate python or pip.

In my opinion QGIS API should provide a way to install packages but i
supose that it's not so easy as I imagine.

How do you use to solve this problem?  There is any example of it?  I
searched it on the documentation and don't appear any reference 
about how to

do it.

Best reggards
Xavier Barnada




Hi Xavier!

We've considered adding a mechanism for plugin dependencies, (I've
recently added a metadata for that purpose) but this seems a bit 
difficult
to work cross-platform, BTW I'd be very happy to see some efforts 
in that

area.

What we normally do is to package and ship dependencies with the 
plugin

itself, you can have a look to this example:
https://github.com/boundlessgeo/qgis-geoserver-plugin/blob/master/pavement.py#L48 



Cheers.

--
Alessandro Pasotti
w3:   www.itopen.it



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

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




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

Re: [Qgis-developer] Plugin requirements

[Arnaud Morvan]

Hello Xavier,

here we are used to create a virtualenv at build time (in make build or 
make install target), in the plugin folder,
the virtual env site-package folder is added to the python sys path in 
plugin __init__.py file before the class factory.


Note that it is necessary to add this virtualenv to the plugin package 
(in make package target for example).


With this the plugin package can be big, but is self contained.

Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 07/12/2016 15:22, Luigi Pirelli a écrit :

Hi Xavier

1) pip can be installed on OSGeo4W (many references on the gis stack
exchange and on the web)
2) paver do nothing other than creating python eggs or now you can use
wheel format, then setting pythonpath starting the plugin allow you to
load module from local modules... paver is a utility to simplify
packaging

cheers

Luigi Pirelli

**
* Boundless QGIS Support/Development: lpirelli AT boundlessgeo DOT com
* LinkedIn: https://www.linkedin.com/in/luigipirelli
* Stackexchange: http://gis.stackexchange.com/users/19667/luigi-pirelli
* GitHub: https://github.com/luipir
* Mastering QGIS 2nd Edition:
* 
https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition
**


On 7 December 2016 at 10:44, Xavier Barnada  wrote:

Hi Alessandro,

Thank you for the example, I will study paver to see how it works

Cheers

2016-12-07 9:49 GMT+01:00 Alessandro Pasotti :

On Wed, Dec 7, 2016 at 9:43 AM, Xavier Barnada  wrote:

Hi all,

I been working on a plugin that uses another python requirements like
lxml.

The problem it's that I solved the requirements problem on Linux calling
pip command but this don't work on QGIS with Windows because Windows don't
incorporate python or pip.

In my opinion QGIS API should provide a way to install packages but i
supose that it's not so easy as I imagine.

How do you use to solve this problem?  There is any example of it?  I
searched it on the documentation and don't appear any reference about how to
do it.

Best reggards
Xavier Barnada




Hi Xavier!

We've considered adding a mechanism for plugin dependencies, (I've
recently added a metadata for that purpose) but this seems a bit difficult
to work cross-platform, BTW I'd be very happy to see some efforts in that
area.

What we normally do is to package and ship dependencies with the plugin
itself, you can have a look to this example:
https://github.com/boundlessgeo/qgis-geoserver-plugin/blob/master/pavement.py#L48

Cheers.

--
Alessandro Pasotti
w3:   www.itopen.it



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

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


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

Re: [Qgis-developer] Field mapper when pasting features

[Arnaud Morvan]

Hi Saber,

Yes, I will try to spent some time on this in Lyon next week.

Cheers

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 08/12/2016 10:36, Saber Razmjooei a écrit :


Hi Arnaud,

As suggested by others, a QEP for mapping fields would be great. Some 
of the Processing tools (e.g. merging layers) can also benefit from that.


Cheers

Saber

*From:*Qgis-developer [mailto:qgis-developer-boun...@lists.osgeo.org] 
*On Behalf Of *Denis Rouzaud

*Sent:* 01 November 2016 08:02
*To:* Arnaud Morvan; qgis-developer
*Subject:* Re: [Qgis-developer] Field mapper when pasting features

Hi Arnaud,

I am really looking forward to seeing such feature implemented in 
QGIS, thanks for bringing up the discussion.


I would also recommend going for a QEP, as it is a largely foreseen 
feature, and I suppose everyone has its own definition and use-case...


From my point of view, we should go for a very generic field-mapping 
mechanism which could be used in many places such as copy-paste. A way 
of reading/saving configuration would be quite useful, like presets.


Best wishes,

Denis

On 10/27/2016 05:52 PM, Arnaud Morvan wrote:

Hello QGIS devs,

I want to add a field mapper dialog when pasting features to layers.

For now when pasting features with attributes names that do not
exists on the destination layer,
those attributes values are lost.

When clipboard fields and destination layer fields are not
strictly the sames,
In QGIS 3.0, I want to popup a new dialog with fields mapper
configurator.

Dialog UI could be similar to Refactor Fields Algorithm, showing a
list of destination fields :

source expression (expression widgets) => destination field

For example, I often use this workflow :
Open a shapefile or autocad file.
Use the refactor fields to get temporary layer with same
fields as destination.
Paste to postgis database layer.

I think having this dialog showing up when needed could really
ease data manipulation from one layer to another.
And I want to give the possibility to use expressions.

Any remarks, do you think I need to open to open a QEP for that ?

Regards





This email and any files transmitted with it are confidential and 
intended solely for the use of the individual or entity to whom they 
are addressed. If you have received this email in error please notify 
the system manager. This message contains confidential information and 
is intended only for the individual named. If you are not the named 
addressee you should not disseminate, distribute or copy this e-mail. 
Please notify the sender immediately by e-mail if you have received 
this e-mail by mistake and delete this e-mail from your system. If you 
are not the intended recipient you are notified that disclosing, 
copying, distributing or taking any action in reliance on the contents 
of this information is strictly prohibited.


Whilst reasonable care has been taken to avoid virus transmission, no 
responsibility for viruses is taken and it is your responsibility to 
carry out such checks as you feel appropriate.


If this email contains a quote or offer to sell products, carry out 
work or perform services then our standard terms and conditions 
<http://www.lutraconsulting.co.uk/downloads/Lutra%20Consulting%20Standard%20Terms%20and%20Conditions.pdf> 
shall apply unless explicitly stated otherwise.


Saber Razmjooei and Peter Wells trading as Lutra Consulting.





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


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

Re: [Qgis-developer] DB manager: Connection to PostGIS from QGIS3 fails

[Arnaud Morvan]

Hello Paolo

I've the same problem, but not found the time to search and fix it.

Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 29/11/2016 12:51, Paolo Cavallini a écrit :

Hi all,
testing QGIS master, I am noticing that the connection to PostGIS
through the browser is OK, whereas the same through DBM attempts to find
a 'g' schema, which is not there, and fails:

SELECT has_schema_privilege('g', 'CREATE'), has_schema_privilege('g',
'USAGE')

Is this a local problem?
All the best.


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

Re: [Qgis-developer] Field mapper when pasting features

[Arnaud Morvan]

Hello Bernd,

Sounds interesting, I put your answer on the list.

Cheers
Arnaud

Le 27/10/2016 19:26, Bernd Vogelgesang a écrit :

Hi Arnaud,
this is great news that you will take care of this feature!
My colleagues (heavy copy/pasters) will be delighted for sure.

If there would be a way to create a "preset" in case you want to copy 
multiple times from one layer to another would be great.
My colleagues often browser through a layers items manually and copy 
things bit by bit to an already existing layer to combine those.
Having to match the fields for every single copy/paste action would 
not help too much in their workflows.


Cheers
Bernd


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

Re: [Qgis-developer] Issue with vector menu and localization

[Arnaud Morvan]

Hi Andreas,

We got the same problem with 2.16 in french.

The processing menu configuration do not support translations.
Once the first run have been made, the config menu entries are set in 
the current translation.


Note that when this will be corrected in transifex and released (Vect&or 
translation for the processing menu entries),
I've added a reset button in processing dialog next to menus node in 
master branch,

but this have not been backported to master_2.
Here is a backport pull request : https://github.com/qgis/QGIS/pull/3686

I think that we should remove the "Vect&or/" prefix in processing config 
for menus.

Adding it dynamically when creating the menu entries would be better.

Chears

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 28/10/2016 08:32, Neumann, Andreas a écrit :


Hi,

There is a localization issue with the vector menu in QGIS 2.18x

When I have english as my GUI language I have one vector menu (nice), 
but when I switch to german I end up with two vector menus: "Vektor" 
(contains plugins) and "Vector" (contains all the Processing stuff 
(former fTools)).


Probably the same in other languages when "Vector" is translated to 
something else ...


Is there something we can do about this situation?

Thanks,
Andreas



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


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

[Qgis-developer] Field mapper when pasting features

[Arnaud Morvan]

Hello QGIS devs,

I want to add a field mapper dialog when pasting features to layers.

For now when pasting features with attributes names that do not exists 
on the destination layer,

those attributes values are lost.

When clipboard fields and destination layer fields are not strictly the 
sames,

In QGIS 3.0, I want to popup a new dialog with fields mapper configurator.

Dialog UI could be similar to Refactor Fields Algorithm, showing a list 
of destination fields :


source expression (expression widgets) => destination field

For example, I often use this workflow :
Open a shapefile or autocad file.
Use the refactor fields to get temporary layer with same fields as 
destination.

Paste to postgis database layer.

I think having this dialog showing up when needed could really ease data 
manipulation from one layer to another.

And I want to give the possibility to use expressions.

Any remarks, do you think I need to open to open a QEP for that ?

Regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

Re: [Qgis-developer] Table Manager discontinuation and missing French translator

[Arnaud Morvan]

Hello

Happy that users know about refactor fields.

+1 to add Refactor Fields in Vector menu processing default entries.
https://github.com/qgis/QGIS/blob/eefeef573d544be5dac3f665b10a4b757762e900/python/plugins/processing/gui/menus.py#L19

Concerning the attribute table button,
note that Refactor Fields comes with a python plugin (if processing is 
installed with QGIS, it is not necessary loaded at runtime).
Not sure that we can easily add a button in attribute table from python 
plugin.


Arnaud

Le 22/09/2016 17:04, Borys Jurgiel a écrit :

Good idea.
What about adding a shorcut to the Vector menu, like it's now with the
algorithms formerly available in fTools?

Borys


  
Dnia czwartek, 22 września 2016 14:32:07 Bernd Vogelgesang pisze:

Just a note on usability:
Though even the Table Manager was not the obvious tool to manipulate
attribute tables, cause it wasn't even installed by default, now the
Refactor Fields function in Processing is even less visible to people.

It would be great if this function could be kind of hard-wired to the
attribute table with a button (like the field calculator, formatting
etc.). I think especially ArcMap-users (but not only) will struggle to
find that function.

Cheers
Bernd

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


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

[Qgis-developer] Backport fixes to exiting release

[Arnaud Morvan]

Hello,

What is the recommended workflow to fix bugs in exiting releases.

Should I make a pull request on each branch (2.16, master_2 and master) ?
Or is there a precise workflow to backport fixes in existing releases ?

Best regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

Re: [Qgis-developer] Python3 and cStringIO in processing

[Arnaud Morvan]

Note that you could also add a symbolic link in .qgis-dev to output dir, 
example :


ln -s /output/python/plugin/* $HOME/.qgis-dev/python/plugins/

And it should load processing plugin from output dir.

No need to make recurrent copies to install dir after that.

Arnaud

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 18/09/2016 13:24, Mathieu Pellerin a écrit :


This error means that your python files have not passed through 2to3. 
Make sure PORT_PLUGINS flag is turned on.


Also right now, the make script only runs 2to3 against the python code 
inside the output directory, it does *not* do so for the python code 
copied to your install location during "make install". You either can 
run QGIS from your output directory, or (as a temporary ugly hack) 
copy the content of the python folder in your output directory onto 
your installation's /share/qgis/python directory.


Hope that helps.


On Sep 18, 2016 6:02 PM, "Richard Duivenvoorde" <mailto:rdmaili...@duif.net>> wrote:


Hi,

testing Python3/Qt5 I get the following message:

  File

"/home/richard/apps/qgis3/master/debug/share/qgis/python/plugins/processing/tools/vector.py",
line 34, in
import cStringIO

According to:

http://stackoverflow.com/questions/30377620/python-3-4-cstringio-vs-stringio

<http://stackoverflow.com/questions/30377620/python-3-4-cstringio-vs-stringio>

there is no cStringIO anymore in py3

do I miss a compile/install step or isn't processing not yet ready to
use yet?

Regards,

Richard Duivenvoorde
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org <mailto:Qgis-developer@lists.osgeo.org>
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
<http://lists.osgeo.org/mailman/listinfo/qgis-developer>
Unsubscribe:
http://lists.osgeo.org/mailman/listinfo/qgis-developer
<http://lists.osgeo.org/mailman/listinfo/qgis-developer>



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


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

Re: [Qgis-developer] Couldn't load plugin fTools

[Arnaud Morvan]

Apparently the release of the plugin is not be compatible with QGIS master,
which is not surprising as API breaks are allowed on master branch for 
QGIS 3.0.


Here is some hints to work with QGIS master :

* do not install QGIS master system wide, better install it in your 
home directory.


* QGIS master use plugins from in $HOME/.qgis-dev/python/plugins 
directory
so you could clone the plugin master branch, build it if 
necessary and install it to $HOME/.qgis-dev/python/plugins directory

and propose pull request on the plugin repo if suitable

Hope this helps.

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 13/09/2016 15:44, Eugenio Trumpy a écrit :


Before make install command I remove qgis dir from /usr/local/share,

after make install (now without error) qgis runs correctly, but I have 
not any plugin.


So I started to reinstall those important for me...someone (e.g 
lizmap) gave me this message in the error box:



The plugin is broken. Python said:

global name 'Qgis' is not defined


hints?


E.



*Da:* Alexander Bruy 
*Inviato:* martedì 13 settembre 2016 15.12
*A:* Eugenio Trumpy
*Cc:* qgis-developer@lists.osgeo.org
*Oggetto:* Re: [Qgis-developer] Couldn't load plugin fTools
I was talking about *installation* directory, not repo.

2016-09-13 16:05 GMT+03:00 Eugenio Trumpy <mailto:frippe12...@hotmail.com>>:


Hi Alexander,


I just clone newly the qgis repo, I started from scratch.

What kind of old file do you mean?


E.




*Da:* Alexander Bruy mailto:alexander.b...@gmail.com>>
*Inviato:* martedì 13 settembre 2016 14.40
*A:* Eugenio Trumpy
*Cc:* qgis-developer@lists.osgeo.org
<mailto:qgis-developer@lists.osgeo.org>
*Oggetto:* Re: [Qgis-developer] Couldn't load plugin fTools
Seems you trying to install QGIS master on the top of existing QGIS
installation without removing old files. fTools plugin was removed in
QGIS 2.16.

Please perform clean install (remove all old files first)

2016-09-13 15:07 GMT+03:00 Eugenio Trumpy mailto:frippe12...@hotmail.com>>:
> Hi all,
>
>
> I just compiled without error qgis master  2.99.0-master d265f33
on ubuntu
> 16.04.
>
> However I got this error on launching the application:
>
>
> Couldn't load plugin fTools due to an error when calling its
classFactory()
> method
>
> ImportError: cannot import name QGis
>
> Traceback (most recent call last):
>   File "/usr/local/share/qgis/python/qgis/utils.py", line 333, in
> startPlugin
> plugins[packageName] = package.classFactory(iface)
>   File
"/usr/local/share/qgis/python/plugins/fTools/__init__.py", line 33,
> in classFactory
> from .fTools import fToolsPlugin
>   File "/usr/local/share/qgis/python/qgis/utils.py", line 642,
in _import
> mod = _builtin_import(name, globals, locals, fromlist, level)
>   File "/usr/local/share/qgis/python/plugins/fTools/fTools.py",
line 33, in
> from qgis.core import QGis
> ImportError: cannot import name QGis
>
>
> Python version: 2.7.12 (default, Jul  1 2016, 15:12:24) [GCC
5.4.0 20160609]
> QGIS version: 2.99.0-Master Master, d265f33
>
> Python Path:
>
> /usr/local/share/qgis/python/plugins/processing
> /usr/local/share/qgis/python
> /home/eugenio/.qgis-dev//python
> /home/eugenio/.qgis-dev//python/plugins
> /usr/local/share/qgis/python/plugins
> /usr/lib/python2.7
> /usr/lib/python2.7/plat-x86_64-linux-gnu
> /usr/lib/python2.7/lib-tk
> /usr/lib/python2.7/lib-old
> /usr/lib/python2.7/lib-dynload
> /usr/local/lib/python2.7/dist-packages
> /usr/lib/python2.7/dist-packages
> /usr/lib/python2.7/dist-packages/PILcompat
> /usr/lib/python2.7/dist-packages/gtk-2.0
> /usr/lib/python2.7/dist-packages/ubuntu-sso-client
> /usr/lib/python2.7/dist-packages/wx-3.0-gtk2
> /home/eugenio/.qgis-dev//python
>
> After that qgis start normally, but all the python plugin were
not loaded.
>
>
> Any hints on this regard?
>
>
> Eugenio
>
>
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
<mailto:Qgis-developer@lists.osgeo.org>
> List info:
http://lists.osgeo.org/mailman/listinfo/qgis-developer
   

Re: [Qgis-developer] Couldn't load plugin fTools

[Arnaud Morvan]

Or, if this is for development, change the CMAKE_INSTALL_PREFIX path to 
install QGIS master in your home, not in your system folders.


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 13/09/2016 14:40, Alexander Bruy a écrit :

Seems you trying to install QGIS master on the top of existing QGIS
installation without removing old files. fTools plugin was removed in
QGIS 2.16.

Please perform clean install (remove all old files first)

2016-09-13 15:07 GMT+03:00 Eugenio Trumpy :

Hi all,


I just compiled without error qgis master  2.99.0-master d265f33 on ubuntu
16.04.

However I got this error on launching the application:


Couldn't load plugin fTools due to an error when calling its classFactory()
method

ImportError: cannot import name QGis

Traceback (most recent call last):
   File "/usr/local/share/qgis/python/qgis/utils.py", line 333, in
startPlugin
 plugins[packageName] = package.classFactory(iface)
   File "/usr/local/share/qgis/python/plugins/fTools/__init__.py", line 33,
in classFactory
 from .fTools import fToolsPlugin
   File "/usr/local/share/qgis/python/qgis/utils.py", line 642, in _import
 mod = _builtin_import(name, globals, locals, fromlist, level)
   File "/usr/local/share/qgis/python/plugins/fTools/fTools.py", line 33, in
 from qgis.core import QGis
ImportError: cannot import name QGis


Python version: 2.7.12 (default, Jul  1 2016, 15:12:24) [GCC 5.4.0 20160609]
QGIS version: 2.99.0-Master Master, d265f33

Python Path:

/usr/local/share/qgis/python/plugins/processing
/usr/local/share/qgis/python
/home/eugenio/.qgis-dev//python
/home/eugenio/.qgis-dev//python/plugins
/usr/local/share/qgis/python/plugins
/usr/lib/python2.7
/usr/lib/python2.7/plat-x86_64-linux-gnu
/usr/lib/python2.7/lib-tk
/usr/lib/python2.7/lib-old
/usr/lib/python2.7/lib-dynload
/usr/local/lib/python2.7/dist-packages
/usr/lib/python2.7/dist-packages
/usr/lib/python2.7/dist-packages/PILcompat
/usr/lib/python2.7/dist-packages/gtk-2.0
/usr/lib/python2.7/dist-packages/ubuntu-sso-client
/usr/lib/python2.7/dist-packages/wx-3.0-gtk2
/home/eugenio/.qgis-dev//python

After that qgis start normally, but all the python plugin were not loaded.


Any hints on this regard?


Eugenio


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





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

Re: [Qgis-developer] Processing qgis:joinattributesbylocation output empty if run from console

[Arnaud Morvan]

Should be fixed with https://github.com/qgis/QGIS/pull/3471

only remove double quotes around "['within']" => ['within'] and it works

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 05/09/2016 19:25, Anita Graser a écrit :

Hi,

qgis:joinattributesbylocation seems to produce correct output when run 
from GUI but empty output when called from the console (just copied 
the command from history):


processing.runalg("qgis:joinattributesbylocation","D:/Documents/Geodata/qgis_sample_data/shapefiles/popp.shp","D:/Documents/Geodata/qgis_sample_data/shapefiles/regions.shp","['within']",0,0,"sum,mean,min,max,median",0,None)

It also runs much faster on the command line, indicating that it's not 
actually processing the data (because it's much slower through the GUI).


Can anyone reproduce? I use OSGeo4W version 2.16.

Best wishes,
Anita




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


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

Re: [Qgis-developer] Processing qgis:joinattributesbylocation output empty if run from console

[Arnaud Morvan]

Hello Anita,

Yes, I can reproduce, with 2.16 on Ubuntu.

As it is probably me that have injected this bug, I will try to fix it.

I will have to take a look at ExtractByLocation and SelectByLocation too.

Regards

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 05/09/2016 19:25, Anita Graser a écrit :

Hi,

qgis:joinattributesbylocation seems to produce correct output when run 
from GUI but empty output when called from the console (just copied 
the command from history):


processing.runalg("qgis:joinattributesbylocation","D:/Documents/Geodata/qgis_sample_data/shapefiles/popp.shp","D:/Documents/Geodata/qgis_sample_data/shapefiles/regions.shp","['within']",0,0,"sum,mean,min,max,median",0,None)

It also runs much faster on the command line, indicating that it's not 
actually processing the data (because it's much slower through the GUI).


Can anyone reproduce? I use OSGeo4W version 2.16.

Best wishes,
Anita




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


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

[Qgis-developer] Incompatible matplotlib and pyparsing on Windows in 2.16.2

[Arnaud Morvan]

Hello,

It seems that the pyparsing version (1.5.2) installed with QGIS 2.16.2 
on Windows in :

 C:/Programmes/QGIS 2.16/apps/Python27/Lib/sites-packages
is too old for the matplotlib one.
matplotlib says : "matplotlib requires pyparsing >= 1.5.6"

Best Regards

Arnaud Morvan
Camptocamp
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] QEP about processing UI refactoring

[Arnaud Morvan]

Hello

I've wrote a QEP about processing UI refactoring.

Comments are welcome.

Best regards.

Arnaud Morvan
Camptocamp
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] QEP about processing UI refactoring

[Arnaud Morvan]

Hello,

I've wrote a QEP about processing UI refactoring :
https://github.com/qgis/QGIS-Enhancement-Proposals/issues/65

Comments are welcome.

Best regards

Arnaud Morvan
Camptocamp
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] [Processing] Parameters and widgets refactoring

[Arnaud Morvan]

I've started the work to define the common widget interface and 
ParametersPanel short methods.
The branch is there : 
https://github.com/arnaud-morvan/QGIS/commit/f28898b181e99b082f7f999a8de4d90430349c40

For now this is a draft, openned for discussion.

Not sure I will have a lot of time to spend on this immediately after 
the hackfest, fortunately 3.0 is not for next week.
I could take some time to keep it alive, and I will probably have some 
time to finish the work in Bonn.
Not sure it will take so much time, the functionnality is there, all to 
be done is dispatching code from ParametersPanel to Widgets, and for 
sure make some tests and bugfix (specially for batch and modeler dialogs).


Arnaud

Le 28/05/2016 09:00, Victor Olaya a écrit :

I like this design a lot. That will give us a cleaner structure and a
code that is easier to understand.

Definitely this has to be in 3.0, although it will be quite a bit of work.

Any additional comment from anyone?

Arnaud, will you at camptocamp be able to dedicate time to this task?
Let's see how we can better organize it

Thanks!

2016-05-27 18:59 GMT+02:00 Arnaud Morvan :

Here the idea to create some class to get processing UI more modular,
flexible and more user friendly.

Actually in processing all parameter's widgets are hardcoded in
ParametersPanel, depending on parameter types. This is not flexible to
create custom parameters and related panels.
On top of this, we have a lot of code duplicated between :
 Standard algorithm dialog/panel
 Batch dialog/panel
 Modeler dialog/panel

A more modular vision could be to create a widget class for each parameter
type.
ParametersPanel would only have to iterate over parameters and create
appropriate widgets.

Using this, it could be easy to create new parameter types and related
widgets without creating a custom dialog and panel class for each case.

Proposed design:
 Parameter class would have a new "widget" attribute (name as string) and
optionally widget options.
 Existing parameters will have a default widget name already given.
 Processing UI will maintain a global a registry for widgets (name =>
class)
 Each processing plugin could register it's own parameters types and
related widgets.
 Each algorithm could override the default widget for a given parameter.

With this design, we have a dissociation between the parameters that are
core related and widgets that are ui oriented. Widgets could be easily
extended without altering the core.

For example, in postgis related algorithms, it could be possible to create
custom widgets, capable of database introspection using existing GeoDB class
to propose drop down lists of schemas and tables or primary keys and
geometry column detection, without altering existing parameter types.

This is a draft, your comments are welcome.

Arnaud Morvan
Camptocamp


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


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

Re: [Qgis-developer] LayerTree group checkbox behavior

[Arnaud Morvan]

Hi DelazJ,

Thanks for raising the difference between my proposal and Franck's ticket.
But this not seems really different.
The ticket says that when unchecking a group all layers should be unchecked.
IMHO another visual sign should be used (layer grayed, like when the 
layer is not visible due to scale).


Regards

Le 27/05/2016 19:32, DelazJ a écrit :

Hi,
I often used the group to group layers about same thematics and 
perform layout by simply checking the group (or viewing maps about 
that thematic). This was before preset and multi styles land in QGIS 
but I suppose I may not be the only one doing this.


Arnaud, I'm not really able to link your comment below to the proposal 
in the issue.


> When I uncheck a layer group, I expect children to be hidden but not 
unchecked.


With your proposal, would that mean that you can have a layer checked, 
but because he is under an unchecked group, none of its features is 
displayed in the canvas? Wouldn't that puzzle users?
If I understand Franck report, with a simple click on group level, 
layers are all unchecked but next click only checks the previously 
checked, not all the layers.

Do I misunderstand something?

2016-05-27 17:08 GMT+02:00 Raymond Nijssen <mailto:r.nijs...@terglobo.nl>>:


Totally agree it should work as requested.

Raymond

    On 27-05-16 17:03, Arnaud Morvan wrote:
>
> Hello,
>
> One of our customer want this ticket to be done :
> https://hub.qgis.org/issues/14547
> For myself, I agree that the actual behavior of layer tree
groups is a
> little disappointing.
>
> When I uncheck a layer group, I expect children to be hidden but not
> unchecked.
> On the same way, when I check a layer group, I expect checked
children
> to be shown, but not those who are'nt checked to be checked.
>
> Does anyone have an opinion on this ?
>
> Regard
>
> Arnaud Morvan
> Camptocamp
> ___
> Qgis-developer mailing list
> Qgis-developer@lists.osgeo.org
<mailto:Qgis-developer@lists.osgeo.org>
> List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

--
Terglobo
Fahrenheitstraat 1
5223 BJ 's-Hertogenbosch
06 25 31 49 83 


___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org <mailto:Qgis-developer@lists.osgeo.org>
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer




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


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

[Qgis-developer] [Processing] Parameters and widgets refactoring

[Arnaud Morvan]

Here the idea to create some class to get processing UI more modular, 
flexible and more user friendly.


Actually in processing all parameter's widgets are hardcoded in 
ParametersPanel, depending on parameter types. This is not flexible to 
create custom parameters and related panels.

On top of this, we have a lot of code duplicated between :
Standard algorithm dialog/panel
Batch dialog/panel
Modeler dialog/panel

A more modular vision could be to create a widget class for each 
parameter type.
ParametersPanel would only have to iterate over parameters and create 
appropriate widgets.


Using this, it could be easy to create new parameter types and related 
widgets without creating a custom dialog and panel class for each case.


Proposed design:
Parameter class would have a new "widget" attribute (name as 
string) and optionally widget options.

Existing parameters will have a default widget name already given.
Processing UI will maintain a global a registry for widgets (name 
=> class)
Each processing plugin could register it's own parameters types and 
related widgets.

Each algorithm could override the default widget for a given parameter.

With this design, we have a dissociation between the parameters that are 
core related and widgets that are ui oriented. Widgets could be easily 
extended without altering the core.


For example, in postgis related algorithms, it could be possible to 
create custom widgets, capable of database introspection using existing 
GeoDB class to propose drop down lists of schemas and tables or primary 
keys and geometry column detection, without altering existing parameter 
types.


This is a draft, your comments are welcome.

Arnaud Morvan
Camptocamp


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

[Qgis-developer] LayerTree group checkbox behavior

[Arnaud Morvan]

Hello,

One of our customer want this ticket to be done : 
https://hub.qgis.org/issues/14547
For myself, I agree that the actual behavior of layer tree groups is a 
little disappointing.


When I uncheck a layer group, I expect children to be hidden but not 
unchecked.
On the same way, when I check a layer group, I expect checked children 
to be shown, but not those who are'nt checked to be checked.


Does anyone have an opinion on this ?

Regard

Arnaud Morvan
Camptocamp
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
List info: http://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Removing overlap creates records with NULL geom

[Arnaud Morvan]

This should fix the problem : https://github.com/qgis/QGIS/pull/3087

Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 20/05/2016 12:12, Paolo Cavallini a écrit :

Hi all,
when digitizing with the Avoid intersection function, if a new polygon
is completely contained in an older one, a record with a NULL geometry
is created. I'm not sure this is a good idea, maybe better to remove the
record and/or warn the user?
All the best, and thanks.


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

[Qgis-developer] [Processing] Table output in .dbf, .csv, ods, xlsx, memory layer, spatialite and postgis

[Arnaud Morvan]

Hello,

I've open a pull request to output tables without geometries using 
VectorWriter and OutputVector classes.
This correspond now perfectly with the input ParameterTable which takes 
all layers, regadless of the geometry column.


Using this, I've improved "Refactor fields" algorithm so we can now 
"refactor" the same way tables with or without geometry.


It also and permit to save tables without geometries in several formats 
using native QGIS providers.

Note that for now, tables can only be saved in "csv" format.

I think it would be good to merge this before freeze.
Could someone take a look at it.

Best regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] Annotations in QgsMapCanvas.saveAsImage

[Arnaud Morvan]

Hello,

I've made a pull request to add annotations in canvas image export here:

https://github.com/qgis/QGIS/pull/3059

This is a small change and we would like it to be merged before freeze.

Could someone make a review ?

Best Regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] Insert/Update in existing layers using Processing

[Arnaud Morvan]

Hello,

Actually in QGIS, updating an existing layer with data coming from other 
layers, files or processing output is not trivial.
=> It is possible to copy/paste features from one layer to another, but 
no user friendly possibility to define fields mapping.
=> Processing algorithms always create new layers, but do not give the 
possibility to update an existing one.
=> "Import to Postgis" algorithm can be used for table creation too, no 
updating options.

=> DB Manager have the same limitations.

We want to improve this in QGIS and we have some propositions (see below).
Let me know what you think about this. Should we create a QEP ?


   1 - Create a new processing algorithm "Import into existing layer"

This algorithm would give the possibility to update / insert into an 
existing layer,

with data coming from file, loaded layer or processing output .
This would give the possibility to wrote, from processing, to any type 
of datasource that support writing.


Here's the inputs :
Source layer (vector): Vector layer for reading
Source identifier (field): Primary key in source layer

Destination layer (vector) : Vector layer for writing
Destination identifier (field) : Primary key in destination layer

Insert new features (checkbox) : Insert new features in 
destination layer.
Update existing features (checkbox) : Update existing features 
in destination layer.
Delete features (checkbox) : Should features, that do not exits 
in source, be deleted from destination ?


No output, as a result, inserted and updated features could be 
selected.


The identifier fields give a one/one relationship between source and 
destination layer.


Regarding the writing method, I see two possibilities :

* Use the edition buffer, the user could see the result before 
saving to disk or database.
Note that this could use a lot of memory in case of treating a 
lot of data.


* Use directly the provider methods.

Maybe we could propose a choice between this two methods.

Thereafter, the user should expect a field mapper from the source to the 
destination (possibly with the use of expressions).

For exporting, we can use the refactorfield as an final output algorithm.
For importing, fields already exists on destination, here it should be 
simpler than refactorfield mapper.
Note that we will not have source/destination fields definition in all 
cases, only for loaded layers.



   2 - Update input layer with "Import into existing layer" after
   existing algorithms


As a second time, we could add an option on some existing algorithm,
with only one input layer,
that do not change fields definition,
that add a checkbox or output option in algorithm dialog, to update the 
input layer as an output.


For example, I want to make a buffer on a selection in a loaded vector 
layer.
In most case, I want this to be done on the source layer, not in a new 
layer.

Processing do not give this possibility for now.

What I propose is not to change existing algorithms, but to add an 
output option in the dialog.
As a result, after running the base algorithm, an import would be done 
using the feature Id as a primary key to update the input layer using 
edition buffer.


This would be user friendly for common operations on existing layers.


   3 - Propose a field mapping for paste operation

This take place on the c++ part of QGIS, in the Edit / Paste features 
(Ctrl+V) action in main menu.
Actually, when we paste features from one layer to another, fields that 
do not exists on destination layer are lost without warning.
We should expect a field mapper dialog here, close to the one in 
previous proposition.


Best regards.

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] Pull request to avoid intersections in reshape tool

[Arnaud Morvan]

Hello,

Can someone take a look at this Pull Request : 
https://github.com/qgis/QGIS/pull/2342
It is proposed for a long time now, and relates with 2 tickets on QGIS 
tracker.


Regards

--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

Re: [Qgis-developer] Create grid in Processing: no polygons?

[Arnaud Morvan]

I use processing directly from master branch (symbolic link in my 
.qgis2/python/plugins folder to QGIS master branch processing folder)
And I run a relatively recent build of QGIS binaries, from master 
branch, under Ubtuntu 14.04.


Arnaud

Le 22/10/2015 11:11, Paolo Cavallini a écrit :

Il 22/10/2015 11:08, Arnaud Morvan ha scritto:

Hi Paolo,

Just tested "Vector grid" and "Create grid" from master,
all seems to works as expected.

Thanks. Tested on more machines, some work, some not.
Trying to understand the reasons.
Which version do you use?
All the best.



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

Re: [Qgis-developer] Create grid in Processing: no polygons?

[Arnaud Morvan]

Hi Paolo,

Just tested "Vector grid" and "Create grid" from master,
all seems to works as expected.

Arnaud

Le 22/10/2015 10:56, Paolo Cavallini a écrit :

Hi all,
with current master from yesterday, the command returns always line
layers, never polygons, regardless of the choice of the user: anyone
confirms?
All the best.


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

Re: [Qgis-developer] Errors with some Processing algorithm

[Arnaud Morvan]

Le 12/10/2015 12:15, matteo a écrit :

Hi all,
I'm doing some test with Processing and I'm facing some problem with
some algorithm:

* Create grid (QGIS).. each time I try to create a grid I receive this
error:
Grid instance has no attribute 'TYPES' See log for more details

while using the same algorithm from the Vector menu of QGIS no problem

This is also a lack due to my recent changes in the translation strings
This can be fixed by https://github.com/qgis/QGIS/pull/2366




* Zonal statistics: input raster (DEM) and a polygon vector (I'm sure
the CRS are the same). This is the error:

in method 'SpatialReference_ImportFromProj4', argument 2 of type 'char
*' See log for more details

same thing of above: the algorithm works if called from the QGIS menu.


Error seems to come from here : 
https://github.com/qgis/QGIS/commit/9fc1b48bf3ff46c2210bdbbc4d9353033da66f6a#commitcomment-13714783

This should also affect HypsometricCurves
@m-kuhn: Can you confirm that we can revert unicode() => str() on this 
specific lines

I'm not confortable with unicode object and binaries binding like osgeo._osr

Arnaud



I'll continue testing..

Thanks

Matteo
___
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] fTools - Processing (again)

[Arnaud Morvan]

Normally, I would be present at next hackfest and interested in working 
on this.


From my point of view, the task would be:

 * Create a customizable vector menu based on Processing like proposed
   by Victor at Nodebo
 * List fTools menu entries, with candidates Processing algorithms, and
   differences that need to be solved.
 * Porting algorithms that has not yet been ported
 * Complete some algorithms with new parameters or createcustom dialogs
   close to fTools ones when need.
 * Get processing plugin fully translated (already working on this)

Arnaud

Le 11/09/2015 13:53, Victor Olaya a écrit :

I think Arnaud also raised the question sometime ago (my fault for not
commenting on that at the time...), and it should be easy to add those
shortcuts, but I haven't had much time lately for working on
Processing.

I wont be attending the next Hackfest, but sounds like a task that can
be accomplished in those days, and I will be happy to help remotely or
take it completely and try to develop a subsystem in Processing to
customize the menus and toolbars with Processing algorithms.

Hope this helps

2015-09-11 12:52 GMT+02:00 Paolo Cavallini :

Il 11/09/2015 11:29, kimaidou ha scritto:

+1 for this !

Hi all,
thanks for raising this point, IMHO a serious one. I'm very much in
favour of removing redundancy. In this case, however, I think we better
be careful before removing fTools, because:

* people are used to it, and for one-shot analyses it is (slightly)
easier to run than Processing (weak argument)
* we do not have enough development resources to make Processing
bulletproof, particularly for 3rd party backends; therefore, we
encounter occasional problems, and we cannot guarantee a smooth user
experience in all cases (strong argument).

First issue can be solved, as suggested, by adding menu shortcuts to
Processing analyses, to mimic existing situation.
Second one is more serious: IMHO we really need a dedicated developer in
this area: any power user (=larger institutions) are willing to take it?
Similar things may be said for GDALTools.
All the best.
--
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
___
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 mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] [Processing] Translate parameter selection content

[Arnaud Morvan]

Could someone take a look at this pull request:

https://github.com/qgis/QGIS/pull/2283

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

Re: [Qgis-developer] [Processing] translate parameter selection dropdown contents

[Arnaud Morvan]

Hello,

without further notice, I will work on processing dropdown parameter 
translation next week, that is:


modify all algorithms in the QGIS group, like this :
https://github.com/arnaud-morvan/QGIS/commit/0286c3d95dffeee65ff7a6b7a69919d700815e4b

and make a Pull Request

Best Regards

Arnaud


Le 26/08/2015 14:50, Arnaud Morvan a écrit :


Hello,

In processing algorithms, I want to translate the ParameterSelection 
dropdown contents.


I've tested two possibilities on SpatialJoin, which both operate,
and I want to get review before editing all algorithms :

First solution: use of QCoreApplication.translate('SpatialJoin', ...) 
in the class header:


https://github.com/arnaud-morvan/QGIS/commit/0daefe838c2b02c8a1bbeffdec551f578424b592 



With this form we have to specify context for each string.
Translation is effectively made at import time because language is 
set before loading processing plugin.


Second solution: use self.tr method in the defineCharacteristics method :

https://github.com/arnaud-morvan/QGIS/commit/0286c3d95dffeee65ff7a6b7a69919d700815e4b 



With this form, we use the self.tr method at runtime on each 
defineCharacteristics call, like it is the case for name, group and 
parameters names.



I think that it is better to make translations at runtime, that is, in 
the defineCharacteristics method.


Does anyone have an opinion on this ?


Arnaud



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

[Qgis-developer] [Processing] translate parameter selection dropdown contents

[Arnaud Morvan]

Hello,

In processing algorithms, I want to translate the ParameterSelection 
dropdown contents.


I've tested two possibilities on SpatialJoin, which both operate,
and I want to get review before editing all algorithms :

First solution: use of QCoreApplication.translate('SpatialJoin', ...) in 
the class header:


https://github.com/arnaud-morvan/QGIS/commit/0daefe838c2b02c8a1bbeffdec551f578424b592

With this form we have to specify context for each string.
Translation is effectively made at import time because language is 
set before loading processing plugin.


Second solution: use self.tr method in the defineCharacteristics method :

https://github.com/arnaud-morvan/QGIS/commit/0286c3d95dffeee65ff7a6b7a69919d700815e4b

With this form, we use the self.tr method at runtime on each 
defineCharacteristics call, like it is the case for name, group and 
parameters names.



I think that it is better to make translations at runtime, that is, in 
the defineCharacteristics method.


Does anyone have an opinion on this ?


Arnaud

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

Re: [Qgis-developer] Fwd: Re: Processing algorithm's groups and names translation.

[Arnaud Morvan]

I've made a pull request here : https://github.com/qgis/QGIS/pull/2213

With the strategy to produce a python file which calls 
QCoreApplication.translate with each string (names, groups).


Could you give me a feedback about this.

Best regards

Arnaud

Le 16/07/2015 14:56, Arnaud Morvan a écrit :


If we make files algnames_.txt
=> translators would have to make pull request for each language to 
translate the name of a new algorithms.
=> It should be difficult to identify not translated names, to update 
text files.


Like alex says: self.tr() method is only applicable to pure-python 
algorithms


So my proposition is to create a python script that generate a 
temporary python file which calls translate() with all names and 
groups (iterating over providers and algorithms).

This file could be parsed normally by pylupdate
And we could call translate at runtime to get displayNames for the 
toolbox
External plugins algorithms will be added to translations (depending 
on plugins installed when running update_ts.sh), but this not seems to 
be blocker.


With simplified interface definition file: algclassification.txt
algorithms not present in this text file do not display in simplified 
interface
if we made one text file for each language, this would cause 
simplified interface to display different algorithms for each language.
This text file can eventually be translated in python to be parsed by 
pylupdate


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 16/07/2015 13:12, Alexander Bruy a écrit :

Hi all,

this topic was raised previously and problem is a bit more complex
then just using self.tr() for algorithm name and group. Such approach
will work only for very limited amount of algorithms with pure-Python
implementation.

But what to do with algorithms, defined via description files? In this
case algorithm name and group written in description files and use
English. Seems we need something more advanced then just
overriding getCommandLineName().


2015-07-15 17:54 GMT+03:00 Victor Olaya :

No need to do that

There is a name property for algorithms (used by the toolbox, not the
cmd line), and then there is the getCommandLineName() method, (which
is used by the command line). The problem here is that, by default,
the getCommandLineName method returns a name that is derived from the
alg name. It is easy to fix that, just override that method for all
algorithms. The problem is that "all" algorithms means hundreds of
them...

It's something that should have been done from the beginning, but I
just assumed that it was a good idea to rely on the base
implementation which uses the human-readable name for the cmd line
name, so there was no need to implement that method ofr every
algorithm, saving some work. Lazy me...

So, in short, the code actually supports what you propose...but
algorithms do not use it,

Let me know if you have questions

2015-07-15 16:01 GMT+02:00 Arnaud Morvan 
:

Thanks for your reponse,

This is functionnal but seems difficult to maintain.
How to verify that every algorithm is translated in each language ?

Why don't add in each algorithm a "title" property

 self.name = '...' # used by command line
 self.title = self.tr('...') # used by toolbox

eventually with a fallback on self.name for compatibility with
algorithms from external plugins.
Like this we could use the normal workflow with transifex.

Does the groups have another use than toobox item text ? can we 
replace :

 self.group = '...'
by
 self.group = self.tr('...')

Arnaud

Le 15/07/2015 12:26, Victor Olaya a écrit :

Arnaud

I added a mechanism for defining alternative names for algorithms.
This allows to provide more informative names than the original ones,
without having to break the cmd interface.

The name replacements are stored in this file:


https://github.com/qgis/QGIS/blob/master/python/plugins/processing/gui/algnames.txt 



If the algorithm is listed there, the alternative name is use.
Otherwise, the original name is used.

Maybe we could have several files like that (algnames_fr.txt,
algnames_es.txt, etc), and just use the one corresponding to the
current language. That would be a very easy solution to implement and
very practical for all translators. It's differnet from how
translation is done now, but I think it will be better for 
Processing,

to be able to rename algorithms without changing their commandline
name

Let me know what you think about this

Cheers



2015-07-15 11:16 GMT+02:00 Arnaud Morvan 
:

Hi,

In processing algorithms, parameters names are translatable (and
completely
translated in french),
but algorithm names and groups couldn't be translated because 
processing

commandline interface use them in english.
On the other side, it would be good for

Re: [Qgis-developer] Fwd: Re: Processing algorithm's groups and names translation.

[Arnaud Morvan]

If we make files algnames_.txt
=> translators would have to make pull request for each language to 
translate the name of a new algorithms.
=> It should be difficult to identify not translated names, to update 
text files.


Like alex says: self.tr() method is only applicable to pure-python 
algorithms


So my proposition is to create a python script that generate a temporary 
python file which calls translate() with all names and groups (iterating 
over providers and algorithms).

This file could be parsed normally by pylupdate
And we could call translate at runtime to get displayNames for the toolbox
External plugins algorithms will be added to translations (depending on 
plugins installed when running update_ts.sh), but this not seems to be 
blocker.


With simplified interface definition file: algclassification.txt
algorithms not present in this text file do not display in simplified 
interface
if we made one text file for each language, this would cause simplified 
interface to display different algorithms for each language.
This text file can eventually be translated in python to be parsed by 
pylupdate


Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

Le 16/07/2015 13:12, Alexander Bruy a écrit :

Hi all,

this topic was raised previously and problem is a bit more complex
then just using self.tr() for algorithm name and group. Such approach
will work only for very limited amount of algorithms with pure-Python
implementation.

But what to do with algorithms, defined via description files? In this
case algorithm name and group written in description files and use
English. Seems we need something more advanced then just
overriding getCommandLineName().


2015-07-15 17:54 GMT+03:00 Victor Olaya :

No need to do that

There is a name property for algorithms (used by the toolbox, not the
cmd line), and then there is the getCommandLineName() method, (which
is used by the command line). The problem here is that, by default,
the getCommandLineName method returns a name that is derived from the
alg name. It is easy to fix that, just override that method for all
algorithms. The problem is that "all" algorithms means hundreds of
them...

It's something that should have been done from the beginning, but I
just assumed that it was a good idea to rely on the base
implementation which uses the human-readable name for the cmd line
name, so there was no need to implement that method ofr every
algorithm, saving some work. Lazy me...

So, in short, the code actually supports what you propose...but
algorithms do not use it,

Let me know if you have questions

2015-07-15 16:01 GMT+02:00 Arnaud Morvan :

Thanks for your reponse,

This is functionnal but seems difficult to maintain.
How to verify that every algorithm is translated in each language ?

Why don't add in each algorithm a "title" property

 self.name = '...' # used by command line
 self.title = self.tr('...') # used by toolbox

eventually with a fallback on self.name for compatibility with
algorithms from external plugins.
Like this we could use the normal workflow with transifex.

Does the groups have another use than toobox item text ? can we replace :
 self.group = '...'
by
 self.group = self.tr('...')

Arnaud

Le 15/07/2015 12:26, Victor Olaya a écrit :

Arnaud

I added a mechanism for defining alternative names for algorithms.
This allows to provide more informative names than the original ones,
without having to break the cmd interface.

The name replacements are stored in this file:


https://github.com/qgis/QGIS/blob/master/python/plugins/processing/gui/algnames.txt

If the algorithm is listed there, the alternative name is use.
Otherwise, the original name is used.

Maybe we could have several files like that (algnames_fr.txt,
algnames_es.txt, etc), and just use the one corresponding to the
current language. That would be a very easy solution to implement and
very practical for all translators. It's differnet from how
translation is done now, but I think it will be better for Processing,
to be able to rename algorithms without changing their commandline
name

Let me know what you think about this

Cheers



2015-07-15 11:16 GMT+02:00 Arnaud Morvan :

Hi,

In processing algorithms, parameters names are translatable (and
completely
translated in french),
but algorithm names and groups couldn't be translated because processing
commandline interface use them in english.
On the other side, it would be good for users to get algorithm names
translated inside toolbox.

Has someone begin to work on this or an idea about how to get names
translated inside toolbox.

Maybe we could help on this ?

Best regards

Arnaud
___
Qgis-developer mailing list

[Qgis-developer] Fwd: Re: Processing algorithm's groups and names translation.

[Arnaud Morvan]

Thanks for your reponse,

This is functionnal but seems difficult to maintain.
How to verify that every algorithm is translated in each language ?

Why don't add in each algorithm a "title" property

self.name = '...' # used by command line
self.title = self.tr('...') # used by toolbox

eventually with a fallback on self.name for compatibility with
algorithms from external plugins.
Like this we could use the normal workflow with transifex.

Does the groups have another use than toobox item text ? can we replace :
self.group = '...'
by
self.group = self.tr('...')

Arnaud

Le 15/07/2015 12:26, Victor Olaya a écrit :

Arnaud

I added a mechanism for defining alternative names for algorithms.
This allows to provide more informative names than the original ones,
without having to break the cmd interface.

The name replacements are stored in this file:

https://github.com/qgis/QGIS/blob/master/python/plugins/processing/gui/algnames.txt

If the algorithm is listed there, the alternative name is use.
Otherwise, the original name is used.

Maybe we could have several files like that (algnames_fr.txt,
algnames_es.txt, etc), and just use the one corresponding to the
current language. That would be a very easy solution to implement and
very practical for all translators. It's differnet from how
translation is done now, but I think it will be better for Processing,
to be able to rename algorithms without changing their commandline
name

Let me know what you think about this

Cheers



2015-07-15 11:16 GMT+02:00 Arnaud Morvan :

Hi,

In processing algorithms, parameters names are translatable (and completely
translated in french),
but algorithm names and groups couldn't be translated because processing
commandline interface use them in english.
On the other side, it would be good for users to get algorithm names
translated inside toolbox.

Has someone begin to work on this or an idea about how to get names
translated inside toolbox.

Maybe we could help on this ?

Best regards

Arnaud
___
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] Processing algorithm's groups and names translation.

[Arnaud Morvan]

Hi,

In processing algorithms, parameters names are translatable (and 
completely translated in french),
but algorithm names and groups couldn't be translated because processing 
commandline interface use them in english.
On the other side, it would be good for users to get algorithm names 
translated inside toolbox.


Has someone begin to work on this or an idea about how to get names 
translated inside toolbox.


Maybe we could help on this ?

Best regards

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

[Qgis-developer] Vector menu - ftools / processing

[Arnaud Morvan]

Hi,

At the hackfest in Nodebo, we talk about ftools plugin replacement by a 
customisable menu based on processing algorithms.


What about this now, is there somebody working on this ? is it started ? 
is it about to be finished ?


If we get some budget for this, is it still time to participate.

Best Regards

Arnaud

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

[Qgis-developer] Publish new plugin SpeadsheetLayers in QGIS official repository

[Arnaud Morvan]

Hello,

I've recently posted a new plugin named SpreadsheetLayer on QGIS 
official repository.
After that I've received some remarks about metadata file content, which 
I've corrected.


So I've reposted the plugin, but it does not seem to have been published 
yet on QGIS plugin repository.


Can somebody take a look at this ?

Best regards.

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

[Qgis-developer] Processing algs documentation and createAlgorithmHelp

[Arnaud Morvan]

Hello

I've see that processing algs documentation have been refactorized, for 
now there is one .rst file for each group instead of one file for each 
algorithm in the past.

This should have been made to reduce the number of files.

On another side, there is some functions in processing.tools.help that 
create the help files for algorithms, the old way, one file for each 
algorithm, with title level shifted compared to new help files.


Accordingly, updating an algorithm documentation is now much more 
complicated.


Furthermore, there is, in QGIS-Documentation master branch an algorithm 
description which should have been merge with old format (one file in a 
group folder).


Regards

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

[Qgis-developer] [QGIS] New processing validity checker before feature freeze

[Arnaud Morvan]

Hello,

I've created a new processing algorithm to check geometry validity at 
the hackfest.

(It is a fTools feature which was missing in processing)

Here is the pull request: https://github.com/qgis/QGIS/pull/2064

I'd like it to be merge before feature freeze.
I think everything is OK to merge, comments are welcome anyway.

I've just rebased it on last successfully tested commit from master to 
avoid conflicts.


Best regards.

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

[Qgis-developer] ftools to processing : geometry validity checker

[Arnaud Morvan]

Today at the hackfest.

I will port the ftool geometry validity checker to processing,
with small implementation differences:
I will add method selector [QGIS|GEOS] as a parameter
I will make two outputs:
one for the valid ones
one for the invalid ones with one supplementary field for the 
reason/comment of the invalid state


Comments are welcome

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

Re: [Qgis-developer] Processing and ftools (Proposal for hackfest)

[Arnaud Morvan]

I should be able to work on this at the hackfest.

IMHO, removing ftools menu entries should be conditioned by the 
translation of groups and names of algorithms in the toolbox.

I rememer that alex started working on this.

Regards

Arnaud

Le 06/05/2015 09:27, Victor Olaya a écrit :

Hi all,

during the last hackfest we discussed about fully integrating ftools
into processing, so eventually we can have all those algorithms in
Processing in the same way as they are now in the ftools plugin, and
remove the latter to avoid redundancy.

Not much was done, but I would like to work on this during the Hackfest.

If anyone has ideas, or anyone that will attend the HF wants to
collaborate on this, please let me know.

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


--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 58 48 20 32

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

Re: [Qgis-developer] Finalising the changelog for 2.8

[Arnaud Morvan]

We've added one entry and altered another to reflect our contributions 
on processing.


Regards

Arnaud

Le 21/02/2015 09:19, Tim Sutton a écrit :

Hi all

We have finished our edits to the visual changelog, the working copy 
is here:


http://changelog.qgis.org/qgis/version/2.8/

Please note that you should not blog / tweet / link to this URL yet, 
Richard will move it onto the main QGIS.org web site from where it can 
be translated and where it is statically rendered.


Before I give Richard the green light to do that, I would like to 
mention that at a recent PSC meeting we agreed to allow feature 
funders and developers to be included in an acknowledgement in the 
changelog. Thus I would request anyone who funded or developed a 
feature listed in the change log to please submit to me a change 
request by email (t...@kartoza.com <mailto:t...@kartoza.com>) with the 
following specific format:


For funders:

**This features was funded by:** [name of your 
organisation](http://website/of/your/organisation)


For developers:

**This feature was developed by:** John Doe, Polly Gone, Sim Boll


Finally, if there is a key feature that you think should be on the 
list, at the above web site, please let me know and I will endeavour 
to add it over the weekend. Please provide clear details about what 
the new feature does etc.


Many thanks

Tim

--
--
Tim Sutton
Visit http://kartoza.com <http://kartoza.com/> to find out about open 
source:

 * Desktop GIS programming services
 * Geospatial web development
* GIS Training
* Consulting Services
Skype: timlinux Irc: timlinux on #qgis at freenode.net 
<http://freenode.net/>

Tim is a member of the QGIS Project Steering Committee
---
Kartoza is a merger between Linfiniti and Afrispatial


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


--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 79 26 57 95

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

Re: [Qgis-developer] What happened to processing.runalg?

[Arnaud Morvan]

I've just meet the same problem

it comes from this commit : 
https://github.com/qgis/QGIS/commit/956c155e8f45cb1a0fc4c5d6204f607f80edc6b0#diff-111
some imports where removes from processing.__init__.py in context of a 
pep8 check

this is the source of the regression

Arnaud

Le 06/02/2015 14:41, Tim Sutton a écrit :

Hi All

Once of our tests in InaSAFE fails against master (see below). Can 
someone tell me where runalg went, and more broadly which parts of the 
processing framework can be relied of as API compatible between 
releases? I guess more and more people will start relying on its API 
and we should make it clear what the best practice is.


Thanks!

Regards

Tim

==
ERROR: Test if line aggregation works
--
Traceback (most recent call last):
  File 
"/home/timlinux/dev/python/inasafe-dev/safe/impact_statistics/test/test_aggregator.py", 
line 596, in test_line_aggregation

impact_layer_attributes=impact_layer_attributes)
  File 
"/home/timlinux/dev/python/inasafe-dev/safe/impact_statistics/test/test_aggregator.py", 
line 338, in _aggregate

aggregator.aggregate(impact_layer)
  File 
"/home/timlinux/dev/python/inasafe-dev/safe/impact_statistics/aggregator.py", 
line 527, in aggregate

qgis_impact_layer, safe_impact_layer)
  File 
"/home/timlinux/dev/python/inasafe-dev/safe/impact_statistics/aggregator.py", 
line 634, in _aggregrate_vector_impact

self._aggregate_line_impact(safe_impact_layer)
  File 
"/home/timlinux/dev/python/inasafe-dev/safe/impact_statistics/aggregator.py", 
line 920, in _aggregate_line_impact

res = self.processing.runalg('qgis:intersection',
AttributeError: 'module' object has no attribute 'runalg'




--
--
Tim Sutton
Visit http://kartoza.com  to find out about open 
source:

 * Desktop GIS programming services
 * Geospatial web development
* GIS Training
* Consulting Services
Skype: timlinux Irc: timlinux on #qgis at freenode.net 


Tim is a member of the QGIS Project Steering Committee
---
Kartoza is a merger between Linfiniti and Afrispatial


___
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] Processing : 2 Pull Requests for merging before feature freeze

[Arnaud Morvan]

For a short description:

With the new parameter type for geometry predicates, the user can choose 
precisely the spatial relation to use in these algorithms :

Select by location
Extract by location
Join attributes by location

With the new fields mapper algorithm, you can completetly reformat layer 
fields with a friendly user interface.
In combination with existing modeler, it becomes relatively easy to 
create full formatted data export models.


Arnaud

Le 23/01/2015 10:12, Arnaud Morvan a écrit :


For now, it seems that all the points raised have been threated
and commits rebased to avoid conflicts

Reminder, here are the links to pull requests:

https://github.com/qgis/QGIS/pull/1753

https://github.com/qgis/QGIS/pull/1759

Regards

Arnaud


Le 21/01/2015 13:14, Victor Olaya a écrit :

Thanks for the reminder. I added comments on both of the PR (once
should be almost ready to merge, the other one might need a bit more
discussion...)

Regards

2015-01-21 12:23 GMT+01:00 Paolo Cavallini :

Il 21/01/2015 12:18, Frédéric Jacon ha scritto:


https://github.com/qgis/QGIS/pull/1753
https://github.com/qgis/QGIS/pull/1759

You can find features description on pull requests pages

We feel both PR carry interesting enhancements and new 
functionnalities

for processing.
Could you please confirm if you think it is OK for merging ?

agreed, interesting stuff, thanks.
Beware build is failing, however:
https://travis-ci.org/qgis/QGIS/builds/47760579
All the best.
--
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
___
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 mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Re: [Qgis-developer] Processing : 2 Pull Requests for merging before feature freeze

[Arnaud Morvan]

For now, it seems that all the points raised have been threated
and commits rebased to avoid conflicts

Reminder, here are the links to pull requests:

https://github.com/qgis/QGIS/pull/1753

https://github.com/qgis/QGIS/pull/1759

Regards

Arnaud


Le 21/01/2015 13:14, Victor Olaya a écrit :

Thanks for the reminder. I added comments on both of the PR (once
should be almost ready to merge, the other one might need a bit more
discussion...)

Regards

2015-01-21 12:23 GMT+01:00 Paolo Cavallini :

Il 21/01/2015 12:18, Frédéric Jacon ha scritto:


https://github.com/qgis/QGIS/pull/1753
https://github.com/qgis/QGIS/pull/1759

You can find features description on pull requests pages

We feel both PR carry interesting enhancements and new functionnalities
for processing.
Could you please confirm if you think it is OK for merging ?

agreed, interesting stuff, thanks.
Beware build is failing, however:
https://travis-ci.org/qgis/QGIS/builds/47760579
All the best.
--
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
___
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 mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] Translation - processing plugin

[Arnaud Morvan]

Hello,

I want to translate processing plugin in french.
But I'm not sure about the exact translation process :

- add self.tr calls in code
- run push_ts.sh script to update english translation file and push it 
to transifex

- translate strings on transifex website
- optionnaly run pull_ts.sh to test my translation, but do not commit 
*.ts files on github


At release time, somebody will run pull_ts.sh to update translation 
files before compiling QGIS


Is this exact, does I miss something ?

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

Re: [Qgis-developer] processing - new geoalgorithme FieldsMapper

[Arnaud Morvan]

and a new one, with new ParameterPredicate class (in progress, but all 
comments are welcome)

https://github.com/qgis/QGIS/pull/1759

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

[Qgis-developer] processing - new geoalgorithme FieldsMapper

[Arnaud Morvan]

I've made a Pull Request in processing plugin with a new GeoAlgorithm.
https://github.com/qgis/QGIS/pull/1753

Can someone take a look at this (volaya or alexbruy), we would like it 
to be pulled before feature freeze.


Objective of this new QGIS GeoAlgorithm is to provide a user friendly 
tool to refactor layer fields.


It takes a standard ParameterVector as input.

From there, the user can :

 * Change some field's names or type.
 * Add or delete some fields.
 * Reorder fields.
 * Calculate new fields based on expressions.

The result is write through a standard OutputVector parameter.

This GeoAlgorithm can be used in modeler.

Arnaud

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

[Qgis-developer] Add all geometry predicates in "XXX by location" processing algorithms

[Arnaud Morvan]

We want to make a pull request on this processing geoalgorithms:
"Select by location", "Extract by location" and "Join by location".

In "Select by location", we have actually three boolean parameters :
- Include input features that touch the selection features
- Include input features that overlap/cross the selection features
- Include input features completely within the selection features
treated in an boolean "or" combination.

Effective treatment depends on type of geometries :
  If layer to select from is a point layer : result = A.intersects(B)
  If one the two layers is a line layer : overlap/cross <=> A.crosses(B)
  else : overlap/cross 
<=> A.overlap(B)


We want to propose all standard predicates to the user :
contains, crosses, disjoins, equals, intersects, overlaps, touches, within

Apparently we could not disable some parameter checkboxes based on 
geometry types without overiding the default generated dialog.


Note : For the disjoins predicate, we have to skip the spatial index filter.

Does anybody have some remarks/objections on this.

Arnaud

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

Re: [Qgis-developer] OGR VRT and SrcSQL

[Arnaud Morvan]

After instrospection,

when adding to QGIS an 'OGR VRT' layer with SrcSql tag,
iterarator with spatial filter return no features.

This seems to be due to OGR.

Does anybody know a workaround about this ?

Arnaud


I am working with OGR VRT files in QGIS.

When I use the "SrcLayer" tag, all works correctly.
But when I use "SrcSQL" tag, QGIS seems to load the file correctly, but
nothing appears in the canvas.

exemple:

  
  test.shp
  SELECT * FROM test
  


all attributes are correctly displayed in the attribute table.
when I use Save As ... to another shapefile, the new file display correctly.
"ogrinfo -la" gives me exactly the same results for "SrcLayer" and
"SrcSQL" tags.
I've also tried the getWKT plugin which give me the same result for
features from the two layers.

I do not understand what  goes wrong : OGR or QGIS ?

Arnaud

___
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] QGIS custom widgets with python plugins

[Arnaud Morvan]

If you install QGIS to custom path, you should set some environment 
variables before calling pyuic4


exemple:

export QGIS_PREFIX_PATH=~/apps/qgis-master
export LD_LIBRARY_PATH=$QGIS_PREFIX_PATH/lib
export PYTHONPATH=$QGIS_PREFIX_PATH/share/qgis/python/:$PYTHONPATH

Personnally, I've added one line in
/usr/lib/python2.7/dist-packages/PyQt4/uic/widget-plugins/qgis_customwidgets.py

pluginType = MODULE
def moduleInformation():
try:
import qgis.gui
return "qgis.gui", dir(qgis.gui)
except ImportError:
*print "impossible to import qgis.gui"*
return "", []

to detect import error

Arnaud


Le 25/11/2014 09:35, Denis Rouzaud a écrit :

Hi Stefan,

To properly compile the UI file using pyuic4, you need to have the uic 
widget plugin installed.


This is the file qgis_customwidgets.py that should be located in 
/usr/lib/python2.7/dist-packages/PyQt4/uic/widget-plugins/


This file can be either installed by the package python-qgis-common or 
using -WITH_CUSTOM_WIDGETS=TRUE if you compile your own QGIS (I 
suppose this requires a make install).


Let me know if this helps.

Best wishes,

Denis

On 14.11.2014 16:13, Ziegler Stefan wrote:

Hi

I sucessfully used QGIS custom widgets (eg. qgscollapsiblegroupbox) 
in qt designer in and then with python plugins with 
"uic.loadUiType()". Using the "traditional" pyuic approach I get an 
missing module error when starting QGIS:


ImportError: No module named qgscollapsiblegroupbox

I'm on Ubuntu 14.04 with an own compiled QGIS master. Do I have to 
put "libqgis_customwidgets.so.2.7.0" somewhere specific? Or is there 
anything else I have to do?


Regards
Stefan

Freundliche Grüsse
Stefan Ziegler
Kantonsgeometer

Amt für Geoinformation
Amtliche Vermessung
Rötistrasse 4
4500 Solothurn

Telefon +41 32 627 75 96
Telefax +41 32 627 75 98
stefan.zieg...@bd.so.ch
http://www.so.ch


___
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


--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 79 26 57 95

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] OGR VRT and SrcSQL

[Arnaud Morvan]

I am working with OGR VRT files in QGIS.

When I use the "SrcLayer" tag, all works correctly.
But when I use "SrcSQL" tag, QGIS seems to load the file correctly, but 
nothing appears in the canvas.


exemple:


test.shp
SELECT * FROM test



all attributes are correctly displayed in the attribute table.
when I use Save As ... to another shapefile, the new file display correctly.
"ogrinfo -la" gives me exactly the same results for "SrcLayer" and 
"SrcSQL" tags.
I've also tried the getWKT plugin which give me the same result for 
features from the two layers.


I do not understand what  goes wrong : OGR or QGIS ?

Arnaud

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

[Qgis-developer] New processing manage fields algorithm

[Arnaud Morvan]

We want to create a new processing algorithm for manage table fields.

This new algorithm would have :
one standard input layer vector parameter
one standard output layer vector parameter

a specific fields mapping parameter composed, in columns
source field name (optionnal)
destination field name
destination type, length, precision
expression (default to the source field name)
the mapping would be initialized, when possible, with the input 
layer fields.


In the dialog :
the first tab would present the normal input/output parameters and 
a button to go to the next tab

the supplementary fields tab would contains the fields mapping

The final goal is to supply a user friendly interface to manage table 
fields :

rename fields
change field types
add new fields based on expression
change fields order
and save the result as a new file, temporary file or memory layer

What do you think about this ? Any remarks ?

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

[Qgis-developer] processing Select By Location operators

[Arnaud Morvan]

Processing algorithms like "Select By Location", "Extract by Location" 
have been enhanced with some new operators :


 * Include input features that touch the selection features
 * Include input features that overlap/cross the selection features
 * Include input features completely within the selection features

Which can be combined in a binary OR manner.


What about exposing each existing geometry operators supported by 
QgsGeometry class :


 * intersect
 * contains
 * disjoin
 * equals
 * touch
 * overlap
 * within
 * crosses


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

Re: [Qgis-developer] Fwd: [Qgis-user] qgis chugiak reshape features tool doesn't work

[Arnaud Morvan]

There is a pull request about this :
https://github.com/qgis/QGIS/pull/1609#issuecomment-57915435

Should be finished before 2.6 release

Arnaud

Le 20/10/2014 10:48, Richard Duivenvoorde a écrit :

Resending this email from the user list to the dev list.

@Umberto: you have more chance to get this done if you sent this to the
developers list. But it is good to advertize this kind of fixes and try
to get this into QGIS in this way! Thank!

@devs: does this look like a viable fix for the problem?

Regards,

Richard Duivenvoorde


 Original Message 
Subject:[Qgis-user] qgis chugiak reshape features tool doesn't work
Date:   Mon, 20 Oct 2014 10:25:18 +0200
From:   umbertofilippo 
To: qgis-u...@lists.osgeo.org



Hello.

I’ve been using qgis since a long time.
With the new version (2.4) I can’t edit my polygons with the reshape
features tool because it rises an “unknown exception”.
I know this is a bug (see hub.qgis.org
<https://hub.qgis.org/issues/10999>). Anyway, it seems they are trying
to fix this problem (on GitHub
<https://github.com/qgis/QGIS/pull/1609/commits>), but, as a
non-programmer user, I cannot figure out if I can actually solve my
issue modifying the code myself (according to the gits) so far.
Since the reshape tool is a must to have tool in my work, I wonder if
someone experienced the same lack and maybe got a hint for me.

  Thanks in advance!

P.S.: I am really sorry to have posted this message many times by my
second (not-registered) mail address (umberto.min...@unimi.it). Please
consider this message and not the others, and sorry again...




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


--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 79 26 57 95

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] utilisation de pg_service.conf sous windows

[Arnaud Morvan]

I want to use the pg service file to define postgis connection 
properties with QGIS, under Windows 7.


I've set the PGSERVICEFILE environment variable to my file path.

But when I try to connect with the service parameter from QGIS, he says 
that the file does'nt exist, with the exact path of the existing file.


Arnaud

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

[Qgis-developer] processing translation

[Arnaud Morvan]

What about to translate the processing plugin ?

There is no doubt that processing is more powerfull/modular than the old 
ftools, but users who don't speek english find that the processing 
module and its toolbox is less accessible than former fTools menu.


Content to translate will be :
- the names, groups
- add a description property (with some illustrations would be the 
best) shown in an new information panel on node selection.


What do you think about this ?

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

[Qgis-developer] HF - ftools issues

[Arnaud Morvan]

Hello

I'm Arnaud Morvan From Camptocamp
It is planned that Camptocamp getthe fTools plugin maintenance
To prepare the hackfest, is there any issues or tasks plannedfor the 
fTools plugin


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

Re: [Qgis-developer] QGIS standalone app - layer tree, legend and co.

[Arnaud Morvan]

Yes, you could use QgsMapLayerRegistry for adding/removing layer and 
QgsLegendInterface for manage groups and layer visiblity,

but new layer tree is more powerfull.

Here is a previous response of Martin on same subject :
http://lists.osgeo.org/pipermail/qgis-developer/2014-June/033529.html

Arnaud

Le 02/07/2014 11:17, Matthias Ludwig a écrit :

Hi,
  
I'am writing a data viewer as python qgis standalone application. Is it possible to use the QgisInterface and other classes (like QgsLegendInterface) for layer handling (adding, removing, activation, etc.)?

Cause I didn't found a way I have written my own layer model based on qts 
QAbstractListModel. It's working more or less, but I would prever a nativ 
solution if available.

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


--
Arnaud Morvan
Ingénieur logiciel
Tél: +33 (0)4 79 26 57 95

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
http://www.camptocamp.com

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

[Qgis-developer] lack of zip_decoder in PIL with Windows 64bits installer

[Arnaud Morvan]

I've meet problems to open PNG files with PIL from plugin on with 
windows 64 bits version of QGIS standalone installer

I don't have zip_decoder in binary _image.pyd (same problem on 2.2 and 2.4)

A workaround is to overwrite PIL in QGIS folder with files from PIL 
standalone installer


I've printed methods PIL.Image.core.*_decoder

on QGIS 2.4 64bits:

bit_decoder
fli_decoder
gif_decoder
hex_decoder
msp_decoder
packbits_decoder
pcd_decoder
pcx_decoder
raw_decoder
sun_rle_decoder
tga_rle_decoder
tiff_lzw_decoder
xbm_decoder

on QGIS 2.2 32bits:

bit_decoder
fli_decoder
gif_decoder
hex_decoder
jpeg_decoder
msp_decoder
packbits_decoder
pcd_decoder
pcx_decoder
raw_decoder
sun_rle_decoder
tga_rle_decoder
tiff_lzw_decoder
xbm_decoder
zip_decoder

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

[Qgis-developer] Settings custom variables overwrite PATH instead of append to it on Windows

[Arnaud Morvan]

In Options/System,
I've added a custom variable to append to environment PATH with value 
";C:\Program Files (x86)\ffmpeg"
when QGIS restart python is disabled because environment PATH is 
strictly equal to ";C:\Program Files (x86)\ffmpeg"


custom variables append correctly on other variables like COMPUTERNAME

tested on Windows7 64 bits with QGIS 2.2 32 bits standalone installer

Seems to work correctly on Linux

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