On 03/21/2013 10:37 PM, Alex Mandel wrote:
On 03/20/2013 11:58 PM, Paolo Cavallini wrote:
Il 21/03/2013 02:26, Alex Mandel ha scritto:
It's user configurable in the Project Properties in 1.8. Did that
change in master?
Or do you mean when projection on the fly is on?
Yes, OTF on, Project
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi all.
On a few machines (QGIS 1.8 on win) defining label size in map units works
correctly
only when zooming in. At lower zooming levels, the labels remain of the same
size on
screen. On Linux qgis master behaves correctly. A local/solved problem,
Hi
On Thu, Mar 21, 2013 at 3:25 PM, Hugo Mercier hugo.merc...@oslandia.com wrote:
Hi,
Sorry if it is not the right place to report.
I have an account on hub.qgis.org. I can add comments to tickets and
create tickets.
But a ticket cannot be assigned to me (which would be the case for
Has anyone tried to use an Instead of trigger in PostGIS from QGIS; Updating
multiple tables in PostGIS from QGIS in one go.
Jakob
--
View this message in context:
http://osgeo-org.1560.n6.nabble.com/Instead-of-trigger-Updating-a-PostGIS-view-from-QGIS-tp5042313.html
Sent from the Quantum
On 03/22/2013 11:53 AM, Jakob Lanstorp wrote:
Has anyone tried to use an Instead of trigger in PostGIS from QGIS; Updating
multiple tables in PostGIS from QGIS in one go.
I did multiple experiments with triggers and QGIS. It is working nicely.
Also QGIS versioning plugin (pgVersion) is using
yes - the PostgreSQL rules and triggers usually work without problems.
I use them often when I deal with views.
Andreas
Am 22.03.2013 11:53, schrieb Jakob Lanstorp:
Has anyone tried to use an Instead of trigger in PostGIS from QGIS; Updating
multiple tables in PostGIS from QGIS in one go.
Hi all,
I am looping over features of a layer using new API (QgsFeatureRequest).
If I modify a feature and do: layer.updateFeature(f), my loop is broken.
Is this a normal behavior?
What should I do if I want to modify features during the loop?
Thanks
Denis
Do you mean your loop is exited when you call updateFeature?
Can you post your code.
- Nathan
On Fri, Mar 22, 2013 at 10:54 PM, Denis Rouzaud denis.rouz...@gmail.comwrote:
Hi all,
I am looping over features of a layer using new API (QgsFeatureRequest).
If I modify a feature and do:
Yes exactly
f = QgsFeature()
iter = vectorLayer.getFeatures( QgsFeatureRequest() )
while iter.nextFeature( f ):
print f.id()
f.setAttribute( field, QVariant( 1 ) )
vectorLayer.updateFeature( f )
Prints only 1
Hi all,
I've seen that metadata.txt is now mandatory for plugins (good thing).
I've seen a side effect on existing installs, when plugin is installed
previously without metadata (FastSQL Layer, Import Project... ), there is no
error message. Installer displays 'installed' status, but plugin is
' rails failed to start'
--
View this message in context:
http://osgeo-org.1560.n6.nabble.com/hub-qgis-org-is-down-tp5042352.html
Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
___
Qgis-developer mailing list
Applications on OS X do not inherit the shell environment. The simplest
solution is to install my psycopg2 python package for the system python. My
QGIS uses the system python so it will automatically find it.
If you compiled QGIS yourself using that python.org python that I see you have
Hi Matthias,
Thanks for your reply.
This should work better, the loop is not exited.
However, my field values are set to NULL using either
changeAttributeValue(s) on vector layer or on data provider.
I tried:
vectorLayer.changeAttributeValue( f.id(), fieldIdx, QVariant(alt) )
If updateFeature is indeed doing feature request inside and breaking the
loop then that is a bug. updateFeature should not affect the feature
request.
- Nathan
On Fri, Mar 22, 2013 at 11:10 PM, Denis Rouzaud denis.rouz...@gmail.comwrote:
Yes exactly
f = QgsFeature()
Also Denis. You can simplify the code a lot like this:
for f in vectorLayer.getFeatures( QgsFeatureRequest() ):
f[field] = 1
vectorLayer.updateFeature(f)
On Fri, Mar 22, 2013 at 11:10 PM, Denis Rouzaud denis.rouz...@gmail.comwrote:
Yes exactly
f = QgsFeature()
Hi Nathan,
On Sat, 23. Mar 2013 at 00:29:03 +1000, Nathan Woodrow wrote:
If updateFeature is indeed doing feature request inside and breaking the loop
then that is a bug. updateFeature should not affect the feature
The misfeature is that there can only be one active iterator per provider (see
Jurgen,
What is the purpose of doing all the extra feature lookup in updateFeature?
Just so that don't change all the attributes everytime? I doubt there would
be that much overhead to really care, I would even say the getFeature
request would be slower then just changing all the attributes each
Hi Denis,
On 22.03.2013 14:43, Denis Rouzaud wrote:
Hi Matthias,
Thanks for your reply.
This should work better, the loop is not exited.
However, my field values are set to NULL using either
changeAttributeValue(s) on vector layer or on data provider.
I tried:
Hi Nathan,
On Sat, 23. Mar 2013 at 01:42:32 +1000, Nathan Woodrow wrote:
What is the purpose of doing all the extra feature lookup in
updateFeature?
To record just the changes on the undo stack and not load of changes that
actually don't change anything.
To me this method is broken an
Hi Jürgen and Nathan,
On Fre 22 Mär 2013 18:00:14 CET, Jürgen E. Fischer wrote:
Hi Nathan,
(... snip )
The bug is that the providers have a notion of active iterators. The
iterators
should be independent, but currently aren't. I think that's going to change
when threading is introduced
On Sat, Mar 23, 2013 at 3:00 AM, Jürgen E. j...@norbit.de wrote:
I don't see a way to currently get what changed without looking up the
current
feature in the process.
Well the feature itself knows what values it started with. Every update in
QgsFeature goes though the setAttribute, etc,
Hi,
Matthias Kuhn and I reworked the QgsColorButton class and refactored it use
across the project. The following is an overview of its new features, a
poll on how to move forward, and a note about a Mac issue.
Features:
* Now one class based off of QPushButton, with two variants: one with an
22 matches
Mail list logo
