As I wrote before, Oslandia kept their promise, and during the 8th QGIS developer meeting, the Atlas plugin was integrated into the QGIS print composer. This has been made possible thanks to the financial support (although not the full amount has been raised) of the following individuals and organisations:
Agence de l’eau Adour-Garonne
City of Uster, Switzerland
Spencer Gardner
Giovanni Allegri
John C. Tull
Bill Williamson
Ujaval Gandhi
The workflow remains the same: create a print layout and specify a coverage layer that will be used to generate the atlas. In addition, it is now possible to create complex labels using the full power of QgsExpression and the attributes of the coverage layer.
Let’s go through the process of making an atlas step by step.
The 8th QGIS developer meeting, also known as the hackfest, has come to an end.
This time there were no presentations, which had been a feature of the previous two meetings, and this had a positive impact on the results: more than 120 commits were made, and about 60 bug reports were closed. Among the most notable changes:
support for geometryless tables in WFS data provider
ability to open layer properties from the Identify Results window
Atlas plugin integration (tests and documentation included)
automatic detection of the transparency band for raster layers
and much more
On top of that:
system update has been performed on the qgis.org server
changes to the documentation structure have been agreed upon, and the process of integrating and restructuring the repository has begun
updated plugin metadata requirements and fixed a number of bugs in the plugins.qgis.org application
a rating system for plugins has been implemented
According to the participants, this meeting was one of the most productive, which was greatly facilitated by the venue — Villa Vogelsang.
QGIS plugin developers should note that it was decided at the hackfest to make changes to the plugin metadata. Namely:
added mandatory author field
added mandatory email field
the authorName field is declared deprecated and is ignored when parsing metadata
The new fields must be present in both the __init__.py file and the metadata.txt file. The purpose of this change is to separate the actual author of the plugin from the plugin maintainer. For plugins already uploaded to the repository, these metadata fields will be filled in automatically using information from LDAP, but for all new plugins, these fields should be set by the developer.
Morning. I’m working in the office when the silence is broken by a ringing telephone. A call from an unknown number. I answer. A man asks if it’s me, and when I say yes, he continues, “It’s FedEx bothering you”. O_o. “Your package has arrived. Should we deliver it to your place, or will you pick it up from our office? My eyes started to twitch. What package? From where? I asked for more details. Hm… they seem to be correct, but that didn’t make things any clearer.
In about 30 minutes, the package was on my desk. I checked the return address and didn’t understand anything. It was only when I took out the contents that it all made sense. There was this T-shirt
Read it. Wonderful book: funny cases, interesting stories and thoughts. Much of what was said is still relevant today, and some things even have gotten worse:
One other thing I could never get them to do was to ask questions. Finally, a student explained it to me: “If I ask you a question during the lecture, afterwards everybody will be telling me, ‘What are you wasting our time for in the class? We’re trying to learn something. And you’re stopping him by asking a question’.” It was a kind of oneupmanship, where nobody knows what’s going on, and they’d put the other one down as if they did know. They all fake that they know, and if one student admits for a moment that something is confusing by asking a question, the others take a highhanded attitude, acting as if it’s not confusing at all, telling him that he’s wasting their time.
Attempts to import SEXTANTE from its SVN repository into the QGIS repository on GitHub with the history intact are a thing of the past. Also in the past were numerous fixes for bugs that appeared after the import, writing build rules, and fixing issues caused by the directory structure change. SEXTANTE is now officially included in QGIS as a core plugin.
Thanks to Victor, Tim, Camilo. Well, and me too :-).
Together with Victor, we have implemented in SEXTANTE the ability to save output to any OGR-compatible format or memory layer and to set the output data encoding. The only thing missing is the ability to set dataset and layer creation options. In the future, we plan to add support for saving results to spatial databases.
In the meantime, we are fixing bugs and making sure that all the algorithms are fully functional. In particular, there are 16 algorithms in fTools that are still untested.
Yesterday this blog turned 5 years old. After that time, I can say that the idea of blogging was not that stupid. And even though I don’t write regularly and it’s more for myself, maybe some of the posts have been useful or just interesting to readers.
I don’t know how many readers I have, but thank you for sticking with me.
Reviewed 19 fTools tools in two groups: cleaned up the code, fixed some bugs, optimised a bit. There is still as much to do. Then test and if no critical bugs appear, I will commit.
