Automating the Generation of Python Bindings in QGIS

If you are a PyQGIS developer, you probably already stumbled upon a situation where you needed to look at the signature of a QGIS function, and you dive into the C++ documentation or source code. This is not the friendliest thing if you are not a C++ developer…

Fortunately, @timlinux developed a tool for generating documentation for the Python API, and thanks to great work of @RouzaudDenis and @_mkuhn, it is now possible to generate the sip files automatically from the header files. Before, the sip file had to be created manually by the developer, which means it was subject to human mistakes (like for instance forgetting to port a function).

To support this automated generation for the entire source code, it is necessary to annotate the headers with the relevant SIP notations. As an example, I may not want to port the dataItem_t pointer to Python, in which case I would annotate the header with SIP_SKIP.

typedef QgsDataItem *dataItem_t( QString, QgsDataItem * ) SIP_SKIP;

A good place to start, before adding automated SIP generation to headers, is to read the SIP Bindings section, of the QGIS coding standards, or even having a look at qgis.h, where all these annotations(macros) are defined.

The sip files which are currently not generated automatically, are defined in autosip_blacklist.sh, so if you want to automate a sip, the first thing would be to remove that file from this list. Then you may run the sipify_all.sh script, which will scan through all the files which are not blacklisted and generate the sip files for them. This will generate a new sip file for the one you removed from the blacklist. If you compare the new file with the old file, in most of the cases the signatures of the functions won’t change. In that case, you don’t need to do anything. If you find differences in the signature, it is because there are some special instructions between “/”, like /Factory/ which you need to support. To do that, you need to add the appropriated annotations in the header file; in that case, do not forget to add the “qgis.h” file, in order to support the macros.

#include "qgis.h"

When you finish annotating the file, run the script again and check if the old and new sip match. If they do match, then you have supported the automated generation of the sip file; otherwise, you need to go again to the header and check what is missing.

There are still many sip files left to be automated, so we encourage you to contribute to QGIS with PR on this matter! 🙂

Advertisement

Easing the Creation of Metadata in QGIS

In a previous blog post, I presented QGIS enhancement #91, which aims at providing the infrastructure in QGIS to author, consume and share standards-based metadata (e.g.: ISO).

In this post I would like to focus on a specific WP which aims at easing the task of authoring metadata. Let’s face it: this is the long face many people put on, when they are told they need to create metadata.

We would like to at least reduce this effort, by letting users create a metadata template, which would then be reused across the project, enabling the automated population of metadata. Having the repetitive bits out of the way, they could focus on the fun parts: creating specific layer metadata, and of course, working with the data.

More specifically, this WP covers the support to two events:

• Filling of the template, which would then be associated to the project; this can happen in one tab of the project settings.
• Automated population of metadata for a layer, based on this template; this can be triggered through the layer properties, or when the user loads/creates a layer.

The mockups bellow illustrate these application scenarios.

Creation of the Metadata Template

Application of the Metadata Template

This template would be based on the QGIS internal schema, developed on WP1. The fields presented on the following mockups are only examples, based on the Dublin Core schema.

One interesting enhancement would be to support the import/export of this template, so that it could be shared across an organization. One user could also have multiple templates, according to the layers he was working on (see image bellow). Both these scenarios would require detaching the template from the project file and storing it in an external format.

Support to External Templates

We envision this WP to deliver the following:

• UI and handlers for creating the template.
• UI and handlers for applying the template.
• UI and handlers for exporting/importing the template (optional).

I will submit a proposal for these developments to the QGIS Grant Applications Programme and will be looking forward to having the support of the community to ease the creation of metadata in QGIS

Welcoming the QGIS Metadata Store

Support to standards based metadata (e.g.: ISO) has been greatly missed in QGIS. We would like for that to no longer be the case in QGIS 3.0, with this enhancement proposal.

This blog post focuses on WP3: “QGIS Metadata Store”, which will introduce an external physical format for storing metadata internally in QGIS. The goal is to support portability, enabling users to share their layer metadata, even in offline scenarios. This WP will build directly on the outputs of WP1, which will define an “internal metadata schema” and WP2, “QGIS metadata API”, which will encode/decode from the internal schema to the supported schemas.

The final goal is for QGIS to support two types of metadata stores: remote and local. In this WP we will focus on local stores, only.

In the diagram below we depict the inheritance model for metadata stores, where an abstract metadata store will have a polymorphic behavior, according to the particular data format. For instance in the case of a PostgreSQL DB, the method “save” will create a table on the database, whether in the case of a Shapefile, it would create an XML file.

Some formats, such as text files, can be more limited than others. As an example, searches in text files can be quite slow. For that reason, we will create a “prime” format, the “QGIS metadata store”, which can accompany more restrictive formats.The prime format will be an SQLite database, because of its lightweight, and because it is well-known within the QGIS community.

As the goal is to support all these different formats in the future, we will design an infrastructure to accommodate that, but in this first iteration we will focus on the simple use case of creating an xml file, and an SQLite data store.

The metadata contents will be passed by the metadata API. In this WP we will implement format translation, but not schema translation.

Along with these developments we will implement a user interface to allow the user to configure serialization/deserialization behavior, e.g.: in which format we should write metadata, and where.

The QGIS metadata store will be synced with any changes that we apply to the metadata. In the moment that we export metadata into XML, it will write those changes to the XML file.

Metadata search will also be polymorphic, according to the data format. In this iteration, as a proof of concept, we will implement some simple text search, which will enable users to query their metadata.

We envision this WP to deliver the following:

• An infrastructure to accommodate the external storage of metadata in QGIS, fully implemented for the use case of XML files.
• Support for searching the metadata store.
• UI for saving/loading metadata.

I will submit a proposal for these developments to the QGIS Grant Applications Programme and will be looking forward to having the support of the community to welcome the QGIS metadata store 🙂

Go On board with a GeoNetwork Container

GeoNetwork is a FOSS catalog for geospatial information. It is used around the world by organizations such as FAO, the Dutch Kadaster or Eurostat, just to mention a few.

As any software service, it may not be trivial to install and configure, which may put people away for giving it a try. This could change with docker.

Docker, which could be defined in a nutshell as infrastructure as code, automates the deployment of Linux applications inside software containers. It relies in a technology, LXC, which provides operating-system-level virtualization on Linux. In less than four years it experienced a massive adoption by the software community, and it has already been taken to production in many use cases.

The docker hub is a massive repository for ready-to-use images. You can find anything from web servers to databases, or even actual operative systems. With a docker pull at the tip of your fingers, you can have them running in your computer in a matter of minutes (depending on your internet connection).

Anyone can upload their docker images to docker hub, but there are some images which are released “officially”.
Official images sources live in the docker repositories, and they are considered good to use (and reuse), because they implement docker best practices, and therefore their code can be seen as an example. They are also heavily documented according to some standards, and they go through a security audit.

Although there are a couple of geonetwork images on the docker repositories, there is no official image yet, so I decided to create one. While the image goes through the approval process, I decided to publish it anyway, so that anyone can benefit from it in the meantime.

These images provides the two latest releases of geonetwork (3.0.5 and 3.2.0), as well as the previous release (3.0.4). By default, geonetwork runs on a local h2 database, but I created a variant which can use a postgresql database as backend, either running on a container or on a bare metal server. This should make it more fit for production.

You can read more about these and other features, such as setting and persisting the data directory, on the docker hub page.

Once the official images get released I will make an announcement here. But in the meantime, there is no excuse to not start playing with geonetwork:

docker pull geocat/geonetwork

Have fun with docker & Geonetwork ! 🙂

Watching a Server through a Container

Lately I have been working a lot with docker, the new kid on the block on cloud computing, which is winning the heart of sysadmins, as well as developers.

The main idea is to setup a Spatial Data Infrastructure, something that has been at the core of other projects such as Georchestra.

Unfortunately having something running on a server is normally not a complete smooth experience, and this sets the ground for the need of a monitoring service.

After searching a bit, I found NewRelic, which provides monitoring on a service basis. I really liked the advanced functionality and the completeness of the dashboards, so it was not hard to convince myself to try it.

NewRelic provides two types of monitoring: application monitoring, and server monitoring, which is what I will cover today on this post. The server monitor, is basically a daemon that runs on the server and collects statistics about various metrics, such as: memory usage, CPU usage, bandwidth, etc. But what really caught my eye about this solution, was the ability to monitor the docker daemon and the different containers that run within it.

Unfortunately this functionality appears to be broken for docker 1.11 (my current version), but with the help of the NewRelic engineers I was able to apply a workaround.

My next step was to dockerize this solution. After all, wouldn’t it be great to spin another container in my SDI, that would monitor the other containers AND the server?

The bad news is that the existing images of Newrelic’s server on docker hub do not implement the workaround. So I went and implement my own image.

You can pull this image from the repository, with:

docker pull doublebyte/newrelic_sysmond

Then you can run it with:

 docker run -d \
–privileged=true –name nrsysmond \
–pid=host \
–net=host \
-v /sys:/sys \
-v /dev:/dev \
–env=”NRSYSMOND_license_key=REPLACE_BY_NEWRELIC_KEY” \
-v /var/run/docker.sock:/var/run/docker.sock \
-v /var/log:/var/log:rw \
newrelic_sysmond

The privileged flag and the bindings to the host directories are necessaries, because we need to be able to watch the docker daemon, and collect the docker metrics.

Note that if you also want to collect memory stats of the containers, it is necessary to configure it in the kernel. The procedure is explained on the docker documentation, but it really comes down to updating the bootloader and restarting. In the case of grub, you would need to add this line to /etc/default/grub:

GRUB_CMDLINE_LINUX=”cgroup_enable=memory swapaccount=1″

Then you need to update grub with:

update-grub

After a restart of the server, the docker memory statistics should be present on the server dashboard:

Spatial Data Mining

Social media streams may generate massive clouds of geolocated points, but how can we extract useful information from these, sometimes huge, datasets? I think machine learning and GIS can be helpful here.

My PechaKucha talk at DataBeers : “Visualizing Geolocated Tweets: a Spatial Data Mining Approach”.

Cluster Explorer Demo

Cluster explorer is a piece of software I was working on, which blends machine learning and GIS to produce a cluster visualization on top of a virtual globe.
The virtual globe uses NasaWorldWind, a Java framework for 3D geo-visualization based on OpenGL, while the clustering algorithms use the ELKI Data mining framework for unsupervised learning.
The tool allows to cluster a bunch of points (for instance geolocated Tweets) using one of two algorithms (or both), and explore the contextual information provided by the geographic layers (e.g.: OSM, Bing).

Visualizing Micro-blogging Data

This time I’m at the Workshop: “Data Visualization from Different Universes”, presenting some work regarding the visualization of Tweets, using a combination of machine learning and GIS.

Big Spatial Data

And here is the actual talk, which contains an overview of approaches regarding Data, Big Data, and Big Spatial Data, in these fast pace days…

Driving Spatial Analysis Beyond the Limits of Traditional Storage

My presentation on the “Conference on Advanced Spatial Modelling and Analysis” consisted in some thoughts regarding Big Spatial Data, and how to handle them in terms of modern technologies.

It was great to see such a motivated crowd from all generations, and to get to know the research developed by CEG, in topics such as Agent Based Modelling and Neural Networks. It was also great to talk again to Arnaud Banos, from the Complex System Institute of Paris Ile-de-France (ISC-PIF).