David Planella's questions

When creating a snap, and according to the syntax of the snapcraft.yaml file, the daemon keyword can be added to specify that the app will be run as a system service. As an example of such declaration:

apps:
  tinyproxy:
    command: ./launcher.sh
    daemon: simple
    plugs: [network,network-bind]

However, what does it exactly mean to be a system service in a snap? That is, what's their lifecycle (start, restart, stop...)? Can I monitor them running in the system with the usual tools (ps, top, etc)?

I've successfully built a snap that I can install on my system.

However, it crashes upon start, and I'm not quite sure where to start debugging that crash.

What are the different approaches to debugging an installed app failing to start?

I have heard about the ability to use snaps in developer mode.

What is this mode and what can I use it for?

Applications packaged as snaps in Ubuntu are installed (mounted) under the /snap/$SNAPPNAME location. Everything under /snap is mounted as a read-only file system, thus applications cannot write to that space, neither in other app's directories nor in their own.

While there is a home interface that snaps can specify to read/write the user's home directory, it is reserved for security reasons and needs to be manually connected (enabled) by the user.

So where can an app inside a snap write its config, data and other files? Are there APIs to access special writeable locations?

I've been trying snappy Ubuntu core on a Raspberry Pi 2 device, and I've heard I can use Snapcraft to do development on it. Now I hear that snaps are available on the desktop and server for Ubuntu 16.04 and other distros.

What is Snapcraft and how can I use it?

I'd like to know how many downloads of a given package in a PPA there have been since it was first published.

I remember there was a bug about it about getting these metrics on the web UI, but as far as I know, it never got implemented.

But I think the number of downloads can nevertheless be obtained via the Launchpad API if I'm the owner of that PPA. Any pointers?

When the Ubuntu SDK preview was announced on the 2nd of January, it was based on the Qt 5 Beta release (as Qt 5 had not yet been released and packaged for Ubuntu). At some point, the Qt 5 release was packaged on a separate PPA and the Ubuntu SDK migrated to be based on the contents of that PPA.

New installs work fine as described on the Ubuntu SDK installation instructions, but I'd like to know how those of us who installed it on release day on the 2nd can migrate to the latest version of the SDK, as the change of PPAs requires a manual upgrade.

This seems to be related to reports of folks who get the "error importing Ubuntu.Components" message when upgrading the SDK.

I've got an issue whereby after each reboot the first time I log into my session it takes ages. That is, it can take up to five minutes from the moment I enter my password until I can use the desktop.

I was trying to use bootchart to debug the issue, but while bootchart itself worked, the graphics output are cropped at 105 seconds, so I cannot even see when compiz starts.

Looking at /etc/init/bootchart.conf, it seems the graphic is cropped indeed after compiz start, but it doesn't seem to be my case

bootchart --format=$format \
    --crop-after=compiz,metacity,mutter,kwin,xfwm4 \
    --annotate=ureadahead,mountall,hostname,hwclock \
    --annotate=Xorg \
    --annotate=gdm-session-worker \
    --output="/var/log/bootchart" "$TARBALL"

Here is some example output.

Is there any way I can see a longer time axis on the graph, at least until the desktop is loaded?

I submitted my app to the Ubuntu Software Centre, and it just got reviewed by the Ubuntu App Review Board (ARB).

I got an e-mail suggesting some fixes, and I'm not sure about the next steps I should follow to fix my app.

My local project has some code that is not working, and I'd like to show it to someone to have a look at and possibly give me a hand with it. This isn't a short snippet or a single file; there are multiple files and directories involved.

How can I best publish my code?

I've come across this bug while writing my app. In short, the app's window size is fixed, which does not work well with smaller screens such as in netbooks.

In those, the lower buttons are off the screen limits and cannot be used. I'd like to take this into account in the user interface, but first of all, I'd like to find out what the standard way to detect screen size is in GTK, if there is one.

So does anyone have any ideas on how to do that?

I'm creating an application that uses symbolic icons from the default theme.

However, there are a few icons that I need that cannot be represented by those from the default theme, so I'm creating my own ones.

What I did was to simply go to /usr/share/icons/gnome/scalable/actions/, copied a few locally into my app's source tree that could serve as a basis, and started editing them.

So far so good. But I've noticed the following: all symbolic icons are of a light grey color when looking at the original .svg file, but when they are put onto a widget, they become darker.

Here's an example, using the /usr/share/icons/gnome/scalable/actions/view-refresh-symbolic.svg icon from the default theme:

• Here's what it looks like when opening the original with Inkscape:

• And here's what it looks like on a toolbar on a running application:

Notice the icon being much darker at runtime. That happens both with the Ambiance and Radiance themes.

I wouldn't mind much, but I noticed it affects my custom icon, whereby parts of it become darker (the inner fill), whereas parts of it remain the same color as the original (the stroke).

So what causes the default symbolic icons to darken and how should implement that for my custom icons?

As an app author, I would like to promote my app in Ubuntu. One nice way to do it is to encourage users to submit their reviews and ratings to the Software Center.

It would be really cool if I could just add a 'Submit review' button in the About dialog of my app to make it extremely easy to write and submit a review.

Is there a Software Center ratings and reviews API that I can access through Python to achieve that?

I'm writing a Python + GObject app that needs to read a non-trivial amount of data from disk upon start. The data is read synchronously and it takes about 10 seconds to finish the read operation, during which time the loading of the UI is delayed.

I'd like to run the task asynchronously, and get a notification when it's ready, without blocking the UI, more or less like:

def take_ages():
    read_a_huge_file_from_disk()

def on_finished_long_task():
    print "Finished!"

run_long_task(task=take_ages, callback=on_finished_long_task)
load_the_UI_without_blocking_on_long_task()

I've used GTask in the past for this sort of thing, but I'm concerned that its code hasn't been touched in 3 years, let alone been ported to GObject Introspection. Most importantly, it's no longer available in Ubuntu 12.04. So I'm looking for an easy way to run tasks asynchronously, either in a standard Python way or in a GObject/GTK+ standard way.

Edit: here's some code with an example of what I'm trying to do. I've tried python-defer as suggested in the comments, but I could not manage to run the long task asynchronously and let the UI load without having to wait for it to finish. Browse the test code.

Is there an easy and widely used way of running asynchronous tasks and get notified when they're finished?

I'd like to write a QtQuick app based on Python (PySide) and QML.

I know Qt apps have got their own translation technology, but I would like to stick to gettext for this one. I'd like to know if it's possible to:

• Mark strings for translation in QML files in a way gettext tools can extract them into a .pot file
• Have gettext translate QML files at runtime.

I know this is done in the Unity 2D code, in C++, so I'm wondering how it can be done with Python.

Note: I'm talking about using exclusively gettext at runtime, not about converting between gettext and Qt Linguist formats.

Recently I've been involved in conversations about creating a Quickly template for a Qt and QtQuick Ubuntu app. The idea is to make it as easy to develop Qt apps from concept to package as it is right now with GTK, on which the Ubuntu application Quickly template is based.

The intention is still to use Python as the base programming language, and the first question that comes to mind is: which Python bindings should we use PyQt or PySide?

I'd like to hear from the folks experienced in both technologies what the pros and cons of each one are, how well maintaned each one is, how well mapped the Qt API is to the bindings, etc.

Thanks!

I've created an application that I'd like to submit through the Ubuntu app developer process. One of the requirements for the process is that when installed, all files go into the /opt/extras.ubuntu.com/<appname> directory.

Through changes in packaging and with help from others I've achieved that, but now I'm facing a new problem: when running from /opt the translations from my app are not being loaded.

What happens is the following: unless specified otherwise, the binary files (MO files) containing translations for an application are loaded from /usr/share/locale (or in the case of Ubuntu, from /usr/share/locale-langpack), but in my app, due to the /opt requirement, they are installed in /opt/extras.ubuntu.com/qreator/locale.

That's in theory fine, as gettext allows specifying an alternative location to load translations, so I add the bindtextdomain call to achieve that:

import gettext
from gettext import gettext as _
gettext.bindtextdomain('qreator', '/opt/extras.ubuntu.com/qreator/share/locale/')
gettext.textdomain('qreator')

That works well up to a point: all messages that are output on the command line then appear translated as expected. But it does not solve the main problem: to load the translations from the UI Glade requires specifying the translation domain once more (the equivalent to the gettext.textdomain('qreator') call above), but it does not let specifying where to load translations from (i.e. the Glade equivalent to bindtextdomain, which sadly does not exist). Here's an extract of the code that I'm using:

builder = Gtk.Builder()
builder.set_translation_domain('qreator')
# There isn't a way to tell glade to load translations from
# somewhere else than /usr/share/locale here
builder.add_from_file(ui_filename)

This essentially means that translations from the UI cannot be loaded if your app is installed in /opt.

I'm pretty much stuck at this point. Any ideas on how to make Glade load translations from /opt/extras.ubuntu.com/qreator/share/locale/ instead of /usr/share/locale?

I've written an application using Python and Glade. The app has got a toolbar with a dark background, and each toolbar button has got a custom image (i.e. no stock icons are used).

I've noticed that when using stock icons instead of custom images as icons, whenever the mouse hovers over the button the button is highlighted, which does not happen with my custom images.

Thunderbird seems to go a step further and rather than highlighting the icons, it inverts their color, which works quite well for a dark theme (see below):

Button not highlighted, normal color

Button highlighted on hover, color inverted

How can I get my custom icon images or the buttons to be highlighted on hover?

I've got a 32-bit Ubuntu installation running on 64-bit hardware. Now that multi-arch has been implemented, I would like to switch to 64-bit without having to reinstall the OS.

This is one of the user stories addressed by the spec:

Shawn installed his system using the 32-bit version of Ubuntu, but his hardware is 64-bit and he wants to switch over. He manually installs the amd64 versions of dpkg and apt, replacing the i386 versions and changing which architecture is used as the default; then he installs the amd64 ubuntu-minimal package; then he installs the amd64 ubuntu-desktop package. Over time the remaining i386 packages are replaced automatically on upgrade.

However, when trying to follow the instructions in there, I cannot find any 64-bit version of dpkg or apt.

Did this user story got implemented in a different way in the final spec, or do I need to do something differently?

In short, how can I switch my 32-bit installation to 64-bit?

I've got a LAMP server available from http://91.xxx.xx.xx/ (just an example IP address), and I'd like to install Wordpress there.

I've gone as far as doing a simple sudo apt-get install wordpress to install Wordpress, but I'm not too sure about the set of best-practice steps after that.

I've read the documentation to install Wordpress in the Ubuntu wiki, but it is out-of-date. Also I couldn't figure it out from the /usr/share/doc/wordpress/README.Debian provided in the wordpress package.

Here are some additional notes of what I'm trying to do with the site:

• The server is running Ubuntu 10.04 LTS
• Wordpress and any other software required should be installed from packages from the official Ubuntu repositories
• I'm not planning to use juju
• Wordpress should be available from http://91.xxx.xx.xx/ directly (i.e. not from http://91.xxx.xx.xx/blog or http://91.xxx.xx.xx/wordpress)

I'd appreciate any help with the clear set of recommended steps after the initial installation of the wordpress package to get the site up and running, as there's lots of info out there, but it seems to be out-of-date.