Michal Čihař - Blog Archives for Gammu

Weblate 2.13

Weblate 2.13 has been released today pretty much on the schedule. The most important change being more fine grained access control and some smaller UI improvements. There are other new features and bug fixes as well.

Full list of changes:

  • Fixed quality checks on translation templates.
  • Added quality check to trigger on losing translation.
  • Add option to view pending suggestions from user.
  • Add option to automatically build component lists.
  • Default dashboard for unauthenticated users can be configured.
  • Add option to browse 25 random strings for review.
  • History now indicates string change.
  • Better error reporting when adding new translation.
  • Added per language search within project.
  • Group ACLs can now be limited to certain permissions.
  • The per project ALCs are now implemented using Group ACL.
  • Added more fine grained privileges control.
  • Various minor UI improvements.

If you are upgrading from older version, please follow our upgrading instructions.

You can find more information about Weblate on https://weblate.org, the code is hosted on Github. If you are curious how it looks, you can try it out on demo server. You can login there with demo account using demo password or register your own user. Weblate is also being used on https://hosted.weblate.org/ as official translating service for phpMyAdmin, OsmAnd, Aptoide, FreedomBox, Weblate itself and many other projects.

Should you be looking for hosting of translations for your project, I'm happy to host them for you or help with setting it up on your infrastructure.

Further development of Weblate would not be possible without people providing donations, thanks to everybody who have helped so far! The roadmap for next release is just being prepared, you can influence this by expressing support for individual issues either by comments or by providing bounty for them.

Tests coverage from Windows builds revisited

Few months ago, I've written about getting coverage information from many platforms into one report. That approach worked, but I've always felt guilty for pushing almost thousand of files to Codecov.

This week I finally found time to revisit this and make it work better and faster. Actually just uploading these files took about 30 minutes. Together with slow tests execution (which took about 30 minutes as well), we were reaching AppVeyor build time limits and builds did timeout quite often, what is not really nice result.

I've started with rewriting the wrapper used to execute OpenCppCoverage. I've originally used Python for that, which is nice, but I thought the overhead must be noticeable. As it is not possible to execute Python script directly from CTest, it was wrapped in simple bat file, adding another overhead. Reimplementing this in C showed that there is indeed overhead, but this is not going to save more time than few minutes.

Next obvious step was to look at uploading coverage files as this is really something what should be fast and not take such enormous time. When writing the original post, I've already tried to merge coverage data using OpenCppCoverage, but that showed to be too slow to make it actually work (the testsuite didn't complete in given 60 minutes).

I was also looking at existing solutions to merge Cobertura XML files, but I've found nothing working reasonably fast. The problem is that all of these always merge two files at one step, making merging thousand files really slow job as you're constantly generating, parsing and processing the resulting xml file for 1000 times. Also these solutions are probably more generic that what I needed.

In the end these files are just simple XML and merging them should not be hard. I was able to quickly write Python script to merge them. It does not support all of the Cobertura attributes, it just merges line based coverage (as this is the only thing which OpenCppCoverage generates), but works pretty fast and reliable.

Overall the build time went down from 60 minutes to 35 minutes and I don't see much space to improve besides improving OpenCppCoverage speed, what is really out of scope for me. Actually older version (0.9.5.3) performs way faster than current one (0.9.6), which is 2-3 times slower.

Gammu 1.38.2

Yesterday Gammu 1.38.2 has been released. This is bugfix release fixing for example USSD or MMS decoding in some situations.

The Windows binaries are available as well. These are built using AppVeyor and will help bring Windows users back to latest versions.

Full list of changes and new features can be found on Gammu 1.38.2 release page.

Would you like to see more features in Gammu? You an support further Gammu development at Bountysource salt or by direct donation.

Life of free software project

During last week I've noticed several interesting posts about challenges being free software maintainer. After being active in open source for 16 years I can share much of the feelings I've read and I can also share my dealings with the things.

First of all let me link some of the other posts on the topic:

I guess everybody involved in in some popular free software project knows it - there is much more work to be done than people behind the project can handle. It really doesn't matter it those are bug reports, support requests, new features or technical debt, it's simply too much of that. If you are the only one behind the project it can feel even more pressing.

There can be several approaches how to deal with that, but you have to choose what you prefer and what is going to work for you and your project. I've used all of the below mentioned approaches on some of the projects, but I don't think there is a silver bullet.

Finding more people

Obviously if you can not cope with the work, let's find more people to do the work. Unfortunately it's not that easy. Sometimes people come by, contribute few patches, but it's not that easy to turn them into regular contributor. You should encourage them to stay and to care about the part of the project they have touched.

You can try to attract completely new contributors through programs as Google Summer of Code (GSoC) or Outreachy, but that has it's own challenges as well.

With phpMyAdmin we're participating regularly in GSoC (we've only missed last year as we were not chosen by Google that year) and it indeed helps to bring new people on the board. Many of them even stay around your project (currently 3 of 5 phpMyAdmin team members are former GSoC students). But I think this approach really works only for bigger organizations.

You can also motivate people by money. It's way which is not really much used on free software projects, partly because lack of funding (I'll get to that later) and partly because it doesn't necessarily bring long time contributors, just cash hunters. I've been using Bountysource for some of my projects (Weblate and Gammu) and so far it mostly works other way around - if somebody posts bounty on the issue, it means it's quite important for him to get that fixed, so I use that as indication for myself. On attracting new developers it never really worked well, even when I've tried to post bounties to some easy to fix issues, where newbies could learn our code base and get paid for that. These issues stayed opened for months and in the end I've fixed them myself because they annoyed me.

Don't care too much

I think this is most important aspect - you simply can never fix all the problems. Let's face it and work according to that. There can be various levels of don't caring. I find it always better to try to encourage people to fix their problem, but you can't expect big success rate in that, so you might find it not worth of the time.

What I currently do:

  • I often ignore direct emails asking for fixing something. The project has public issue tracker on purpose. Once you solve the issue there others will have chance to find it when they face similar problem. Solving things privately in mails will probably make you look at similar problems again and again.
  • I try to batch process things. It is really easier to get focused when you work on one project and do not switch contexts. This means people will have to wait until you get to their request, but it also means that you will be able to deal them much more effectively. This is why Free hosting requests for Hosted Weblate get processed once in a month.
  • I don't care about number of unread mails, notifications or whatever. Or actually I try to not get much of these at all. This is really related to above, I might to some things once in a month (or even less) and that's still okay. Maybe you're just getting notifications for things you really don't need to get notified on? Do you really need notification for new issues? Isn't it better just to look at the issue tracker once in a time than constantly feeling the pressure of not read notifications?
  • I don't have to fix every problem. When it seems like something what could be as well fixed by the reporter, I just try to give them guidance how to dig deeper into the issue. Obviously this can't work for all cases, but getting more people on board always helps.
  • I try to focus on things which can save time in future. Many issues turn out to be just some unclear things and once you figure out that, spend few more minutes to improve your documentation to cover that. It's quite likely that this will save your time in future.

If you still can't handle that, you should consider abandoning the project as well. Does it bring something to you other than frustration of not completed work? I know it can be hard decision, in the end it is your child, but sometimes it's the best think you can do.

Get paid to do the work

Are you doing your fulltime job and then work on free software on nights or weekends? It can probably work for some time, but unless you find some way to make these two match, you will lack free time to relax and spend with friends or family. There are several options to make these work together.

You can find job where doing free software will be natural part of it. This worked for me pretty well at SUSE, but I'm sure there are more companies where it will work. It can happen that the job will not cover all your free software activities, but this still helps.

You can also make your project to become your employer. This can be sometimes challenging to make volunteers and paid contractors to work on one project, but I think this can be handled. Such setup currently works currently quite well for phpMyAdmin (we will announce second contractor soon) and works quite well for me with Weblate as well.

Funding free software projects

Once your project is well funded, you can fix many problems by money. You can pay yourself to do the work, hire additional developers, get better infrastructure or travel to conferences to spread word about it. But the question is how to get to the point of being well funded.

There are several crowdfunding platforms which can help you with that (Liberapay, Bountysource salt, Gratipay or Snowdrift to mention some). You can also administer the funding yourself or using some legal entity such as Software Freedom Conservancy which handles this for phpMyAdmin.

But the most important thing is to persuade people and companies to give back. You know there are lot of companies relying on your project, but how to make them fund the project? I really don't know, I still struggle with this as I don't want to be too pushy in asking for money, but I'd really like to see them to give back.

What kind of works is giving your sponsors logo / link placement on your website. If your website is well ranked, you can expect to get quite a lot of SEO sponsors and the question is where to draw a line what you still find acceptable. Obviously the most willing to pay companies will have nothing to do with what you do and they just want to get the link. The industry you can expect is porn, gambling, binary options and various MFA sites. You will get some legitimate sponsors related to your project as well. We felt we've gone too far with phpMyAdmin last year and we've stricten the rules recently, but the outcome is still not visible on our website (as we've just limited new sponsors, but existing contracts will be honored).

Another option is to monetize your project more directly. You can offer consulting services or provide it as a service (this is what I currently do with Weblate). It really depends on the product if you can build customer base on that or not, but certainly this is not something what would work well for all projects.

Thanks for reading this and I hope it's not too chaotic, as I've moved parts there and back while writing and I'm afraid it got too long in the end.

Gammu 1.38.1

Today Gammu 1.38.1 has been released. This is bugfix release fixing several minor bugs which were discovered in 1.38.0.

The Windows binaries will be available shortly. These are built using AppVeyor and will help bring Windows users back to latest versions.

Full list of changes and new features can be found on Gammu 1.38.1 release page.

Would you like to see more features in Gammu? You an support further Gammu development at Bountysource salt or by direct donation.

Gammu 1.38.0

Today Gammu 1.38.0 has been released. Changes in last two testing releases have been stabilized and this is the outcome. You can expect changes in API or SMSD tables as well as some additional features.

Also this is first stable release after several years which comes with Windows binaries. These are built using AppVeyor and will help bring Windows users back to latest versions.

Full list of changes and new features can be found on Gammu 1.38.0 release page.

Would you like to see more features in Gammu? You an support further Gammu development at Bountysource salt or by direct donation.

Gammu 1.37.90

Yesterday Gammu 1.37.90 has been released. This release brings quite a lot of changes and it's for testing purposes. Hopefully stable 1.38.0 will follow soon as soon as I won't get negative feedback on the changes.

Besides code changes, there is one news for Windows users - there is Windows binary coming with the release. This was possible to automate thanks to AppVeyor, who does provide CI service where you can download built artifacts. Without this, I'd not be able to do make this as I don't have single Windows computer :-).

Full list of changes:

  • Improved support Huawei K3770.
  • API changes in some parameter types.
  • Fixed various Windows compilation issues.
  • Fixed several resource leaks.
  • Create outbox SMS atomically in FILES backend.
  • Removed getlocation command as we no longer fit into their usage policy.
  • Fixed call diverts on TP-LINK MA260.
  • Initial support for Oracle database.
  • Removed unused daemons, pbk and pbk_groups tables from the SMSD schema.
  • SMSD outbox entries now can have priority set in the database.
  • Added SIM IMSI to the SMSD status table.
  • Added CheckNetwork directive.
  • SMSD attempts to power on radio if disabled.
  • Fixed processing of AT unsolicited responses in some cases.
  • Fixed parsing USSD responses from some devices.

Would you like to see more features in Gammu? You an support further Gammu development at Bountysource salt or by direct donation.

Weblate 2.8

Quite on schedule (just one day later), Weblate 2.7 is out today. This release brings Subversion support or improved zen mode.

Full list of changes:

  • Documentation improvements.
  • Translations.
  • Updated bundled javascript libraries.
  • Added list_translators management command.
  • Django 1.8 is no longer supported.
  • Fixed compatibility with Django 1.10.
  • Added Subversion support.
  • Separated XML validity check from XML mismatched tags.
  • Fixed API to honor HIDE_REPO_CREDENTIALS settings.
  • Show source change in zen mode.
  • Alt+PageUp/PageDown/Home/End now works in zen mode as well.
  • Add tooltip showing exact time of changes.
  • Add option to select filters and search from translation page.
  • Added UI for translation removal.
  • Improved behavior when inserting placeables.
  • Fixed auto locking issues in zen mode.

If you are upgrading from older version, please follow our upgrading instructions.

You can find more information about Weblate on https://weblate.org, the code is hosted on Github. If you are curious how it looks, you can try it out on demo server. You can login there with demo account using demo password or register your own user. Weblate is also being used on https://hosted.weblate.org/ as official translating service for phpMyAdmin, OsmAnd, Aptoide, FreedomBox, Weblate itself and many other projects.

Should you be looking for hosting of translations for your project, I'm happy to host them for you or help with setting it up on your infrastructure.

Further development of Weblate would not be possible without people providing donations, thanks to everybody who have helped so far! The roadmap for next release is just being prepared, you can influence this by expressing support for individual issues either by comments or by providing bounty for them.

CI coverage from Windows, Linux and OSX

Once I got CI working on multiple platforms the obvious next step was to be able to aggregate coverage reports across them. This should not be that hard, right? Well I've spent couple of hours on that during last few days.

On Linux and OSX it was pretty much straightforward. Both GCC and Clang do support coverage, so it's just matter of configuring them properly and collect the coverage reports. I've used own solution for that in past and that was really far from working well (somehow I never managed to get coverage fully uploaded to Codecov). Fortunately there exists CMake script called CMake-codecov which does all needed work and works out of the box on GCC and Clang (even on OSX). Well it works on Travis only once you update the compilers and install llvm-cov tool.

The Windows part on AppVeyor was much harder for me. This can be heavily accounted to lack of my experience with Windows and especially development on Windows in past ten years (probably even more). First challenge was to find something what can generate code coverage there.

After lot of googling I've settled down on OpenCppCoverage what seems to be the only free solution I was able to find. The good thing is that it can generate coverage in Cobertura format that Codecov undestands. There are also bad things that I've learned. First of all it's quite hard to integrate this with CTest. There is no support for wrapping test calls in custom commands, so I've misused the memory checks for that purpose. I've written small python script which pretends the valgrind interface and does call OpenCppCoverage in the background.

Now I had around 800 coverage files (one for each test case) and we need to deal with them somehow. The Codeconv command line client doesn't deal wit this out of the box so the obvious choice was to merge them before upload. There even seems to be script doing that, but unfortunately trying that on our coverage data make it nowhere near completion within hour, so that's not really good choice. Second thing I've tried was merging binary coverage in OpenCppCoverage and then exporting to Cobertura format. Obviously Gammu is special project as all I got from this attempt was crashing OpenCppCoverage (it did merge some of the coverages, but it failed in the end without indicating any error).

In the end I've settled down to uploading files in chunks to Codecov. This seems to work quite okay, though is a bit slow, mostly due to way how Codecov bash uploader prepares data to upload (but this will be hopefully fixed soon).

Anyway the goal has been reached, both Windows and Linux code shows in coverage reports.

Continuous integration on multiple platforms

Over the weekend I've played with continuous integration for Gammu to make it run on more platforms. I had to remember many things from the Windows world on the way and the solution is not yet complete, but the basic build is working, the only problematic part are external dependencies.

First of all we already have Linux builds on Travis CI. These cover compilation with both GCC and Clang compilers, hopefully covering most of the possible problems.

Recently I've added OS X builds on Travis CI, what was pretty much painless and worked out of the box.

The next major architecture to support is Windows. Once I've discovered AppVeyor I thought it might be the way to go. The have free plans for open-source projects (though it has only one parallel build compared to four provided by Travis CI).

As our build system is cross platform based on CMake, it should work pretty much out of the box, right? Well almost, tweaking the basics took some time (unfortunately there is no CMake support on AppVeyor, so you have to script it a bit).

The most painful things on the way:

  • finding our correct way to invoke build and testsuite
  • our code was broken on Windows, making the testsuite to fail
  • how to work with power shell (no, I'm not going to like it)
  • how to download and install executable to PATH
  • test output integration with AppVeyor - done using XSLT transformation and uploading test results manually
  • 32-bit / 64-bit mess, CMake happily finds 32-bit libs during the 64-bit build and vice versa, what makes the build fail later when linking - fixed by trying if code can be built with given library
  • 64-bit code crashes in dummy driver, causing testsuite failures (this has to be something Windows specific as the code works fine on 64-bit Linux) - this seems to be caused by too big allocations on stack, moving them to heap will fix this

You can check our current appveyor.yml in case you're going to try something similar. Current build results are on AppVeyor.

As a nice side effect, we now have up to date Windows binaries for Gammu.