On Monday, 10 September 2018 at 10:50:16 UTC, Joakim wrote:
> Nice work. I wonder about some of your results, as it says that dub itself doesn't build with all of the dmd versions, but somehow the tests pass sometimes (shouldn't be possible if you can't build dub itself). I just tested with `dub fetch dub; dub build dub` with the dub 1.10 that comes alongside dmd 2.081.1 on linux/x64 and it built the latest dub 1.11 without a problem.

Thank you, fixed.

For some reason, libcurl was being linked in only with dub build, and despite the fact that the test harness itself uses libcurl, dub build couldn't find it.

I cleared out all failed builds just to be certain -- about six hours lost total, and I'll trade six hours for bad data any day.

> Regardless, you should talk to Sonke about getting this integrated with code.dlang.org and to Mike about putting up a funding target to pay for both permanent server resources and your time coding this up. I've already said I'd donate towards such a target:

Thank you.

On Monday, 10 September 2018 at 13:58:37 UTC, 0xEAB wrote:
> May I ask why some packages are missing (e.g. `midi-gamepad`)?

midi-gamepad has no releases. It has 0.1.1-alpha, which is a prelease version, and ~master, which is a branch, and I can't rely on ~master being consistent in successive builds.

> A question, how this it currently handle link libraries?

It blindly takes the results of dub build and dub test.

On Monday, 10 September 2018 at 15:46:28 UTC, Neia Neutuladh wrote:
> midi-gamepad has no releases. It has 0.1.1-alpha

So it doesn't test pre-release versions.
Thanks for clarification.

On Monday, 10 September 2018 at 15:46:28 UTC, Neia Neutuladh wrote:
> It blindly takes the results of dub build and dub test.

Another question:
How does it deal with targetType set to "sourceLibrary"?

Moreover, a future feature could be build logs, so one could check why something failed to build.

On Monday, 10 September 2018 at 19:19:56 UTC, 0xEAB wrote:
> On Monday, 10 September 2018 at 15:46:28 UTC, Neia Neutuladh wrote:
>> It blindly takes the results of dub build and dub test.
>
> Another question:
> How does it deal with targetType set to "sourceLibrary"?

As of five minutes ago, for source libraries, the test result is copied to the build result.

> Moreover, a future feature could be build logs, so one could check why something failed to build.

I'm hesitant to do much in this regard because I don't want people to treat this as a CI system. It would certainly be useful for finding bugs in the tester, though.

On Monday, 10 September 2018 at 20:36:44 UTC, Neia Neutuladh wrote:
> On Monday, 10 September 2018 at 19:19:56 UTC, 0xEAB wrote:
>> Moreover, a future feature could be build logs, so one could check why something failed to build.
>
> I'm hesitant to do much in this regard because I don't want people to treat this as a CI system. It would certainly be useful for finding bugs in the tester, though.

A possible solution to this problem could be not testing updated packages for a compiler version they have already been tested with. This comes with some downsides, though - but it prevents abusing your service as CI.


By the way, thanks for all your explanations :)

On Tuesday, 11 September 2018 at 13:48:36 UTC, 0xEAB wrote:
> By the way, thanks for all your explanations :)

No problem! If it's inscrutable, it's not very useful.

On Monday, 10 September 2018 at 01:27:20 UTC, Neia Neutuladh wrote:
> Not on dlang.org anywhere, but I built a crude version of this. Results are available at http://ikeran.org/report/.

A quick status update:

Per-package reports and build badges are now a thing. http://ikeran.org/report/package/accessors.html Dunno if anybody would want a build badge, but it's there. And a bit big. Oh well.

Backfill has gotten all .0 releases from 2.069.0 forward. I'm cutting off the backfill for point releases at 2.080.1, about two days from now. 19 compilers rather than 43 makes this project a bit more sustainable. (It was previously taking six hours to build a single compiler release, but with some changes, it started taking 16.)

After the backfill finishes, I'll start grabbing new dub packages, which will be trivial. The next tasks are auto-detecting new DMD versions (pretty straightforward) and making machine-readable output.

As an aside, the current output is about 50MB of HTML and SVG, served as static files. I thought of using arsd's cgi.d, but there really isn't a point for now.

On Thursday, 20 September 2018 at 02:51:52 UTC, Neia Neutuladh wrote:
> On Monday, 10 September 2018 at 01:27:20 UTC, Neia Neutuladh wrote:
>> Not on dlang.org anywhere, but I built a crude version of this. Results are available at http://ikeran.org/report/.
>
> A quick status update:

And source code is available at https://git.ikeran.org/dhasenan/dubautotester

Please don't judge me.

On Thursday, 20 September 2018 at 04:16:41 UTC, Neia Neutuladh wrote:
> And source code is available at https://git.ikeran.org/dhasenan/dubautotester
>
> Please don't judge me.

Nice work.