This checks the local CPU settings to see if it has the right virtualizing
CPU instructions, and only uses VirtualBox hardware virtualization on
setups where it is available. VMs instances usually do not provide these
instructions. Where the CPU features are not available, it will use VB's
software virtualization.
Also, when cloning, no need to specify the branch and only download that
one. We already have only a single branch. And forcing master isn't
necessary.
This greatly speeds up checkupdates when UCM:Tags is used on git
repositories, especially on ones with lots of tags. The old method ran a
command per tag. The new method runs a single command and doesn't
require any sorting.
Test runs show that `fdroid checkupdates org.adaway` is down from ~13s
to ~10s on my laptop.
The point of removing these was to make sure that gradlew wasn't used,
and that our gradle was used instead. Removing the scripts already
accomplishes this.
Removing gradle/ should be harmless, but some apps like I2P re-use this
directory to also hold other stuff that is actually needed. So be safer
and don't remove it at all.
Fix pubkey extraction on update
Replacement of !86.
Fix pubkey extraction in case of non-empty _JAVA_OPTIONS. Fixes#133.
I didn't actually run the test suite (it looks like there are some preparations to be done for that), but I checked the commands from `test_fdroid_popen_stderr_redirect` in ipython.
See merge request !103
We weren't using ^ and $, so stuff like java-8-openjdk-i386 would match
both the Arch and the Debian regexes. A list with one regex per line in
the same format is also easier to read and maintain.
This might be why java8 isn't being detected properly on our Debian
stable buildserver.
This avoids the following incorrect error:
fdroidserver.common.FDroidException: Only one local metadata file allowed! Found: .fdroid.keystorepass.txt .fdroid.keypass.txt
Allow commonsware and gradle plugin repos
This is in a MR because I'm uneasy about adding a semi-random AWS url to the list. But seems like the commonsware guy only publishes jars there, which doesn't make any sense to me.
Any idea @eighthave?
See merge request !102
Fix update crash in case of unset dates in APK; add an option to use APK's entry date
I encountered an error when I tried to add [TrackID APK](https://play.google.com/store/apps/details?id=com.sonyericsson.trackid) (files in the APK don't have dates set): as far as I can remember `datetime(*manifest.date_time)` was raising `ValueError`.
Also add an option to use date from APK as this APK's entry date (may be useful in case of a simple binary repo): `fdroid update --use-date-from-apk`.
I should have split this commit, but I hope this would be OK too.
See merge request !104
Debian import freeze
These are the issues as discussed in today's meeting. More in the commit messages, and mostly, in the related issues.
See merge request !101
serverwebroot has long supported uploading to multiple servers, this bit of
metadata communicates those official mirrors to the client so that it can
automatically do something useful with that information.
closes#14https://gitlab.com/fdroid/fdroidserver/issues/14
Full BSD/OSX port
This finishes up the OSX port by getting the full test suite running on OSX. It also sets up an OSX CI job travis-ci.org, run as part of the github mirror here: https://github.com/f-droid/fdroidserver You can see those CI builds here: https://travis-ci.org/f-droid/fdroidserver
This should translate entirely to *BSD since OSX's user land is merged with FreeBSD.
See merge request !100
