- Aur-dev - lists.archlinux.org

[aur-dev] AUR new package notification
by Lukas Fleischer 13 Apr '11

13 Apr '11

I'm not subscribed to new package notifications but it seems like AUR this feature has been disabled for a while now. At least, I did neither find a cronjob running "newpackage-notify" nor the "tupkgs.conf" configuration file that is required to run this script. Browsing the aur-dev archives, I also found a mail [1] from 2008 reporting that the cronjob was obviously disabled. Do we still need this at all? Given that there were no complaints for more than two years and that there's a RSS feed [2] for new packages, I'd say we can drop the notification script as well as the field on the account page. Objections? [1] http://mailman.archlinux.org/pipermail/aur-dev/2008-June/000323.html [2] https://aur.archlinux.org/rss.php

3 3

[aur-dev] [PATCH] Fix performance issues with new PackageDepends lookups
by Dan McGee 12 Apr '11

12 Apr '11

We do a lookup by DepName in the package details view, but I made the silly mistake of forgetting this index addition in the upgrade steps. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- Lukas- said missing index was the cause of the increased CPU usage/load on sigurd, it appears. I already created this index in production as I was the one who forgot it in my last set of patches, whoops! -Dan UPGRADING | 4 ++++ support/schema/aur-schema.sql | 1 + 2 files changed, 5 insertions(+), 0 deletions(-) diff --git a/UPGRADING b/UPGRADING index da85a8d..7878f63 100644 --- a/UPGRADING +++ b/UPGRADING @@ -7,6 +7,10 @@ From 1.8.2 to 1.9.0 1. Translation files are now gettext compatible and need to be compiled after each AUR upgrade by running `make install` in the "po/" directory. +2. Fix up issues with depends performance on large dataset. + +ALTER TABLE PackageDepends ADD INDEX (DepName); + From 1.8.1 to 1.8.2 ------------------- diff --git a/support/schema/aur-schema.sql b/support/schema/aur-schema.sql index d8c8560..da6a1f9 100644 --- a/support/schema/aur-schema.sql +++ b/support/schema/aur-schema.sql @@ -125,6 +125,7 @@ CREATE TABLE PackageDepends ( DepName VARCHAR(64) NOT NULL, DepCondition VARCHAR(20), INDEX (PackageID), + INDEX (DepName), FOREIGN KEY (PackageID) REFERENCES Packages(ID) ON DELETE CASCADE, ) ENGINE = InnoDB; -- 1.7.4.4

3 4

[aur-dev] Testing patches introducing database layout changes
by Lukas Fleischer 12 Apr '11

12 Apr '11

Hi, folks! Just as a recommendation to (regular) contributors: Any patches changing the database layout or the way information is stored in the database should be tested by rebuilding and reloading dummy data into the database in addition to usual testing before submitting them to aur-dev: ---- $ cd support/schema/ $ ./gendummydata.py dummy-data.sql && ./reloadtestdb.sh ---- This way, we can ensure that the patch does neither break the schema file nor the dummy data generation script. Maybe we could also make "./gendummydata.py" a bit more modular and run a slimmed down version to avoid delays here. I came up with this cause I just pushed two patches fixing regressions in "aur-schema.sql" and "gendummydata.py" once again. Please don't get the wrong end of the stick: This is no accusation but a simple proposition to improve our workflow. Affected patches are broken because those changes are hard to spot by just skimming through the source code and because I forgot to do these tests before signing off and pushing.

1 0

[aur-dev] some patches to gendummydata (take 3)
by elij 12 Apr '11

12 Apr '11

Updated with changes Lukas requested. - split logging and mysql removal into separate patches - update logging patch with uppercase variable convention - drop tabs -> spaces patch - clarified commit messages

2 6

[aur-dev] [PATCH 1/2] Use HTTPS for transifex URLs
by Dan McGee 11 Apr '11

11 Apr '11

Signed-off-by: Dan McGee <dan(a)archlinux.org> --- TRANSLATING | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/TRANSLATING b/TRANSLATING index 5764fb5..52c69c9 100644 --- a/TRANSLATING +++ b/TRANSLATING @@ -50,8 +50,8 @@ Updating an Existing Translation 1. Download current translation files from Transifex. You can also do this using transifex-client which is available through the AUR: -$ tx init --host=http://www.transifex.net -$ tx set --auto-remote http://www.transifex.net/projects/p/aur +$ tx init --host=https://www.transifex.net +$ tx set --auto-remote https://www.transifex.net/projects/p/aur/ $ tx pull -r aur.aurpot -a 2. Update the existing translation file using an editor or a tool like poedit: -- 1.7.4.4

1 1

[aur-dev] AUR migrating to gettext and Transifex
by Lukas Fleischer 10 Apr '11

10 Apr '11

Hi, folks! I've just merged my gettext wip branch into master and pushed that to gerolde. From now on, AUR translations will be managed via Transifex [1]. That removes the need for sending git-formatted patches to aur-dev and improves collaboration. It'll hopefully also support cross-project work (pacman has been using Transifex for some time already). All current translation maintainers are encouraged to sign up for an account and request the addition of a new language team. You can find our project page on [2]. Cheers! [1] http://www.transifex.net/ [2] http://www.transifex.net/projects/p/aur/

2 3

[aur-dev] AUR 1.8.2 released
by Lukas Fleischer 10 Apr '11

10 Apr '11

1.8.2 is out [1]. Yet another maintainance release before we dive into 1.9.0. This time, most of the changes are internal ones. There's still some changes visible to the end user: * Translation fixes for German and French. * Option to search for exact name matches only (FS#23556). * More stats in the front page table (thanks, Dan!) * Stricter checks for uploaded source tarballs. * Bumping up default per-page value to 50 (thanks again, Dan!) * Fixes for 14 XSS vulnerabilities in total. Check the Git log [2] for a complete list. The official Arch Linux AUR setup [3] will be upgraded soon! [1] http://projects.archlinux.org/aur.git/commit/?id=66882622 [2] http://projects.archlinux.org/aur.git/log/?id=66882622 [3] https://aur.archlinux.org/

1 0

[aur-dev] Using gettext for translations?
by Lukas Fleischer 07 Apr '11

07 Apr '11

Hi, I thought about this sometime ago. The existing i8n files shouldn't be too hard to convert and we could move translation work to some more appropriate platform like Transifex which is used by pacman already. PHP gettext integration doesn't seem to be too bad and our PHP is compiled with gettext support by default, also. Any opinions/objections?

3 7

[aur-dev] Patch for French translation
by Cédric Girard 07 Apr '11

07 Apr '11

Hi, Here is a patch for the French translation. What have been done : * Translation of "notify" has been changed to the literal translation "notifier" which seems more appropriate (cf. recent thread on AUR-general) * Uniformization of the translation of "flag" * Addition of "TU" besides the translation of Trusted User as user may not understand that Trusted User and "Utilisateur de confiance" are the same thing * Some capitalization correction (in French you don't put capital letters in the middle of a sentence as you do eg for title in English). * "unsupported" translation has been replaced by "unsupported" (with quotes) as it makes no sense to translate it * "Safe" translation has been replaced by "sûr" has it seems more appropriate than "sain" (sane) I think that sum it up. I won't try to attach the patch as it seems not working very well ... Here is a pastebin link instead: http://pastebin.com/Gp8TBz1C Regards, -- Cédric Girard

2 2

[aur-dev] some patches to gendummydata (take 2)
by elij 06 Apr '11

06 Apr '11

Updated patchset to take comments and criticisms into account. - tabs to spaces is now its own patch, with appropriate commit message - removal of DBUG variable - removal of straggling 're' module import - removal of 'working...' log messages (collapsed later patch into earlier) - broke 'wrap sql' patch into its own ---- Some patches for gendummydata script. - remove need for sql connection. this allows someone to run the script on a dev box with no sql connection (for example) and then ship the output sql wherever needed. - remove need to have category names. only the actual numbers are needed, and if you are using dummy data, you are likely using the base schema. even if that is not the case, as long as the base number of categories _at least_ are present, the dummy data is still 'fine' (eg. if more categories are added, then no dummy packages will just use those categories until the counter in the script is incremented) - use logging module - remove 'progress' logging output. the script doesn't run slow enough to warrant the extra noise - use spaces in the python script. spaces in python are _a damn good idea_

2 10