fa35eadf47
[entropy.fetchers] do not generate two separate requests per download
...
It has been discovered that each download() call (urllib handler)
caused the code to generate two separete HTTP requests for the same
URL. This commit hopefully fixes it without causing any regression
2012-07-23 19:20:29 +02:00
13b5bcd018
[Rigo] serialize MTed download to the same URLs (avoid flooding the WS)
2012-07-23 19:19:49 +02:00
ef64634def
[entropy.dump] dumpobj: use temporary files, fixes possible races when MT
2012-07-23 15:47:19 +02:00
f301ae60c7
[entropy.graph] Graph: drop __del__ (just use destroy() instead) to help automatic GC
2012-07-23 13:10:49 +02:00
61950f3af0
[entropy.client.interfaces.db] sort added list to improve determinism (good for caches)
2012-07-23 13:08:25 +02:00
e8c6af8331
[entropy.misc] LogFile: remove __del__
2012-07-23 13:08:00 +02:00
defb58574d
[entropy.cache] EntropyCacher: remove __del__
2012-07-23 13:07:51 +02:00
0dbc40777e
[todo] update TODO
2012-07-23 00:30:07 +02:00
db5265e3d5
[entropy.client.package] use retrieveContentIter()
2012-07-23 00:29:27 +02:00
a1c5e2bc12
[entropy.db] introduce retrieveContentIter()
2012-07-23 00:29:11 +02:00
ce4d28ef47
[entropy.client] if xcache=False, force EntropyCacher "stashing cache" to OFF
2012-07-22 19:59:37 +02:00
9472978e60
[entropy.cache] EntropyCacher: make possible to disable the "stashing cache" at runtime
2012-07-22 19:44:18 +02:00
de52fdb453
[equo] fix undefined reference to installed_repo when --resume is used
2012-07-22 12:50:46 +02:00
574801e2f9
Tagging Entropy version 1.0_rc126
2012-07-20 12:59:21 +02:00
ae94f49e79
Release Entropy 1.0_rc126
2012-07-20 12:59:20 +02:00
dc27444749
[entropy.server] use Server.repositories()
2012-07-19 21:10:26 +02:00
499aee4feb
[entropy.server] automatically cleanup the list of trashed SPM UIDs
...
This commit tries to safely reduce the amount of entries inside the
Trashed SPM UIDs table. At least, the entries are now kept under
control.
2012-07-19 21:09:10 +02:00
2f9fafc128
[entropy.db] EntropyRepository: create index for trashedcounters
2012-07-19 21:08:38 +02:00
3dc413e80f
[entropy.server] fix typo in comment
2012-07-19 20:59:48 +02:00
b9b5157817
[entropy.server] resolve_spm_package_uid() already returns int()
2012-07-19 20:49:31 +02:00
c2c0787169
[entropy.spm] PortagePlugin: resolve_spm_package_uid() must return int
2012-07-19 20:48:53 +02:00
8b5606ea06
[entropy.server] use resolve_spm_package_uid()
2012-07-19 20:27:47 +02:00
00a3d87b7d
[entropy.spm] PortagePlugin: add resolve_spm_package_uid()
2012-07-19 20:27:15 +02:00
5a0d8cf7f5
[entropy.db] add removeTrashedUids() and listAllTrashedSpmUids() methods
2012-07-19 20:10:52 +02:00
4d89935ef6
[Rigo] give Icons and Rating worker threads to collect more keys during a batch cycle
2012-07-19 13:27:43 +02:00
749e3ff29d
[entropy.client] do not remove current repository database before download
...
As reported in bug #3495 , do not remove the old repository database
file before validating the download outcome (UrlFetcher download
result)
2012-07-17 11:12:40 +02:00
60d4de50e7
[entropy.client] change error string
...
"repository does not exist online" is for some people confusing,
better saying "cannot download the repository".
2012-07-17 10:53:56 +02:00
ece0b58858
[entropy.client] remove orphan function
2012-07-17 10:51:38 +02:00
4e3b5e6106
Tagging Entropy version 1.0_rc125
2012-07-16 10:40:37 +02:00
60fce6a94b
Release Entropy 1.0_rc125
2012-07-16 10:40:35 +02:00
5189bf51bc
[Rigo] update translations
2012-07-16 10:40:15 +02:00
b504f4ce5f
[Rigo] update rigo.pot
2012-07-16 10:38:31 +02:00
0c4d3a2b0a
[Rigo] when App is updatable, show "Update" instead of "Remove"
...
This commit fixes bug #3417
2012-07-15 11:37:08 +02:00
09b8110790
[Rigo] Add "Black on White" and "White on Black" terminal colors support
...
This commit fixes bug #3433
2012-07-15 11:17:35 +02:00
f573957172
[Rigo] avoid race with finally block and GC wrt file descriptors
...
It has been observed that subprocess.call() called by open_url()
might fail due to Error 9 (Bad File Descriptor) due to a race
between the garbage collector thread (run in the MainThread) and
our own subprocess.call() thread. For this reason, make sure to
call open_url() outside the try/finally block. This seems to be
enough.
2012-07-14 18:22:48 +02:00
afd2608153
[Rigo] rewrite NoticeBoard TreeView behaviour
...
Previously, the Notice object was shown directly inside the
cell renderer, which made things a bit clumsy and buggy.
Now, notices are loaded through xdg-open in html format.
2012-07-14 18:19:59 +02:00
3912ca0471
[RigoDaemon] _noticeboards_available_signal: convert dbus.String() to Python unicode
2012-07-14 18:19:42 +02:00
7135ba3a01
[RigoDaemon] make sure to keep the Installed Packages Repository closed
...
This helps in case of uncommitted transactions that could cause
any sqlite3 db consumer to lock until timeout with "Database is locked".
2012-07-14 16:10:18 +02:00
0ce6633733
[entropy.client] add public method: close_installed_repository()
2012-07-14 16:05:50 +02:00
941bfaf1e8
[services] optimize execution time of _expand_cpvs()
2012-07-09 19:32:34 +02:00
9227b17744
[services] portage-repository-converter: add support to Portage md5 eclass cache
2012-07-09 19:32:10 +02:00
9b673fad9b
[entropy.output] fix is_stdout_a_tty when sys.stdout has no fileno()
2012-07-07 18:15:34 +02:00
bdcc0821fc
Tagging Entropy version 1.0_rc124
2012-07-07 14:11:59 +02:00
89713f2528
Release Entropy 1.0_rc124
2012-07-07 14:11:58 +02:00
0e54ec96ec
Revert "[entropy.server] keep repo closed during early init stages"
...
This reverts commit 4c8d6c74e3
2012-07-07 14:05:57 +02:00
a312a6146e
[entropy.server] reimplement connection pooling
...
This commit should hopefully fix any deadlock reported in
Entropy Server due to weird sqlite3 behaviours when multiple
idle connections are open on the same file
2012-07-07 13:58:25 +02:00
b8976f4c0c
[entropy.db] EntropyRepository: expose connection and cursor pools
...
This commit makes possible for subclasses to reimplement the
connection and cursor pool logic.
2012-07-07 13:29:29 +02:00
017f2b162c
[entropy.db] EntropyRepository: hold __connlock during pool setup
...
This commit ensures that __connlock is held during the whole
__cursor_cache and __connection_cache initialization. This avoids
any possible race condition, especially on dbapi2.connect() which
could cause weird random anomalies
2012-07-07 09:34:21 +02:00
b9c7a2dd92
Tagging Entropy version 1.0_rc123
2012-07-06 21:43:32 +02:00
9d2b329836
Release Entropy 1.0_rc123
2012-07-06 21:43:31 +02:00
Fabio Erculiani