haskell/serversession
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
93 Commits 2 Branches 10 Tags 427 KiB
1aeacdebea
Commit Graph

93 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Felipe Lessa
1aeacdebea Fix acid-state test on GHC < 7.10. 2015-06-01 03:41:59 -03:00
Felipe Lessa
d291e72ba3 Add note about tests on README.md. 2015-06-01 03:32:33 -03:00
Felipe Lessa
5a7b4ce08f
Test GHC 7.10 via base package. 2015-06-01 03:15:31 -03:00
Felipe Lessa
9b1be63b8a Don't test AcidStorage on local storage if on GHC < 7.10. 
I give up.  I hope the tests finally go green on Travis.
 2015-06-01 02:49:00 -03:00
Felipe Lessa
e4f9f422d8 Travis: Don't bother with GHC HEAD. 2015-06-01 02:08:36 -03:00
Felipe Lessa
d23c5ac4c6 Use a HashMap on acid-state implementation + Travis. 
Also dial back on maximum heap size as 2 GiB with -F1.5 still makes OOM
kill the process on Travis.
 2015-06-01 02:06:36 -03:00
Felipe Lessa
9a471a90ea Travis: Tweak acid-state test suite RTS options again. 
C'mon, GHC 7.8... Why do you need so much more memory than GHC 7.10?
 2015-06-01 00:55:43 -03:00
Felipe Lessa
1b3bec4778 Travis: Allow a slightly bigger heap for acid-state test suite. 
Although it uses 580M on my GHC 7.10, it blew up the 1 GiB limit on Travis.

Travis currently allows usage of 3 GiB, and GHC tends to use 2x the the
maximum residency (even with the compacting option, WTF?).
 2015-05-31 23:50:53 -03:00
Felipe Lessa
774bc18fa1 Force the test suites to use a lot less memory. 
Works fine on my machine using GHC 7.10 with these limits.  Note that
I've activated the in-place compaction GC.
 2015-05-31 23:22:21 -03:00
Felipe Lessa
0cca9cd086 Use HashMap for SessionMap. 
It's a bit faster and uses a bit less memory.
 2015-05-31 23:09:02 -03:00
Felipe Lessa
e127371df6 Travis: Explicitly run each test suite. 
Travis is timing out and I can't see where in the test suite we're
having problems.
 2015-05-31 21:35:42 -03:00
Felipe Lessa
4ff73d87d8 Travis: Use PostgreSQL 9.3. 
The persistent test is failing and I have no idea why, as it's not
displaying any error message.  This commit changes a few things with
PostgreSQL just to see if they make any difference.

Note that the persistent test suite passes with flying colors both
before and after this commit on my machine using PostgreSQL 9.4.2.
 2015-05-31 16:30:44 -03:00
Felipe Lessa
fcdb14e2bf More GHC 7.8 fixes: Functor constraint. 2015-05-31 15:42:10 -03:00
Felipe Lessa
e9c4ea69a5 Travis: run tests sequentially. 
They're too intensive and already make good use of the availabe CPU and RAM.
 2015-05-31 15:14:46 -03:00
Felipe Lessa
1770734cdb More consistent badge style. 2015-05-31 15:05:17 -03:00
Felipe Lessa
8f719a3cbb Add Travis and Hackage badges to main README.md. 2015-05-31 15:02:44 -03:00
Felipe Lessa
4e3f27cb36 Setup.lhs x 7. 2015-05-31 14:56:39 -03:00
Felipe Lessa
d967f2b4a7 More GHC 7.8 fixes: DeriveDataTypeable, imports, (Functor m). 2015-05-31 14:53:26 -03:00
Felipe Lessa
a731cb2a8a Travis: GHC 7.10.1 instead of GHC 7.10.2. 2015-05-31 14:38:33 -03:00
Felipe Lessa
16c4248355 More GHC 7.8 fixes (import mempty, derive Typeable). 2015-05-31 14:36:33 -03:00
Felipe Lessa
8a222d4a42 Travis: First try at having Redis and PostgreSQL ready. 2015-05-31 13:53:24 -03:00
Felipe Lessa
fdbce5f767 More missing Applicative imports for GHC 7.8. 2015-05-31 13:46:49 -03:00
Felipe Lessa
f0e1005f18 Add Applicative imports for GHC 7.8. 2015-05-31 12:57:52 -03:00
Felipe Lessa
6190544cc4 Travis: Don't bother with GHC 7.6. 2015-05-31 12:39:34 -03:00
Felipe Lessa
70e4353b3e Travis: --run-tests needs Cabal 1.20. 2015-05-31 12:03:53 -03:00
Felipe Lessa
4bab3dfa2c Travis: I thought I didn't need alex and happy. 2015-05-31 11:57:53 -03:00
Felipe Lessa
c7d8187149 First stab at Travis + cabal-meta. 
Travis never works on the first try for me, though :(.
 2015-05-31 11:49:46 -03:00
Felipe Lessa
a1769bce1c Add README note about custom session data types. 2015-05-31 11:29:31 -03:00
Felipe Lessa
3e33c58af0
Generalize session data (huge commit). 2015-05-31 11:07:23 -03:00
Felipe Lessa
821016a382 Remove outdated comment. 2015-05-30 11:03:52 -03:00
Felipe Lessa
34f55d571f Add workaround to fpco/stack#120. 2015-05-29 17:30:20 -03:00
Felipe Lessa
dc34513b3a First stab at a stack.yaml file. 2015-05-29 15:41:08 -03:00
Felipe Lessa
6324f2ddd5 Add single huge unit test to saveSession. 2015-05-29 00:56:09 -03:00
Felipe Lessa
823f992646 Tests for session invalidation. 2015-05-29 00:40:55 -03:00
Felipe Lessa
43b431a6c3 New support for timeout resolution optimization. 2015-05-28 17:07:57 -03:00
Felipe Lessa
b12b3e7cd1 Do not save empty sessions if possible. 2015-05-28 16:39:13 -03:00
Felipe Lessa
3757b4dad9 Test for nextExpires. 2015-05-28 16:04:30 -03:00
Felipe Lessa
ef0d3a707b Add note about limits to README. 2015-05-28 14:40:22 -03:00
Felipe Lessa
6895fbee4f Allow Redis to store sessions with more than 512k keys. 2015-05-28 14:34:47 -03:00
Felipe Lessa
faae95312a Use cereal for 2x improvement in time and space for persistent. 
Now the peak memory usage of the persistent test suite sits at
2 GiB (752 MB max residency).  It also uses a lot less CPU time.
 2015-05-28 13:34:45 -03:00
Felipe Lessa
11197f6180 Add note about SQL index. 2015-05-28 13:16:45 -03:00
Felipe Lessa
a35a10db0d Fewer stress tests but to the same effect. 2015-05-28 13:16:25 -03:00
Felipe Lessa
ec4623da34 Do not run large data tests in parallel. 
Brings peak memory usage of the persistent test suite from 6 GiB to
4 GiB.  Still very bad, though.
 2015-05-28 13:02:54 -03:00
Felipe Lessa
11bb9cafa4 Add storage tests for size and representation limits. 2015-05-28 12:08:00 -03:00
Felipe Lessa
1a28fc50dd Write more core tests, log mock operations. 2015-05-28 03:26:15 -03:00
Felipe Lessa
803aed691a Add note about Redis expires to main README. 2015-05-28 02:57:31 -03:00
Felipe Lessa
c9e080d5b8 Better time format for Redis serialization. 2015-05-28 02:44:46 -03:00
Felipe Lessa
ff6c0efaab Redis backend is all green. 2015-05-28 02:42:22 -03:00
Felipe Lessa
92fbee20da Fix copy-paste. 2015-05-28 02:23:00 -03:00
Felipe Lessa
bdd6db9e2a Do not lose precision when serializing UTCTime to Redis. 2015-05-28 02:20:17 -03:00