psycopg/psycopg2
1
1
Fork 0
mirror of https://github.com/psycopg/psycopg2.git synced 2024-12-03 06:03:44 +03:00
Code Issues Packages Projects Releases Wiki Activity
2,337 Commits 14 Branches 87 Tags 11 MiB
eeef58b581
Commit Graph

582 Commits

Author SHA1 Message Date
Greg Ward
12317557da Always raise OperationalError when connection was closed externally. 
From the DB-API (https://www.python.org/dev/peps/pep-0249/):

  OperationalError

  Exception raised for errors that are related to the database's
  operation and not necessarily under the control of the programmer,
  e.g. an unexpected disconnect occurs, [...]

Additionally, psycopg2 was inconsistent, at least in the async case:
depending on how the "connection closed" error was reported from the
kernel to libpq, it would sometimes raise OperationalError and
sometimes DatabaseError. Now it always raises OperationalError.
 2017-03-14 12:14:00 +00:00
Greg Ward
b203a7c775 Always detect when a connection is closed behind psycopg2's back. 
There's a race condition that only seems to happen over Unix-domain
sockets. Sometimes, the closed socket is reported by the kernel to
libpq like this (captured with strace):

  sendto(3, "Q\0\0\0\34select pg_backend_pid()\0", 29, MSG_NOSIGNAL, NULL, 0) = 29
  recvfrom(3, "E\0\0\0mSFATAL\0C57P01\0Mterminating "..., 16384, 0, NULL, NULL) = 110
  recvfrom(3, 0x12d0330, 16384, 0, 0, 0)  = -1 ECONNRESET (Connection reset by peer)

That is, psycopg2/libpq sees no error when sending the first query
after the connection is closed, but gets an error reading the result.
In that case, everything worked fine.

But sometimes, the error manifests like this:

  sendto(3, "Q\0\0\0\34select pg_backend_pid()\0", 29, MSG_NOSIGNAL, NULL, 0) = -1 EPIPE (Broken pipe)
  recvfrom(3, "E\0\0\0mSFATAL\0C57P01\0Mterminating "..., 16384, 0, NULL, NULL) = 110
  recvfrom(3, "", 16274, 0, NULL, NULL)   = 0
  recvfrom(3, "", 16274, 0, NULL, NULL)   = 0

i.e. libpq received an error when sending the query. This manifests as
a slightly different exception from a slightly different place. More
importantly, in this case connection.closed is left at 0 rather than
being set to 2, and that is the bug I'm fixing here.

Note that we see almost identical behaviour for sync and async
connections, and the fixes are the same. So I added extremely similar
test cases.

Finally, there is still a bug here: for async connections, we
sometimes raise DatabaseError (incorrect) and sometimes raise
OperationalError (correct). Will fix that next.
 2017-03-14 12:08:03 +00:00
Daniele Varrazzo
3c124a0b87 Flake8 complaints 2017-03-14 12:06:46 +00:00
Daniele Varrazzo
a5382d7b72 Skip the tests according to the earliest of libpq built/linked. 2017-03-02 19:32:44 +00:00
Daniele Varrazzo
3a441467c0 Merge branch 'no-unittest2' 
Close #516
 2017-03-02 12:41:33 +00:00
Daniele Varrazzo
5a65ff5cf4 Don't try to use unittest2 
Modules importing unittest directly would miss the monkeypatch: see
ticket #516.
 2017-03-01 20:44:04 +00:00
Daniele Varrazzo
44d8edfd8c Ignore None arguments passed to make_dsn() 
Close #517.
 2017-03-01 20:12:13 +00:00
Daniele Varrazzo
3638d349ba Skip test on PG version with bad interval rounding 
On these ancient versions:

    =# select '999999:00:00.1'::interval;
                interval
    --------------------------------
     41666 days 15:00:00.0999999046
 2017-03-01 02:14:35 +00:00
Daniele Varrazzo
691df4952b Handle overflow in interval parsing 2017-02-24 13:23:56 +00:00
Daniele Varrazzo
834e9996da Parse interval only using integers 
(almost... except for micros rounding)

While this is probably an improvement on the previous implementation,
I am largely waving a dead chicken at windows, which keeps failing to
pass the seconds overflow test. If it doesn't pass now either I'll start
blaming Python's timedelta.
 2017-02-24 03:48:41 +01:00
Daniele Varrazzo
a15dfbbd4f Try to fix the same problem in windows 
Previous commit doesn't pass on Windows: it looks like window's floor()
has an integer overflow.
 2017-02-24 02:41:34 +01:00
Daniele Varrazzo
14fe3ad8c9 Fixed integer overflow in interval typecaster 
Close #512.
 2017-02-24 02:10:27 +01:00
Daniele Varrazzo
d50ed48807 Added readonly and deferrable attributes 2017-02-16 12:46:35 +00:00
Daniele Varrazzo
b5d80b609d Revert pre-2.7b1 behaviour of silent rollback on conn.set_isolation_level() 
Legacy method is legacy.
 2017-02-16 12:46:35 +00:00
Daniele Varrazzo
20c9c17457 connection.isolation_level is now writable 2017-02-16 12:46:35 +00:00
Daniele Varrazzo
3e12522bc9 Don't convert '{}'::unknown into an empty list 
Close #506.
 2017-02-08 08:43:53 +00:00
Daniele Varrazzo
791befca18 Use pydll to poke into extension library 
Causes an error in Py 3.6 in debug mode.

Close #505
 2017-02-07 12:29:06 +00:00
Daniele Varrazzo
28c489f17e Merge branch 'no-set-default-session' 2017-02-07 00:58:54 +00:00
Daniele Varrazzo
7b11b95908 Skipped a couple of tests failing on old dbs 
I don't even know why - worth investigating.
 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
7f593d5eb0 Skip tests with server not supporting NULL in array 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
2e18b42db0 Don't test with server not implementing INSERT with VALUES list 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
3873c6c09f Deal consistently with E'' quotes in tests 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
d23fe42873 Work around a race condition in async cancel test 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
e599da6308 assertDsnEqual moved as TestSuite method 2017-02-06 21:40:05 +00:00
Daniele Varrazzo
9f467231d9 Merge branch 'conform-subclass-adapter' 2017-02-05 12:13:52 +01:00
Daniele Varrazzo
c6af16c3ae Slower timeout for a test that sometimes fail 
At least it should fail with an error after 2 seconds, not waiting for
Travis timeout of 10 minutes.
 2017-02-05 12:06:55 +01:00
Daniele Varrazzo
3fbff5d848 Give precedence to '__conform__()' over superclasses choosing adapter 
Close #456
 2017-02-05 11:54:50 +01:00
Daniele Varrazzo
9054eeccc0 Set default_transaction_* GUC if session state is changed in autocomit 2017-02-04 15:19:41 +00:00
Daniele Varrazzo
665e9dc665 Exposing ISOLATION_LEVEL_DEFAULT to Python 
This is now the state that is returned to Python if nothing has been
explicitly set.
 2017-02-04 14:43:54 +00:00
Daniele Varrazzo
ca59fd8b3f Test looking the transactions characteristics instead of the default 
So we test the effect, not the implementation. Tests pass on master too
this way, three tests fail in this branch, related to autocommit
(sort-of-obviously).
 2017-02-04 10:57:30 +00:00
Daniele Varrazzo
de8b335d80 Merge branch 'master' into sql-compose 2017-02-03 04:56:02 +00:00
Daniele Varrazzo
1911b250e3 Merge branch 'async-keyword' 
Close #495
 2017-02-03 04:45:17 +00:00
Daniele Varrazzo
ce9be69615 Added async_ as an alias for async 
Added in argument for psycopg2.connect() and connection.__init__, and
for the connection.async attribute.
 2017-02-03 04:43:49 +00:00
Daniele Varrazzo
8baf6aa372 Convert warnings into errors on test 2017-02-03 04:41:32 +00:00
Daniele Varrazzo
6e89db020c Merge branch 'fast-executemany' 2017-02-03 04:40:34 +00:00
Daniele Varrazzo
dc1b4fff90 Avoid an useless encode/decode roundtrip in execute_values() 
Tests moved into a separate module.
 2017-02-02 17:29:17 +00:00
Daniele Varrazzo
d8b1fbd905 Further skipping of slow tests 2017-02-02 03:02:29 +00:00
Daniele Varrazzo
626e57acda Merge branch 'fast-executemany' 2017-02-02 02:40:28 +00:00
Daniele Varrazzo
88a21689ce Added note about finding slow tests 2017-02-02 02:09:59 +00:00
Daniele Varrazzo
f24de0357f Allow skipping the slow test 
It's not so much about tests being slow: some just get stuck and timeout
travis.

Skipped all tests taking about more than 0.2s to run on my laptop.
Fast testing takes about 8s instead of 24.
 2017-02-02 01:53:50 +00:00
Daniele Varrazzo
9bac37baf7 Fixed execute_values with unicode 
Also added unicode tests.
 2017-02-01 02:47:59 +00:00
Daniele Varrazzo
2e2dcd536b Fixed fast execute functions with Python 3 2017-02-01 02:36:54 +00:00
Daniele Varrazzo
a95fd3df1a Added execute_batch and execute_values functions 2017-02-01 01:59:47 +00:00
Daniele Varrazzo
a8a3a298f8 Autonumbered args not available in Python 2.6 2017-01-03 17:53:02 +01:00
Daniele Varrazzo
71a168797c Several improvements to the sql objects 
Comparable, iterable, content accessible
 2017-01-03 17:27:01 +01:00
Daniele Varrazzo
a76e665567 Use {} instead of %s placeholders in SQL composition 2017-01-03 15:02:34 +01:00
Daniele Varrazzo
cf40bff2e2 Dropped sql.compose 
Use a SQL % operator instead.
 2017-01-01 08:39:02 +01:00
Daniele Varrazzo
4a55b8018a Adding sql module documentation 2017-01-01 08:12:05 +01:00
Daniele Varrazzo
41b9bfe401 Added test for sql percents treatment 2017-01-01 06:37:09 +01:00
Daniele Varrazzo
828415d476 Typo: composible -> composable 2017-01-01 06:32:18 +01:00
Daniele Varrazzo
ad2643266f Fixed sql.compose with no args and added tests 2017-01-01 06:26:54 +01:00
Daniele Varrazzo
8c020ca47a Fixed sql.Composed representation 2017-01-01 06:26:42 +01:00
Daniele Varrazzo
600416aafc Fixed sql stuff in Py3 2017-01-01 05:59:21 +01:00
Daniele Varrazzo
f11e6d82b0 Added basic sql module implementation 2017-01-01 05:23:15 +01:00
Daniele Varrazzo
fad5100079 Adding some boilerplate for the new module 2017-01-01 03:53:08 +01:00
Daniele Varrazzo
faaef61c27 Merge branch 'master' into named-callproc 2016-12-26 01:49:42 +01:00
Daniele Varrazzo
c22093ddd4 Skip replication tests in green mode 2016-12-25 19:00:30 +01:00
Daniele Varrazzo
b73115ac41 Added test to verify bug #482 2016-12-25 18:30:27 +01:00
Daniele Varrazzo
a478ba9a47 Fixed tests failing on Python 2.6 2016-12-24 01:03:57 +01:00
Daniele Varrazzo
643ba70bad Added ipaddress objects conversion 
Close #387
 2016-10-11 04:54:41 +01:00
Daniele Varrazzo
706ad2f177 Conver network array types into array of strings by default 2016-10-11 02:31:45 +01:00
Daniele Varrazzo
05627ac0f9 Fix unicode mogrify test on python 2 2016-10-11 00:22:23 +01:00
Daniele Varrazzo
91d2158de7 Python source cleanup using flake8 2016-10-11 00:11:55 +01:00
Daniele Varrazzo
78649f8e90 Dropped use of b() "macro" and 2to3 fixer 
Just use the b"" strings syntax supported from python 2.6.
 2016-08-15 01:56:36 +01:00
Daniele Varrazzo
1d950748af Merge branch 'replication-protocol' 2016-08-14 21:09:54 +01:00
Daniele Varrazzo
bd95269c69 Merge branch 'master' into nul-terminator 2016-08-07 02:49:13 +01:00
Jonathan Ross Rogers
cde19c4d59 Make Range pickleable 2016-08-07 02:38:52 +01:00
Daniele Varrazzo
2a4d6027a4 Merge branch 'master' into replication-protocol 
Conflicts:
	tests/testconfig.py
 2016-08-07 01:53:21 +01:00
Alexander Schrijver
03824a1dba Throw an exception when a NUL character is used as a parameter. 2016-07-18 22:56:55 +02:00
Daniele Varrazzo
7566af145b Merge branch 'conn-get-parameters' 2016-07-01 20:12:01 +01:00
Daniele Varrazzo
5bcaf11f9d Allow adapting bytes using QuotedString on Python 3 too 
Close #365.
 2016-07-01 19:14:00 +01:00
Daniele Varrazzo
70af49c0a2 Fixed encoding tests on Py3 2016-07-01 18:50:24 +01:00
Daniele Varrazzo
4a450b63c4 Don't hope to encode stuff in an arbitrary encoding 
libpq's PQescapeString will use the same encoding it has seen before in
a connection (static_client_encoding).

So I think I'll leave this feature here for people who know what is
doing, but won't really document it as a feature: it can't really work
in a generic way (unless adding some disgusting hack like creating a
fake connection with the encoding we want to call PQescapeStringConn
instead of PQescapeString).
 2016-07-01 18:02:20 +01:00
Daniele Varrazzo
2e8e61b8d4 Test moved to the right module, cleanup, but same problem 2016-07-01 18:01:16 +01:00
Daniele Varrazzo
bada1f1f8e Work in progress on writable encoding 
Would help using adapt(unicode) to quote strings without a connection,
see ticket #331.

Currently in heisenbug state: if test_connection_wins_anyway and
test_encoding_default run (in this order), the latter fail because the
returned value is "'\xe8 '", with an extra space. Skipping the first
test, the second succeed.

The bad value is returned by the libpq:

    ql = PQescapeString(to+eq+1, from, len);

just returns len = 2 and an extra space in the string... meh.
 2016-07-01 18:00:05 +01:00
Greg Ward
3ed2c54790 Fix scattered grammar/spelling errors in comments, debug output, etc. 2016-06-30 21:09:15 +01:00
Daniele Varrazzo
eb687103b4 Skip null array test on Postgres versions not supporting it 2016-03-10 12:04:41 +00:00
Oleksandr Shulgin
a7887fab07 Merge remote-tracking branch 'zalando/feature/replication-protocol' into feature/replication-protocol-c-connection-object 2016-03-08 18:56:58 +01:00
Oleksandr Shulgin
da6e061ee8 Use python-defined make_dsn() for ReplicationConnection class 2016-03-08 18:52:21 +01:00
Oleksandr Shulgin
cb7032554e Merge branch 'master' into feature/replication-protocol-c-connection-object 2016-03-04 10:52:10 +01:00
Daniele Varrazzo
c9fd828f8a Allow make_dsn to take no parameter 
The behaviour of connect() is unchanged: either dsn or params must be
specified.
 2016-03-03 17:09:15 +00:00
Daniele Varrazzo
7aab934ae5 Validate output result from make_dsn() 
The output is not necessarily munged anyway: if no keyword is passed,
validate the input but return it untouched.
 2016-03-03 17:09:15 +00:00
Daniele Varrazzo
7155d06cdc Test that the empty dsn is a valid make_dsn input 2016-03-03 17:09:15 +00:00
Daniele Varrazzo
52087a79d9 Added test suite specific for make_dsn 2016-03-03 15:31:37 +00:00
Daniele Varrazzo
2c55a1bd53 Verify that the dsn is not manipulated by make_dsn if not necessary 2016-03-03 15:07:38 +00:00
Daniele Varrazzo
1c4523f0ac Implementation of make_dsn in Python 
This is equivalent to what proposed in #363, but with a much simpler
implementation.
 2016-03-03 04:33:59 +00:00
Oleksandr Shulgin
5d33b39829 Fix error test for invalid START_REPLICATION command. 2016-01-21 15:56:27 +01:00
Oleksandr Shulgin
09a4bb70a1 Allow retrying start_replication after syntax or data error. 2016-01-05 12:31:57 +01:00
Daniele Varrazzo
5fd0f6c4ee Fixed race condition on import in errorcodes.lookup 
Fixes #382.
 2015-12-16 12:03:10 +00:00
Daniele Varrazzo
7302f348bc Added test with objects without length as callproc param 2015-12-12 17:51:08 -05:00
Daniele Varrazzo
d297976d6d Raise TypeError if the dict in callproc param contains non-strings 
Check-and-conversion chain fixed and simplified. 'spname' was a
reference leak.
 2015-12-12 17:51:08 -05:00
mrmilosz
c205f140a0 callproc: tests, docs, and comment/error-reporting touchups. 2015-12-12 17:49:59 -05:00
Oleksandr Shulgin
051e6d1364 Add skip_before_libpq for test_get_dsn_parameters 2015-10-30 13:02:45 +01:00
Oleksandr Shulgin
602fefcae3 Fix typo in a new test name 2015-10-30 11:38:28 +01:00
Oleksandr Shulgin
a4cbb088fe Add connection.get_dsn_parameters() 2015-10-30 11:10:41 +01:00
Oleksandr Shulgin
fbcf99ad07 Move replication connection to C level. 2015-10-27 18:21:24 +01:00
Oleksandr Shulgin
433fb957cb Merge branch 'feature/connect2' into feature/replication-protocol 2015-10-27 17:37:18 +01:00
Oleksandr Shulgin
7aba8b3ed0 Rework psycopg2.connect() interface. 2015-10-27 17:35:57 +01:00
Oleksandr Shulgin
8b79bf43ac Drop ReplicationCursor.flush_feedback(), rectify pq_*_replication_*() interface. 2015-10-23 18:30:18 +02:00
Oleksandr Shulgin
22cbfb26d6 Actually add replication tests to the test suite. 2015-10-20 13:05:43 +02:00