]> git.ipfire.org Git - thirdparty/psycopg.git/log
thirdparty/psycopg.git
2 years agofix: sync pipeline after implicit BEGIN 401/head
Denis Laxalde [Tue, 4 Oct 2022 15:53:26 +0000 (17:53 +0200)] 
fix: sync pipeline after implicit BEGIN

Prior to this change, the following code:

    with conn.pipeline():
       conn.execute("select 'x'").fetchone()
       conn.execute("select 'y'")

would produce the following libpq trace:

    F 13 Parse  "" "BEGIN" 0
    F 14 Bind  "" "" 0 0 1 0
    F 6 Describe  P ""
    F 9 Execute  "" 0
    F 18 Parse  "" "select 'x'" 0
    F 14 Bind  "" "" 0 0 1 0
    F 6 Describe  P ""
    F 9 Execute  "" 0
    F 4 Flush
    B 4 ParseComplete
    B 4 BindComplete
    B 4 NoData
    B 10 CommandComplete  "BEGIN"
    B 4 ParseComplete
    B 4 BindComplete
    B 33 RowDescription  1 "?column?" NNNN 0 NNNN 65535 -1 0
    B 11 DataRow  1 1 'x'
    B 13 CommandComplete  "SELECT 1"
    F 13 Parse  "" "BEGIN" 0
    F 14 Bind  "" "" 0 0 1 0
    F 6 Describe  P ""
    F 9 Execute  "" 0
    F 18 Parse  "" "select 'y'" 0
    F 14 Bind  "" "" 0 0 1 0
    F 6 Describe  P ""
    F 9 Execute  "" 0
    B 4 ParseComplete
    B 4 BindComplete
    B 4 NoData
    B NN NoticeResponse  S "WARNING" V "WARNING" C "25001" M "there is already a transaction in progress" F "SSSS" L "SSSS" R "SSSS" \x00
    F 4 Sync
    B 10 CommandComplete  "BEGIN"
    B 4 ParseComplete
    B 4 BindComplete
    B 33 RowDescription  1 "?column?" NNNN 0 NNNN 65535 -1 0
    B 11 DataRow  1 1 'y'
    B 13 CommandComplete  "SELECT 1"
    B 5 ReadyForQuery  T

where we can see that the BEGIN statement (due to the connection being
in non-autocommit mode) is emitted twice, as notified by the server.

This is because the transaction state after the implicit BEGIN is not
"correct" (i.e. should be INTRANS, but is IDLE) since the result from
respective statement has not been received yet.

By syncing after the BEGIN, we fetch result from this command thus get
the transaction state INTRANS for following queries. This is similar to
what happens with explicit transaction, by using nested pipelines.

2 years agochore: bump to version number 3.1.3
Daniele Varrazzo [Tue, 4 Oct 2022 16:43:27 +0000 (17:43 +0100)] 
chore: bump to version number 3.1.3

2 years agoMerge branch 'fix-399'
Daniele Varrazzo [Tue, 4 Oct 2022 16:23:17 +0000 (17:23 +0100)] 
Merge branch 'fix-399'

2 years agofix: fix handling of queries with %% in ClientCursor
Daniele Varrazzo [Mon, 3 Oct 2022 23:28:07 +0000 (00:28 +0100)] 
fix: fix handling of queries with %% in ClientCursor

Close #399.

2 years ago#397: Fix typo in install.rst
Darren Ho [Sat, 1 Oct 2022 17:33:59 +0000 (18:33 +0100)] 
#397: Fix typo in install.rst

2 years agoRemove warning about missing binary package for M1
Magnus Watn [Fri, 30 Sep 2022 13:44:28 +0000 (15:44 +0200)] 
Remove warning about missing binary package for M1

Binary packages for Apple M1 was added in #162.

2 years agoMerge branch 'stream-close'
Daniele Varrazzo [Wed, 28 Sep 2022 18:25:08 +0000 (19:25 +0100)] 
Merge branch 'stream-close'

2 years agofix: make sure to terminate query on gen.close() from Cursor.stream()
Daniele Varrazzo [Wed, 21 Sep 2022 10:40:25 +0000 (11:40 +0100)] 
fix: make sure to terminate query on gen.close() from Cursor.stream()

Fix #382

2 years agoperf: use functools.cache where available
Daniele Varrazzo [Wed, 28 Sep 2022 01:01:52 +0000 (02:01 +0100)] 
perf: use functools.cache where available

Faster than lru_cache; use it when we don't suppose we will ever evict
values.

2 years agoMerge branch 'fix-escape-nametuple-field'
Daniele Varrazzo [Wed, 28 Sep 2022 00:46:48 +0000 (01:46 +0100)] 
Merge branch 'fix-escape-nametuple-field'

2 years agodocs: mention named tuples regression fix in change log 388/head
Daniele Varrazzo [Tue, 27 Sep 2022 08:54:49 +0000 (09:54 +0100)] 
docs: mention named tuples regression fix in change log

2 years agotest: more exhaustive test of difficult names/attrs in composites
Daniele Varrazzo [Tue, 27 Sep 2022 08:51:11 +0000 (09:51 +0100)] 
test: more exhaustive test of difficult names/attrs in composites

2 years agofix: restore pre-3.1 behaviour converting fields to named tuples attributes
Jolbas [Sun, 25 Sep 2022 14:11:36 +0000 (16:11 +0200)] 
fix: restore pre-3.1 behaviour converting fields to named tuples attributes

Let more valid identifiers through _as_python_identifier(). Also 2.5x faster if no change is needed

Close #386

2 years agotest: check unusual column names and namedtuple
Jolbas [Sun, 25 Sep 2022 12:32:42 +0000 (14:32 +0200)] 
test: check unusual column names and namedtuple

2 years agofix: avoid using type variables in typevar bounds
Daniele Varrazzo [Tue, 27 Sep 2022 08:15:21 +0000 (09:15 +0100)] 
fix: avoid using type variables in typevar bounds

It is not allowed by the specs. Mypy tolerates it, replacing the
argument with Any, according to @erictraut. Pyright reports it as an
error instead.

See #365, microsoft/pyright#3984

2 years agochore: upgrade mypy to 0.981
Daniele Varrazzo [Wed, 28 Sep 2022 00:28:41 +0000 (01:28 +0100)] 
chore: upgrade mypy to 0.981

Most changes needed by https://github.com/python/typeshed/pull/8446.

2 years agodocs: fix plural
Daniele Varrazzo [Sun, 25 Sep 2022 01:41:52 +0000 (02:41 +0100)] 
docs: fix plural

2 years agodocs: mention macOS arm64 packages distribution
Daniele Varrazzo [Sun, 25 Sep 2022 01:39:29 +0000 (02:39 +0100)] 
docs: mention macOS arm64 packages distribution

2 years agobuild: fix starting Postgres in macOS build script
Daniele Varrazzo [Sun, 25 Sep 2022 01:31:22 +0000 (02:31 +0100)] 
build: fix starting Postgres in macOS build script

The brew command fails with:

    Could not enable service: 125: Domain does not support specified action
    Error: Failure while executing; `/bin/launchctl enable gui/501/homebrew.mxcl.postgresql@14` exited with 125.

2 years agodocs: typo messge -> message
Daniele Varrazzo [Wed, 21 Sep 2022 11:49:25 +0000 (12:49 +0100)] 
docs: typo messge -> message

Thank you, codespell 💀💀💀

2 years agodocs: add link to document how to recover the stream from ACTIVE state
Daniele Varrazzo [Wed, 21 Sep 2022 11:46:07 +0000 (12:46 +0100)] 
docs: add link to document how to recover the stream from ACTIVE state

See #382.

2 years agochore: bump to next dev version number
Daniele Varrazzo [Mon, 19 Sep 2022 17:31:58 +0000 (18:31 +0100)] 
chore: bump to next dev version number

2 years agochore: bump to version number 3.1.2 3.1.2
Daniele Varrazzo [Mon, 19 Sep 2022 11:08:22 +0000 (12:08 +0100)] 
chore: bump to version number 3.1.2

2 years agofix: fix queries with repeated named parameters in ClientCursor
Daniele Varrazzo [Sun, 18 Sep 2022 21:45:48 +0000 (22:45 +0100)] 
fix: fix queries with repeated named parameters in ClientCursor

Close #378.

2 years agotest: allow to trace libpq to STDERR, rename option to --pq-trace
Denis Laxalde [Thu, 15 Sep 2022 10:24:43 +0000 (12:24 +0200)] 
test: allow to trace libpq to STDERR, rename option to --pq-trace

By using STDERR, we get the trace directly in pytest output for failing
tests. Though this requires care when used in combination with
--capture=sys because this would make sys.stderr an in-memory file.

When STDERR is used, we don't insert the header since the test case
identifier will be interleaved already.

2 years agofix: also sync nested pipeline when a transaction is active
Denis Laxalde [Thu, 15 Sep 2022 16:28:21 +0000 (18:28 +0200)] 
fix: also sync nested pipeline when a transaction is active

Upon enter, transaction() checks the in-transaction status to determine
if a BEGIN (when IDLE) or a SAVEPOINT (otherwise) should be inserted.
However, in pipeline mode, if the pipeline is in "implicit transaction"
(i.e. it has no active BEGIN) and statements have been sent, the
in-transaction status might be ACTIVE if those statements have not yet
completed (typically, when results have not been fetched).

By issuing a sync() before entering a nested pipeline (which will happen
when entering transaction(), if already within a pipeline), we force
completion of those statements, and thus get a predictable
in-transaction status before entering the transaction() block.

Closes #374.

2 years agotest: fail test session in case of segfault
Daniele Varrazzo [Thu, 15 Sep 2022 01:07:57 +0000 (02:07 +0100)] 
test: fail test session in case of segfault

Before that, tox reruns would have made tests pass, because failed tests
wouldn't have gotten a chance to be written in the cache.

2 years agoMerge branch 'fix-372'
Daniele Varrazzo [Wed, 14 Sep 2022 21:30:44 +0000 (22:30 +0100)] 
Merge branch 'fix-372'

2 years agofix(psycopg_c): propagate exceptions in Transformer.set_pgresult()
Daniele Varrazzo [Wed, 14 Sep 2022 21:02:38 +0000 (22:02 +0100)] 
fix(psycopg_c): propagate exceptions in Transformer.set_pgresult()

Fix #372

2 years agotest: add test to verify the behavior of adapters failing init
Daniele Varrazzo [Wed, 14 Sep 2022 20:57:05 +0000 (21:57 +0100)] 
test: add test to verify the behavior of adapters failing init

An exception in loader init doesn't get raised and causes a segfault.
This reproduces the issue in #372.

2 years agoMerge branch 'docs/pgconn'
Daniele Varrazzo [Wed, 14 Sep 2022 14:17:55 +0000 (15:17 +0100)] 
Merge branch 'docs/pgconn'

2 years agodocs: document PGconn's tracing methods 373/head
Denis Laxalde [Wed, 14 Sep 2022 12:20:05 +0000 (14:20 +0200)] 
docs: document PGconn's tracing methods

2 years agodocs: document PGconn.encrypt_password()
Denis Laxalde [Wed, 14 Sep 2022 12:02:43 +0000 (14:02 +0200)] 
docs: document PGconn.encrypt_password()

Including an example of usage.

2 years agorefactor(psycopg_c): define loggers at module level
Daniele Varrazzo [Wed, 14 Sep 2022 14:14:32 +0000 (15:14 +0100)] 
refactor(psycopg_c): define loggers at module level

2 years agofix(psycopg_c): add missing logger declaration
Daniele Varrazzo [Wed, 14 Sep 2022 13:40:41 +0000 (14:40 +0100)] 
fix(psycopg_c): add missing logger declaration

2 years agoMerge branch 'fix-371'
Daniele Varrazzo [Wed, 14 Sep 2022 12:24:55 +0000 (13:24 +0100)] 
Merge branch 'fix-371'

2 years agodocs: mention Windows timezone problem fixed in change log
Daniele Varrazzo [Wed, 14 Sep 2022 11:59:14 +0000 (12:59 +0100)] 
docs: mention Windows timezone problem fixed in change log

2 years agofix: catch any type of error if ZoneInfo creation fails
Daniele Varrazzo [Wed, 14 Sep 2022 11:50:05 +0000 (12:50 +0100)] 
fix: catch any type of error if ZoneInfo creation fails

OSError would be enough to solve the problem reported in #371. However,
in case of error, Windows C implementation segfaults. I would prefer to
not chase this error just because of weirdness in the tzdata external
library, so let's cast a wider net.

Fix #371

2 years agotest: add bad timezone to simulate a reported timezone issue
Daniele Varrazzo [Wed, 14 Sep 2022 11:33:23 +0000 (12:33 +0100)] 
test: add bad timezone to simulate a reported timezone issue

The bad timezone causes OSError on Windows.

See #371

2 years agochore(pool): bump to next dev version number
Daniele Varrazzo [Thu, 8 Sep 2022 12:59:47 +0000 (13:59 +0100)] 
chore(pool): bump to next dev version number

2 years agochore: bump psycopg-pool version to 3.1.2 pool-3.1.2
Daniele Varrazzo [Wed, 7 Sep 2022 11:01:04 +0000 (12:01 +0100)] 
chore: bump psycopg-pool version to 3.1.2

2 years agoMerge branch 'fix-370'
Daniele Varrazzo [Wed, 7 Sep 2022 10:56:43 +0000 (11:56 +0100)] 
Merge branch 'fix-370'

2 years agofix(pool): continue growing if the pool is not full yet
Daniele Varrazzo [Tue, 6 Sep 2022 23:49:14 +0000 (00:49 +0100)] 
fix(pool): continue growing if the pool is not full yet

Previously, starting from an empty pool, we would have stopped at the
first connection created.

2 years agofix(pool): allow growing again after failing a growing connection attempt
Daniele Varrazzo [Tue, 6 Sep 2022 16:53:40 +0000 (17:53 +0100)] 
fix(pool): allow growing again after failing a growing connection attempt

Fix #370

2 years agochore: bump to next dev version number
Daniele Varrazzo [Tue, 6 Sep 2022 11:49:19 +0000 (12:49 +0100)] 
chore: bump to next dev version number

2 years agochore: bump to version number 3.1.1 3.1.1
Daniele Varrazzo [Mon, 5 Sep 2022 12:23:19 +0000 (13:23 +0100)] 
chore: bump to version number 3.1.1

2 years agoMerge branch 'macos-broken'
Daniele Varrazzo [Mon, 5 Sep 2022 12:21:37 +0000 (13:21 +0100)] 
Merge branch 'macos-broken'

2 years agodocs: add news entry about Homebrew fix
Daniele Varrazzo [Mon, 5 Sep 2022 11:28:05 +0000 (12:28 +0100)] 
docs: add news entry about Homebrew fix

2 years agorefactor: add find_libpq_full_path() to use both in library and tests
Daniele Varrazzo [Mon, 5 Sep 2022 11:09:33 +0000 (12:09 +0100)] 
refactor: add find_libpq_full_path() to use both in library and tests

2 years agofix(macos): find the libpq in non-standard places
Daniele Varrazzo [Mon, 5 Sep 2022 09:56:06 +0000 (10:56 +0100)] 
fix(macos): find the libpq in non-standard places

2 years agochore: upgrade QEMU to v2 for build
Daniele Varrazzo [Sat, 3 Sep 2022 18:24:52 +0000 (19:24 +0100)] 
chore: upgrade QEMU to v2 for build

Also make sure to use a version not affected by bug #304

2 years agofix: make sure to pass a port number to getaddrinfo
Daniele Varrazzo [Fri, 2 Sep 2022 15:30:23 +0000 (16:30 +0100)] 
fix: make sure to pass a port number to getaddrinfo

Was passing an empty string by mistake, if no port was specified. Mostly
not a problem, except on some platforms, where it causes an
"unrecognized service" error.

Close #366.

2 years agoBump to next dev release
Daniele Varrazzo [Fri, 2 Sep 2022 15:29:28 +0000 (16:29 +0100)] 
Bump to next dev release

2 years agodocs: fix title level in news
Daniele Varrazzo [Tue, 30 Aug 2022 22:46:46 +0000 (23:46 +0100)] 
docs: fix title level in news

2 years agochore: bump version number to release 3.1 3.1
Daniele Varrazzo [Mon, 29 Aug 2022 23:01:14 +0000 (00:01 +0100)] 
chore: bump version number to release 3.1

2 years agochore: bump to version 3.0.17
Daniele Varrazzo [Mon, 29 Aug 2022 22:52:03 +0000 (23:52 +0100)] 
chore: bump to version 3.0.17

2 years agoMerge branch 'crdb-ci'
Daniele Varrazzo [Mon, 29 Aug 2022 08:58:57 +0000 (09:58 +0100)] 
Merge branch 'crdb-ci'

2 years agoci: run CRDB tests in CI
Daniele Varrazzo [Sun, 28 Aug 2022 16:56:08 +0000 (17:56 +0100)] 
ci: run CRDB tests in CI

2 years agotest(crdb): consider Decimal("Infinity") not supported in CRDB < 22
Daniele Varrazzo [Mon, 29 Aug 2022 01:23:03 +0000 (02:23 +0100)] 
test(crdb): consider Decimal("Infinity") not supported in CRDB < 22

2 years agotest(crdb) Skip tests affected by BEGIN READ ONLY bug
Daniele Varrazzo [Mon, 29 Aug 2022 00:43:50 +0000 (01:43 +0100)] 
test(crdb) Skip tests affected by BEGIN READ ONLY bug

The connection doesn't go in read-only mode if the statement is executed
using the advanced protocol.

See https://github.com/cockroachdb/cockroach/issues/87012

2 years agofix: load bytea as bytes, not memoryview, in Python implementation
Daniele Varrazzo [Mon, 29 Aug 2022 01:10:27 +0000 (02:10 +0100)] 
fix: load bytea as bytes, not memoryview, in Python implementation

2 years agochore(crdb): fix file name where to update OIDs
Daniele Varrazzo [Sun, 28 Aug 2022 16:35:33 +0000 (17:35 +0100)] 
chore(crdb): fix file name where to update OIDs

2 years agofix: use a "finally" to clear the result pointer after handling notifies
Daniele Varrazzo [Sat, 27 Aug 2022 20:29:56 +0000 (21:29 +0100)] 
fix: use a "finally" to clear the result pointer after handling notifies

The chance for this to fail failed is slim (an error in the logger? A
ctrl-c just there?) but a bit of paranoia has never hurt anyone.

2 years agoperf(pq_ctypes): use the faster byref() rather than pointer()
Daniele Varrazzo [Sat, 27 Aug 2022 20:28:22 +0000 (21:28 +0100)] 
perf(pq_ctypes): use the faster byref() rather than pointer()

2 years agoMerge branch 'pgnotifyfixes'
Daniele Varrazzo [Sun, 28 Aug 2022 11:56:52 +0000 (12:56 +0100)] 
Merge branch 'pgnotifyfixes'

2 years agodocs: add entry about segfault on fork fixed 360/head
Daniele Varrazzo [Sun, 28 Aug 2022 10:00:12 +0000 (11:00 +0100)] 
docs: add entry about segfault on fork fixed

2 years agofix: use a static notification function for PQsetNoticeReceiver.
Florian Apolloner [Sat, 27 Aug 2022 15:36:33 +0000 (17:36 +0200)] 
fix: use a static notification function for PQsetNoticeReceiver.

The connection is passed via the optional *arg.

Fixes #300

2 years agobuild: build ldap with sasl and ssl support
Daniele Varrazzo [Sat, 27 Aug 2022 11:37:45 +0000 (12:37 +0100)] 
build: build ldap with sasl and ssl support

Pass the libpq build prefix to configure in order to find just built
libraries.

2 years agobuild: upgrade system packages before building libpq
Daniele Varrazzo [Sat, 27 Aug 2022 12:12:55 +0000 (13:12 +0100)] 
build: upgrade system packages before building libpq

2 years agochore: bump version to 3.1 beta 1 to test building packages
Daniele Varrazzo [Sat, 27 Aug 2022 09:57:09 +0000 (10:57 +0100)] 
chore: bump version to 3.1 beta 1 to test building packages

2 years agochore: upgrade libpq dependencies
Daniele Varrazzo [Sat, 27 Aug 2022 10:01:48 +0000 (11:01 +0100)] 
chore: upgrade libpq dependencies

2 years agodocs: fix link to numeric-to-float recipe
Daniele Varrazzo [Wed, 24 Aug 2022 17:50:18 +0000 (19:50 +0200)] 
docs: fix link to numeric-to-float recipe

2 years agofix: avoid all uses of PQconn.send_query in pipeline mode
Daniele Varrazzo [Tue, 23 Aug 2022 22:39:40 +0000 (00:39 +0200)] 
fix: avoid all uses of PQconn.send_query in pipeline mode

We stopped using it in normal querying because, until libpq 14.4, the
libpq complained about "message type 0x33 arrived from server while
idle" (see #314). But usage remained for internal queries such as BEGIN.

In libpq 14.5, Postgres devs tried to fix the issue... and broke
PQsendQuery altogether :(

Hopefully the issue will be solved in future libpq versions, but we can
work around it completely.

Close #350.

2 years agofeat: add pq._debug module to investigate libpq issues
Daniele Varrazzo [Tue, 23 Aug 2022 22:18:56 +0000 (00:18 +0200)] 
feat: add pq._debug module to investigate libpq issues

2 years agoci: don't install modules in editable mode to run packages test
Daniele Varrazzo [Mon, 22 Aug 2022 09:42:52 +0000 (11:42 +0200)] 
ci: don't install modules in editable mode to run packages test

Something else that broke after setuptools 64 release. See tox-dev/tox#2479
for more details.

Reported something at pypa/setuptools#3557

2 years agotest: add missing asserts
Daniele Varrazzo [Sat, 13 Aug 2022 21:42:05 +0000 (23:42 +0200)] 
test: add missing asserts

2 years agochore: fix typos spotted by codespell 2.2.1
Denis Laxalde [Fri, 19 Aug 2022 17:05:05 +0000 (19:05 +0200)] 
chore: fix typos spotted by codespell 2.2.1

2 years agoFix typo: shoul => should
Sergey Fedoseev [Wed, 17 Aug 2022 16:30:42 +0000 (21:30 +0500)] 
Fix typo: shoul => should

2 years agodocs: fix a typo in FileWriter docstring
Denis Laxalde [Tue, 16 Aug 2022 14:17:29 +0000 (16:17 +0200)] 
docs: fix a typo in FileWriter docstring

2 years agofix: consider pipeline mode not supported with libpq 14.5
Daniele Varrazzo [Mon, 15 Aug 2022 11:52:04 +0000 (13:52 +0200)] 
fix: consider pipeline mode not supported with libpq 14.5

Improve diagnostics about the reason why the pipeline mode is not
supported. Improve not supported tests.

2 years agotest: allow to use the != operator in make.libpq
Daniele Varrazzo [Mon, 15 Aug 2022 11:46:46 +0000 (13:46 +0200)] 
test: allow to use the != operator in make.libpq

2 years agofeat: use version() as __build_version__ for pq python implementation
Daniele Varrazzo [Mon, 15 Aug 2022 11:41:01 +0000 (13:41 +0200)] 
feat: use version() as __build_version__ for pq python implementation

Having __build_version__ to be null makes version checking needlessly
convoluted and version() has to be used anyway as fallback.

2 years agorefactor(psycopg_c): define and use an ARRAYSIZE macro
Daniele Varrazzo [Sat, 13 Aug 2022 01:44:17 +0000 (03:44 +0200)] 
refactor(psycopg_c): define and use an ARRAYSIZE macro

2 years agorefactor(psycopg_c): drop deprecated use of DEF in Cython
Daniele Varrazzo [Sat, 13 Aug 2022 01:19:35 +0000 (03:19 +0200)] 
refactor(psycopg_c): drop deprecated use of DEF in Cython

See cython/cython#4310

2 years agoperf(psycopg_c): avoid to create a tuple in fetch_many() loop
Daniele Varrazzo [Mon, 15 Aug 2022 00:50:50 +0000 (02:50 +0200)] 
perf(psycopg_c): avoid to create a tuple in fetch_many() loop

2 years agorefactor: expose optimized functions directly from the origin modules
Daniele Varrazzo [Mon, 15 Aug 2022 00:49:24 +0000 (02:49 +0200)] 
refactor: expose optimized functions directly from the origin modules

Don't check for the availability of the optimized module in every module
using potentially optimized functions.

2 years agofix(psycopg_c): guard PQflush calls for invalid connections
Daniele Varrazzo [Mon, 15 Aug 2022 00:46:19 +0000 (02:46 +0200)] 
fix(psycopg_c): guard PQflush calls for invalid connections

Unlike other libpq functions, PQflush() doesn't have a guard for invalid
connections and will segfault if the connection gets closed.

Also check the return value of the PQflush calls: as we are not using
the wrapped version pgconn.flush(), the function doesn't automatically
throw an exception on failure.

2 years agochore: use `license_files` entry in metadata
Daniele Varrazzo [Sat, 13 Aug 2022 00:32:21 +0000 (02:32 +0200)] 
chore: use `license_files` entry in metadata

`license_file` is deprecated.

2 years agoci: specify non-editable install in tox tests
Daniele Varrazzo [Fri, 12 Aug 2022 23:35:41 +0000 (01:35 +0200)] 
ci: specify non-editable install in tox tests

Work around tox-dev/tox#2479 (editable installs in tox broken by setuptools
v64) and tox-dev/tox#2480 (failure to use `toxinidir` in `deps`).

Use -e consistently in the linting steps instead, because they are
defined with skip_install=True so it seems a good idea to use it. These
steps don't seem affect by the setuptools issue.

2 years agorefactor: drop repeated operation in sync/async TypeInfo.fetch()
Daniele Varrazzo [Fri, 12 Aug 2022 02:00:12 +0000 (04:00 +0200)] 
refactor: drop repeated operation in sync/async TypeInfo.fetch()

2 years agodocs: fix typo "flaot"
Daniele Varrazzo [Fri, 12 Aug 2022 01:43:59 +0000 (03:43 +0200)] 
docs: fix typo "flaot"

2 years agodocs: reword blurb about binary copy and make it a warning
Daniele Varrazzo [Fri, 12 Aug 2022 01:40:20 +0000 (03:40 +0200)] 
docs: reword blurb about binary copy and make it a warning

2 years agodocs: fix typo: "reurned"
Daniele Varrazzo [Fri, 12 Aug 2022 01:25:42 +0000 (03:25 +0200)] 
docs: fix typo: "reurned"

Reurned... ⚱️ ->⚱️

2 years agodocs: fix example and reword around multiple-query statements
Daniele Varrazzo [Fri, 12 Aug 2022 01:25:15 +0000 (03:25 +0200)] 
docs: fix example and reword around multiple-query statements

2 years agodocs: link to the "from psycopg2" page from the "basic usage" top section
Daniele Varrazzo [Fri, 12 Aug 2022 01:24:25 +0000 (03:24 +0200)] 
docs: link to the "from psycopg2" page from the "basic usage" top section

2 years agodocs: quote examples of pip install with extras
Daniele Varrazzo [Fri, 5 Aug 2022 23:47:54 +0000 (01:47 +0200)] 
docs: quote examples of pip install with extras

zsh doesn't parse the square brackets as expected.

See #347.

3 years agoFix syntax in example
Sergey Fedoseev [Sat, 30 Jul 2022 12:39:47 +0000 (17:39 +0500)] 
Fix syntax in example

3 years agobuild: use "latest" version of github builders
Daniele Varrazzo [Thu, 28 Jul 2022 13:19:16 +0000 (15:19 +0200)] 
build: use "latest" version of github builders

3 years agoMerge branch 'build-macos-arm64'
Daniele Varrazzo [Thu, 28 Jul 2022 11:24:42 +0000 (13:24 +0200)] 
Merge branch 'build-macos-arm64'

3 years agodocs: mention macOS arm64 build script in news file
Daniele Varrazzo [Thu, 28 Jul 2022 11:17:23 +0000 (13:17 +0200)] 
docs: mention macOS arm64 build script in news file