Single approach to ext types encode/decode #252
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
DifferentialOrange
changed the base branch from
master
to
DifferentialOrange/gh-250-unittest-warnings
Merged
DifferentialOrange
force-pushed
the
DifferentialOrange/gh-250-unittest-warnings
branch
from
cb12b7b to
7baa481
Compare
Base automatically changed from
DifferentialOrange/gh-250-unittest-warnings
to
master
October 27, 2022 12:22
DBAPI2 compliance tests are not implemented here but inherited from external module [1]. Two tests from this module open a connection and forget to close it. The issue had been filed together with patch PR to module repository [2], but the last update was 7 years ago so it is possibly that it would never be merged. This patch adds this PR change with method overwrite. 1. https://pypi.org/project/dbapi-compliance/ 2. baztian/dbapi-compliance#5 Part of #250
This option is required so it would be possible to decode Datetime with external function without constructing excessive pandas.Timestamp object. Follows #204
Extract tarantool.Datetime encode and decode to external functions. This is a breaking change, but since there is no tagged release with Datetime yet and API was more internal rather than public, it shouldn't be an issue. Follows #204
Extract tarantool.Interval encode and decode to external functions. This is a breaking change, but since there is no tagged release with Interval yet and API was more internal rather than public, it shouldn't be an issue. Follows #229
DifferentialOrange
force-pushed
the
DifferentialOrange/ext-classes-code-health
branch
from
07f6d46 to
db9e5b8
Compare
DifferentialOrange
deleted the
DifferentialOrange/ext-classes-code-health
branch
DifferentialOrange
added a commit
that referenced
this pull request
Nov 9, 2022
Overview This release introduces the support of extention types (decimal, uuid, error, datetime, interval) in MessagePack, various IProto features support (feature discovery and push protocol) and major infrastructure updates (scm version computation, full documentation for external and internal API both as code docstrings and readthedocs HTML, deb and RPM packages, and everything is processed with CI/CD pipelines). Breaking changes This release should not break any existing behavior. New features - Backport ConnectionPool support for Python 3.6 (PR #245). - Support iproto feature discovery (#206). - Decimal type support (#203). - UUID type support (#202). - Support extra information for iproto errors (#232). - Error extension type support (#232). - Datetime type support and tarantool.Datetime type (#204, PR #252). Tarantool datetime objects are decoded to `tarantool.Datetime` type. `tarantool.Datetime` may be encoded to Tarantool datetime objects. You can create `tarantool.Datetime` objects either from MessagePack data or by using the same API as in Tarantool: ```python dt1 = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321) dt2 = tarantool.Datetime(timestamp=1661969274) dt3 = tarantool.Datetime(timestamp=1661969274, nsec=308543321) ``` `tarantool.Datetime` exposes `year`, `month`, `day`, `hour`, `minute`, `sec`, `nsec`, `timestamp` and `value` (integer epoch time with nanoseconds precision) properties if you need to convert `tarantool.Datetime` to any other kind of datetime object: ```python pdt = pandas.Timestamp(year=dt.year, month=dt.month, day=dt.day, hour=dt.hour, minute=dt.minute, second=dt.sec, microsecond=(dt.nsec // 1000), nanosecond=(dt.nsec % 1000)) ``` Use `tzoffset` parameter to set up offset timezone: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tzoffset=180) ``` You may use `tzoffset` property to get timezone offset of a datetime object. Use `tz` parameter to set up timezone name: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tz='Europe/Moscow') ``` If both `tz` and `tzoffset` is specified, `tz` is used. You may use `tz` property to get timezone name of a datetime object. `timestamp_since_utc_epoch` is a parameter to set timestamp convertion behavior for timezone-aware datetimes. If ``False`` (default), behaves similar to Tarantool `datetime.new()`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640984400.0 ``` Thus, if ``False``, datetime is computed from timestamp since epoch and then timezone is applied without any convertion. In that case, `dt.timestamp` won't be equal to initialization `timestamp` for all timezones with non-zero offset. If ``True``, behaves similar to `pandas.Timestamp`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 03:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640995200.0 ``` Thus, if ``True``, datetime is computed in a way that `dt.timestamp` will always be equal to initialization `timestamp`. - Datetime interval type support and tarantool.Interval type (#229). Tarantool datetime interval objects are decoded to `tarantool.Interval` type. `tarantool.Interval` may be encoded to Tarantool interval objects. You can create `tarantool.Interval` objects either from MessagePack data or by using the same API as in Tarantool: ```python di = tarantool.Interval(year=-1, month=2, day=3, hour=4, minute=-5, sec=6, nsec=308543321, adjust=tarantool.IntervalAdjust.NONE) ``` Its attributes (same as in init API) are exposed, so you can use them if needed. - Datetime interval arithmetic support (#229). Valid operations: - `tarantool.Datetime` + `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Datetime` = `tarantool.Interval` - `tarantool.Interval` + `tarantool.Interval` = `tarantool.Interval` - `tarantool.Interval` - `tarantool.Interval` = `tarantool.Interval` Since `tarantool.Interval` could contain `month` and `year` fields and such operations could be ambiguous, you can use `adjust` field to tune the logic. The behavior is the same as in Tarantool, see [Interval arithmetic RFC](https://github.com/tarantool/tarantool/wiki/Datetime-Internals#interval-arithmetic). - `tarantool.IntervalAdjust.NONE` -- only truncation toward the end of month performed (default mode). ```python >>> dt = tarantool.Datetime(year=2022, month=3, day=31) datetime: Timestamp('2022-03-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.NONE) >>> dt + di datetime: Timestamp('2022-04-30 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.EXCESS` -- overflow mode, without any snap or truncation to the end of month, straight addition of days in month, stopping over month boundaries if there is less number of days. ```python >>> dt = tarantool.Datetime(year=2022, month=1, day=31) datetime: Timestamp('2022-01-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.EXCESS) >>> dt + di datetime: Timestamp('2022-03-02 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.LAST` -- mode when day snaps to the end of month, if happens. ```python >>> dt = tarantool.Datetime(year=2022, month=2, day=28) datetime: Timestamp('2022-02-28 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.LAST) >>> dt + di datetime: Timestamp('2022-03-31 00:00:00'), tz: "" ``` - Full documentation of internal and external API (#67). Bugfixes - Allow any MessagePack supported type as a request key (#240). - Make connection close idempotent (#250). Infrastructure - Use git version to set package version (#238). - Test pip install from branch (PR #241). - Pack and publish pip, RPM and deb packages with GitHub Actions (#164, #198). - Publish on readthedocs with CI/CD (including PRs) (#67).
Merged
DifferentialOrange
added a commit
that referenced
this pull request
Nov 9, 2022
Overview This release introduces the support of extention types (decimal, uuid, error, datetime, interval) in MessagePack, various IProto features support (feature discovery and push protocol) and major infrastructure updates (scm version computation, full documentation for external and internal API both as code docstrings and readthedocs HTML, deb and RPM packages, and everything is processed with CI/CD pipelines). Breaking changes This release should not break any existing behavior. New features - Backport ConnectionPool support for Python 3.6 (PR #245). - Support iproto feature discovery (#206). - Decimal type support (#203). - UUID type support (#202). - Support extra information for iproto errors (#232). - Error extension type support (#232). - Datetime type support and tarantool.Datetime type (#204, PR #252). Tarantool datetime objects are decoded to `tarantool.Datetime` type. `tarantool.Datetime` may be encoded to Tarantool datetime objects. You can create `tarantool.Datetime` objects either from MessagePack data or by using the same API as in Tarantool: ```python dt1 = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321) dt2 = tarantool.Datetime(timestamp=1661969274) dt3 = tarantool.Datetime(timestamp=1661969274, nsec=308543321) ``` `tarantool.Datetime` exposes `year`, `month`, `day`, `hour`, `minute`, `sec`, `nsec`, `timestamp` and `value` (integer epoch time with nanoseconds precision) properties if you need to convert `tarantool.Datetime` to any other kind of datetime object: ```python pdt = pandas.Timestamp(year=dt.year, month=dt.month, day=dt.day, hour=dt.hour, minute=dt.minute, second=dt.sec, microsecond=(dt.nsec // 1000), nanosecond=(dt.nsec % 1000)) ``` Use `tzoffset` parameter to set up offset timezone: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tzoffset=180) ``` You may use `tzoffset` property to get timezone offset of a datetime object. Use `tz` parameter to set up timezone name: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tz='Europe/Moscow') ``` If both `tz` and `tzoffset` is specified, `tz` is used. You may use `tz` property to get timezone name of a datetime object. `timestamp_since_utc_epoch` is a parameter to set timestamp convertion behavior for timezone-aware datetimes. If ``False`` (default), behaves similar to Tarantool `datetime.new()`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640984400.0 ``` Thus, if ``False``, datetime is computed from timestamp since epoch and then timezone is applied without any convertion. In that case, `dt.timestamp` won't be equal to initialization `timestamp` for all timezones with non-zero offset. If ``True``, behaves similar to `pandas.Timestamp`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 03:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640995200.0 ``` Thus, if ``True``, datetime is computed in a way that `dt.timestamp` will always be equal to initialization `timestamp`. - Datetime interval type support and tarantool.Interval type (#229). Tarantool datetime interval objects are decoded to `tarantool.Interval` type. `tarantool.Interval` may be encoded to Tarantool interval objects. You can create `tarantool.Interval` objects either from MessagePack data or by using the same API as in Tarantool: ```python di = tarantool.Interval(year=-1, month=2, day=3, hour=4, minute=-5, sec=6, nsec=308543321, adjust=tarantool.IntervalAdjust.NONE) ``` Its attributes (same as in init API) are exposed, so you can use them if needed. - Datetime interval arithmetic support (#229). Valid operations: - `tarantool.Datetime` + `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Datetime` = `tarantool.Interval` - `tarantool.Interval` + `tarantool.Interval` = `tarantool.Interval` - `tarantool.Interval` - `tarantool.Interval` = `tarantool.Interval` Since `tarantool.Interval` could contain `month` and `year` fields and such operations could be ambiguous, you can use `adjust` field to tune the logic. The behavior is the same as in Tarantool, see [Interval arithmetic RFC](https://github.com/tarantool/tarantool/wiki/Datetime-Internals#interval-arithmetic). - `tarantool.IntervalAdjust.NONE` -- only truncation toward the end of month performed (default mode). ```python >>> dt = tarantool.Datetime(year=2022, month=3, day=31) datetime: Timestamp('2022-03-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.NONE) >>> dt + di datetime: Timestamp('2022-04-30 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.EXCESS` -- overflow mode, without any snap or truncation to the end of month, straight addition of days in month, stopping over month boundaries if there is less number of days. ```python >>> dt = tarantool.Datetime(year=2022, month=1, day=31) datetime: Timestamp('2022-01-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.EXCESS) >>> dt + di datetime: Timestamp('2022-03-02 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.LAST` -- mode when day snaps to the end of month, if happens. ```python >>> dt = tarantool.Datetime(year=2022, month=2, day=28) datetime: Timestamp('2022-02-28 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.LAST) >>> dt + di datetime: Timestamp('2022-03-31 00:00:00'), tz: "" ``` - Full documentation of internal and external API (#67). Bugfixes - Allow any MessagePack supported type as a request key (#240). - Make connection close idempotent (#250). Infrastructure - Use git version to set package version (#238). - Test pip install from branch (PR #241). - Pack and publish pip, RPM and deb packages with GitHub Actions (#164, #198). - Publish on readthedocs with CI/CD (including PRs) (#67).
DifferentialOrange
added a commit
that referenced
this pull request
Nov 9, 2022
Overview This release introduces the support of extention types (decimal, uuid, error, datetime, interval) in MessagePack, various IProto features support (feature discovery and push protocol) and major infrastructure updates (scm version computation, full documentation for external and internal API both as code docstrings and readthedocs HTML, deb and RPM packages, and everything is processed with CI/CD pipelines). Breaking changes This release should not break any existing behavior. New features - Backport ConnectionPool support for Python 3.6 (PR #245). - Support iproto feature discovery (#206). - Decimal type support (#203). - UUID type support (#202). - Support extra information for iproto errors (#232). - Error extension type support (#232). - Datetime type support and tarantool.Datetime type (#204, PR #252). Tarantool datetime objects are decoded to `tarantool.Datetime` type. `tarantool.Datetime` may be encoded to Tarantool datetime objects. You can create `tarantool.Datetime` objects either from MessagePack data or by using the same API as in Tarantool: ```python dt1 = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321) dt2 = tarantool.Datetime(timestamp=1661969274) dt3 = tarantool.Datetime(timestamp=1661969274, nsec=308543321) ``` `tarantool.Datetime` exposes `year`, `month`, `day`, `hour`, `minute`, `sec`, `nsec`, `timestamp` and `value` (integer epoch time with nanoseconds precision) properties if you need to convert `tarantool.Datetime` to any other kind of datetime object: ```python pdt = pandas.Timestamp(year=dt.year, month=dt.month, day=dt.day, hour=dt.hour, minute=dt.minute, second=dt.sec, microsecond=(dt.nsec // 1000), nanosecond=(dt.nsec % 1000)) ``` Use `tzoffset` parameter to set up offset timezone: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tzoffset=180) ``` You may use `tzoffset` property to get timezone offset of a datetime object. Use `tz` parameter to set up timezone name: ```python dt = tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, sec=54, nsec=308543321, tz='Europe/Moscow') ``` If both `tz` and `tzoffset` is specified, `tz` is used. You may use `tz` property to get timezone name of a datetime object. `timestamp_since_utc_epoch` is a parameter to set timestamp convertion behavior for timezone-aware datetimes. If ``False`` (default), behaves similar to Tarantool `datetime.new()`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=False) >>> dt datetime: Timestamp('2022-01-01 00:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640984400.0 ``` Thus, if ``False``, datetime is computed from timestamp since epoch and then timezone is applied without any convertion. In that case, `dt.timestamp` won't be equal to initialization `timestamp` for all timezones with non-zero offset. If ``True``, behaves similar to `pandas.Timestamp`: ```python >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 00:00:00'), tz: "" >>> dt.timestamp 1640995200.0 >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow', ... timestamp_since_utc_epoch=True) >>> dt datetime: Timestamp('2022-01-01 03:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow" >>> dt.timestamp 1640995200.0 ``` Thus, if ``True``, datetime is computed in a way that `dt.timestamp` will always be equal to initialization `timestamp`. - Datetime interval type support and tarantool.Interval type (#229). Tarantool datetime interval objects are decoded to `tarantool.Interval` type. `tarantool.Interval` may be encoded to Tarantool interval objects. You can create `tarantool.Interval` objects either from MessagePack data or by using the same API as in Tarantool: ```python di = tarantool.Interval(year=-1, month=2, day=3, hour=4, minute=-5, sec=6, nsec=308543321, adjust=tarantool.IntervalAdjust.NONE) ``` Its attributes (same as in init API) are exposed, so you can use them if needed. - Datetime interval arithmetic support (#229). Valid operations: - `tarantool.Datetime` + `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Interval` = `tarantool.Datetime` - `tarantool.Datetime` - `tarantool.Datetime` = `tarantool.Interval` - `tarantool.Interval` + `tarantool.Interval` = `tarantool.Interval` - `tarantool.Interval` - `tarantool.Interval` = `tarantool.Interval` Since `tarantool.Interval` could contain `month` and `year` fields and such operations could be ambiguous, you can use `adjust` field to tune the logic. The behavior is the same as in Tarantool, see [Interval arithmetic RFC](https://github.com/tarantool/tarantool/wiki/Datetime-Internals#interval-arithmetic). - `tarantool.IntervalAdjust.NONE` -- only truncation toward the end of month performed (default mode). ```python >>> dt = tarantool.Datetime(year=2022, month=3, day=31) datetime: Timestamp('2022-03-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.NONE) >>> dt + di datetime: Timestamp('2022-04-30 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.EXCESS` -- overflow mode, without any snap or truncation to the end of month, straight addition of days in month, stopping over month boundaries if there is less number of days. ```python >>> dt = tarantool.Datetime(year=2022, month=1, day=31) datetime: Timestamp('2022-01-31 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.EXCESS) >>> dt + di datetime: Timestamp('2022-03-02 00:00:00'), tz: "" ``` - `tarantool.IntervalAdjust.LAST` -- mode when day snaps to the end of month, if happens. ```python >>> dt = tarantool.Datetime(year=2022, month=2, day=28) datetime: Timestamp('2022-02-28 00:00:00'), tz: "" >>> di = tarantool.Interval(month=1, adjust=tarantool.IntervalAdjust.LAST) >>> dt + di datetime: Timestamp('2022-03-31 00:00:00'), tz: "" ``` - Full documentation of internal and external API (#67). Bugfixes - Allow any MessagePack supported type as a request key (#240). - Make connection close idempotent (#250). Infrastructure - Use git version to set package version (#238). - Test pip install from branch (PR #241). - Pack and publish pip, RPM and deb packages with GitHub Actions (#164, #198). - Publish on readthedocs with CI/CD (including PRs) (#67).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
api: pandas way to build datetime from timestamp
This option is required so it would be possible to decode Datetime with
external function without constructing excessive pandas.Timestamp
object.
Follows #204
api: extract datetime encode/decode from class
Extract tarantool.Datetime encode and decode to external functions. This
is a breaking change, but since there is no tagged release with Datetime
yet and API was more internal rather than public, it shouldn't be an
issue.
Follows #204
api: extract interval encode/decode from class
Extract tarantool.Interval encode and decode to external functions. This
is a breaking change, but since there is no tagged release with Interval
yet and API was more internal rather than public, it shouldn't be an
issue.
Follows #229