Expand tests #64
Merged
Expand tests #64
Conversation
This file contains 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
aastafiev
pushed a commit
to aastafiev/clickhouse-connect
that referenced
this pull request
Mar 4, 2023
* Add additional Python versions to tests * Fix typo * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Remove arrow temporarily * Finalize test matrix
genzgd
added a commit
that referenced
this pull request
Mar 6, 2023
* Additional table engine definitions * Add directory, fix test * More distribution updates * Add passthrough of connection options, clean up TLS settigns * Replaced all dict typing with Dict * Replaced all dict typing with Dict * Fix SqlAlchemy table exists method * Workflows (#5) * Add first test workflow * Fix cache typo * Try to fix python setup * Fix another typo * Fix cache * requirements.txt? * Hack superset directories * Hack superset dependencies * Fix another typo * Fix Jinja dependency * More mucking with test dependencies * Build cython extensions in tests * Attempt to fix cython build * Hack for old Python IPv6Addresses * Exclude Python 3.7 from tests because Superset * Don't test Date32 (not supported in CH 21.8) * Make tests work with ClickHouse 21.8 * Exclude Python 3.10 from tests temporarily * Skip multiple test runs on pull request * Add workflow for publish releases to PyPI (#6) * Cleaning up directories and improving superset testing infrastructure. * Install dependencies for pylint * Fix typo * Testing of release workflow * Fix typo * Testing release 1 * Testing release 2 * Testing release 3 * Testing release 4 * Testing release 5 * Testing release 6 * Testing windows * Testing windows 3 * Testing windows 4 * Testing windows 5 * Test with artifacts * Test with artifacts 2 * Test with artifacts 3 * Test with artifacts 4 * Test with artifacts 6 * Test with artifacts 7 * Test with artifacts 8 * Test with pypi 1 * Final test * Final test 2 * Final test 3 * Publish Fixes for PyPI (#7) * Test new release publish * Test new release publish 2 * Test new release publish 3 * Test new release publish 4 * Test new release publish 5 * Test new release publish 6 * Test new release publish 7 * Test new release publish 8 * Test new release publish 8 * Test new release publish 9 * Test new release publish 10 * Test new release publish 11 * Test new release publish 12 * Test new release publish 12 * Test new release publish 14 * Test new release publish 15 * Fix Release Issues on PyPI * Add code to handle LIMIT 0 queries to return names and types (DBT compatibility) * Add `metadata_only` to query settings to retrieve only column names * Incorporate PR review comments * Stop accidentially using root logger * Add wheel for lint workflow * Try to make pylint happy * Add pylint extension * Stop removing newlines from SQL queries, pin Pylint version * One more pylint tweak * Add SQL comment tests * First work on storing nested elements * Fix some deprecation, including implicit coercion to float/int for array.array inserts * Lint fix * Refactor session params, promote to beta (#12) * Refactor settings validation and passthrough, promote to Beta status * Tiny cleanup * Quiet linter * Fix non-local test code * Include separate POST data with client.command (#14) * One data attempt * Lint fixes * More lint fixes * Send all SQL as POST data for client.command * Add validation utility for entrypoints * Fix lint * Fix timeout and retry issues (#16) * Only retry queries, not inserts or commands, and use progress headers as keep alive for long running queries * Ensure string data is encoded as utf8 * Lint fixes * Add chinese character test * Decode ClickHouse Error Response to UTF-8 (#19) * Return ClickHouse error messages as decoded UTF-8 * Return ClickHouse error messages as decoded UTF-8 * Don't try to send readonly settings (even for send_progress or compression) (#20) * Make sure we don't send read only settings, remove brotli (seems broken), fix ping * Fix Lint * Don't modify keyword arguments * Add socket options for TCP keep alive to HTTP adapter (#21) * Add socket options for TCP keep alive to HTTP adapter * Tight up filtering of query parameters/settings that are not server settings * Fix lint * Arrow and modest Client method refactoring (#22) * Add arrow integration step 1 * Fix lint * Add socket options for TCP keep alive to HTTP adapter * Tight up filtering of query parameters/settings that are not server settings * Fix lint * modest cleanup and refactoring of client methods * swap abstract method parameters * Fix arrow doc strings * Support for Nested columns (#23) * Checkpoint for nested column support * Nested column support basic functionality * Fix lint * More lint fixes * Low card fix (#25) * Fix issues with empty inserts and numeric LowCardinality types * Checkpoint for nested column support * Nested column support basic functionality * Fix lint * More lint fixes * Fix issues with empty inserts and LowCardinality numeric types * Fix lint * Try to fix tests for ClickHouse 21.8 * Try to fix tests for ClickHouse 21.8 take 2 * Try to fix tests for ClickHouse 21.8 take 3 * Try to fix tests for ClickHouse 21.8 take 5 * Add changelog * Update changelog * Include pip installation (#26) * Format control/JSON/Improvements (#28) * Format control checkpoint * Checkpoint on data format handling * Lint fixes * Lint fixes * Format control checkpoint * Fix lint * Fix lint * Format checkpoint * Fix named tuples * Named tuple tweaks and test * Fix lint * JSON support checkpoint * Enhance JSON library selection and usage, use QueryContext for all queries * Fix lint * Formatting fixes plus more format tests * Fix lint * Fix lint * Expand read format tests * Fix lint, remove spaces from python JSON serialization * format doc checkpoint * Lint fix * Fix SQL comment related issues * Fix comment and sqlalchemy issues * Lint fixes * Update changelog * Fix sql comment issues in DBAPI cursor (#29) * Changelog update * Zero limit fix (#32) * Fix sql comment issues in DBAPI cursor * Fix 0 limit issue. * Arrow inserts (#33) * Arrow table inserts, backtick quote all column and table identifiers * Tweak developer version name * Remove pip local install * Update test requirements * Build cython for lint * Add raw insert test and identifier quote tweak * Fix multibyte utf-8 writes (#34) * added wait_end_of_query in valid_transport_settings (#36) `SELECT name, value, changed, description, type, readonly FROM system.settings` do not return `wait_end_of_query` setting, so it added in HttpClient.valid_transport_settings fixes error handling for errors during queries which use `ON CLUSTER` ClickHouse/ClickHouse#15124 * Test updates (#37) * Update test matrix * First pass at adding cloud tests * Minor updates, put timestamp first in test database * Fix DBI cursor, parameterized query fix and updates (#39) * Fix datetime formatting for parameterization * Fix cursor description problem with commands, support simple format query strings * Sqla 1.14 update (#40) * Fix sqlalchemy 1.4 issue, update superset testing version * fix typo, update changelog * Tweak test requirements * Tweak test requirements * Tweak test requirements (#42) * add query_retries client parameter (#43) * Tweak test requirements * Add query_retries parameter * All the wheels (#45) * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update Version * First cibuildwheel test * Fix yaml * Update Version * Limit test builds * Update Version * Test more versions, test pypi upload * Update Version * Add macos builds * Yaml fixes * Update Version * Tweak build versions * Fix yaml * Update Version * Add non-x86 linux builds * Update Version * Test push only * Conditional git push to facilitate testing * More workflow fixes * Test publish all * Test publish all * Test publish all * Test publish all * Test publish all * Update VERSION * Test publish all * Test publish all * Mac OS build test * Update VERSION * Test Mac OS x86 * Test Mac OS x86 * Test Mac OS x86 * Update VERSION * All builds again with Cython * Update VERSION * All builds again with Cython * Update VERSION * Mac only test * Update VERSION * Mac only Python 3.11 test * Update VERSION * Full Python 3.11 test * Update version * Cleanup for improved builds release * Fix release job * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Remove invalid python 3.11 * Fix validation errors in Superset connection dialog (#46) * Retry keep alive disconnects (#48) * Append unixtime millis to the test db name when using Cloud (#50) * Append unixtime millis to the test db name when using Cloud * Fix create_db database name * Update ignored test branches Co-authored-by: Geoff Genz <geoff@clickhouse.com> * Add Python code highlighting (#52) * Remove rb (#54) * Remove row_binary format * Append unixtime millis to the test db name when using Cloud (#50) * Append unixtime millis to the test db name when using Cloud * Fix create_db database name * Update ignored test branches Co-authored-by: Geoff Genz <geoff@clickhouse.com> * Cleanup for row_binary removal Co-authored-by: Serge Klochkov <3175289+slvrtrn@users.noreply.github.com> * Valid params fix (#55) * Fix various superset issues * Fix lint * Type fixes (#58) * Fix UInt64 type bug, fix Superset big integers * Fix typos * Fix lint * Release 0.3.2 (#60) * Handle queries that are actually commands, simply and clean up tests * Update version * Fix test ordering * Test fixes * Fix command regex for 'SHOW CREATE' (#62) * Remove exists and explain from commands * SQLAlchemy Fixes (#63) * Fix JSON insert for SQLAlchemy, update related tests * Clean up JSON test configuration * Remove UUID from JSON * SQLAlchemy cleanup to provide better 1.4 compatibility * Update changelog * Update README * Update README.md * Update README.md * Expand tests (#64) * Add additional Python versions to tests * Fix typo * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Remove arrow temporarily * Finalize test matrix * Fix classifier * Version 0.3.5 build Python 3.11 wheels * Fix query_context.updated_copy method, add first examples (#66) * Chunked inserts (#67) * Chunked and compressed inserts * Fix chunk writes * Fix lint * One more chunk fix * Avoid copying when building insert chunks * Fix typo * Fix compression bug * Insert context (#70) * Insert context checkpoint * Pandas rework * Handle internal error while streaming inserts * Add query result in column format * Reusable insert context work * Improve InsertContext, add pandas tests * Fix lint * Update changelog * Use CaseInsensitiveDict for sqla_type_map (#71) * Use CaseInsensitiveDict for sqla_type_map * Fix type * Release 0 4 1 (#72) * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Add select connect_version to allowing checking from Superset SQLLab * Fix some numpy big int and datetime issues * Ensure little endian numpy types * Clean up settings, minor fixes (#78) * Various settings rework * Fix arrow string setting * Change Pandas NaN handling * Allow setting http client HTTPAdapter, include VERSION in distribution (#79) * Send all recognized transport headers regardless if recognized * update Changelog to follow a common pattern (#81) * Fix common settings bug (#82) * Bug fixes (#83) * Fix JSON inserts using ujson library (#85) * File insert, server side bindings (#87) * Add tool functions for uploading files, support dsn for client creation * Add dsn test and get_client_setting method * Fix dsn host * Add support for server side parameter binding * Fix dbapi issue, update test matrix * Fix lint * Fixes to align with doc * Improve QueryContext usability * Cleanup for contexts and raw_query method * Additional tweaks to match updated documentation * Update test matrix, removed unused docs directory * Update year in LICENSE (#88) * Streaming, C improvements, refactoring (#92) * Spiking around streaming reads * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Add .c files * Native streaming checkpoint * Native streaming checkpoint * Native streaming checkpoint * Tiny cleanup * Streaming updates and refactoring * Fix Python 3.7 issues * Fix versioning glitch * Add fixed string optimizations * Update changelog, uncomment C optimizations * Make sure json allowed for tests. * Make sure json allowed for tests (working version) * Address some PR comments * Noting that 0.5.0 is not yet released * Compression updates, performance improvements (#95) * Spiking around streaming reads * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Add .c files * Native streaming checkpoint * Native streaming checkpoint * Native streaming checkpoint * Tiny cleanup * Streaming updates and refactoring * Fix Python 3.7 issues * Fix versioning glitch * Add fixed string optimizations * Update changelog, uncomment C optimizations * Make sure json allowed for tests. * Make sure json allowed for tests (working version) * Address some PR comments * Remove requests layer * Compression updates (in progress) * Checkpoint for urllib migration * Clean up TLS stuff * Fix settings issue with SQLAlchemy connections, compression updates * Fix certifi requirement * Revert sqlalchemy version for tests * Compression reorganization, QueryResult and Cython buffer cleanup * Client construction cleanup, minor tweaks * Clarify "certifi" package use * Clean up compression config, add DSN test * Fix lint * Throw exception on unsupported compression type * Fix source build issue * Fix source build issue * Still trying to fix builds * Fix ping * Fix handling of None database (#97) * Fix QueryResult "with" statement context (#101) * Fix QueryResult with context * Add unit test for QueryResult with statement * Prep for 0.5.3 patch release * add a PR template (#103) * add pr template * Exclude github templates from tests * Exclude github templates from tests take 2 Co-authored-by: Geoff Genz <geoff@clickhouse.com> * v0.5.4 -- Numpy/Pandas improvements, stream refactoring (#105) * Add C conversion for IPv4 native * Tiny cleanup * Add DateTime to c data_conv * First work on specialized numpy columns * numpy rework checkpoint * numpy rework checkpoint * date calculation checkpoint * date calculation checkpoint * Cleanup checkpoint * query context cleanup * numpy checkpoint * Numpy read checkpoint * context rework checkpoint * format rework checkpoint * format rework checkpoint * progress toward pd/np reads * progress toward pd/np reads * stream cleanup checkpoint * Add C optimization for UUID * Minor cleanup, ready for tests * Fix lint * First work on optimizing None/NULL handling in Cython * Tweak retries/connection resets * Context cleanup checkpoint * Fix lint * Finalize for 0.5.4 release * Fix lint * Fix weird ORDER BY in test * Tweak Cython UUID transform * Add product_name, fix Numpy for Superset * Downgrade Numpy Build for Python 3.7 * Clean up numpy build dependencies * Fix numpy typo * Don't build PyPy 3.7 binary wheels * build debugging * Tweak release/test actions * Fix client_name refs * Update test matrix, try to make cloud tests less flaky * One more attempt at flaky cloud tests * Cleanup changelog * Fix numpy loc card (#110) * Fix numpy lowcardinality exception * Bump version * Fix dataframe insert mutation (#113) * Restore proxy support (#115) * update perf_test, tweak pandas test * Return empty df instead of list (#119) * Add query_tz and column_tzs query arguments (#121) * Add query_tz and column_tzs query arguments * Move get_client docstring, add preliminary support for client_protocol_version * Streaming fixes (#124) * zstd problem debugging * zstd problem debugging * Stream error handling cleanup * Optimize null strings, bug fixes (#125) * zstd problem debugging * zstd problem debugging * Stream error handling cleanup * Optimize nullable strings * Respect use_none=False, add tests * Fix stream error extraction * Make stream error extraction a little tighter * Close urllib3 connections on exit * Close urllib3 connections on exit * Better error message for invalid numpy DateTime64 types (#127) * added column_oriented check to the named_results function (#129) * SSH Tunnel example, support for client_protocol_version (#131) * Better error message for invalid numpy DateTime64 types * tweak test for Python 3.7 * Server host name (#134) * Fix DSN database parsing * Add server_host_name * Fix ssh tunnel example * Update issue templates * Update issue templates * Fix insert test * Fix lint * Panda nulls (#137) * Handle pandas timezones * use_na_values checkpoint * use_na_values checkpoint * use_na_values checkpoint * use_na_values checkpoint * Pandas nulls checkpoint * Pandas nulls checkpoint * Minor cleanup * Fix bad pandas import * Make optional import failures less scary * init * fix * fix * fix * added support for version 19.x * fixes playtest * fixes Host header when server_host_name set * fixes tests for 19.x server version code refactoring * revert back playtest * removed noqa comments --------- Co-authored-by: Geoff Genz <geoff@clickhouse.com> Co-authored-by: guykohen <guy@clickhouse.com> Co-authored-by: guykoh <104740726+guykoh@users.noreply.github.com> Co-authored-by: Michael Harris <mharris@luabase.com> Co-authored-by: Artem Popov <artpop92@gmail.com> Co-authored-by: Serge Klochkov <3175289+slvrtrn@users.noreply.github.com> Co-authored-by: Chun-Sheng, Li <peter279k@gmail.com> Co-authored-by: EugeneTorap <evgenykrutpro@gmail.com> Co-authored-by: Mikhail Shustov <restrry@gmail.com> Co-authored-by: Ivan Blinkov <ivan@clickhouse.com> Co-authored-by: John McCann Cunniff Jr <johncunniff1248@gmail.com> Co-authored-by: Stas <reijnnn@gmail.com> Co-authored-by: Aleksey Astafiev <a.astafyev@crazypanda.ru>
genzgd
added a commit
that referenced
this pull request
Jan 18, 2024
* Additional table engine definitions * Add directory, fix test * More distribution updates * Add passthrough of connection options, clean up TLS settigns * Replaced all dict typing with Dict * Replaced all dict typing with Dict * Fix SqlAlchemy table exists method * Workflows (#5) * Add first test workflow * Fix cache typo * Try to fix python setup * Fix another typo * Fix cache * requirements.txt? * Hack superset directories * Hack superset dependencies * Fix another typo * Fix Jinja dependency * More mucking with test dependencies * Build cython extensions in tests * Attempt to fix cython build * Hack for old Python IPv6Addresses * Exclude Python 3.7 from tests because Superset * Don't test Date32 (not supported in CH 21.8) * Make tests work with ClickHouse 21.8 * Exclude Python 3.10 from tests temporarily * Skip multiple test runs on pull request * Add workflow for publish releases to PyPI (#6) * Cleaning up directories and improving superset testing infrastructure. * Install dependencies for pylint * Fix typo * Testing of release workflow * Fix typo * Testing release 1 * Testing release 2 * Testing release 3 * Testing release 4 * Testing release 5 * Testing release 6 * Testing windows * Testing windows 3 * Testing windows 4 * Testing windows 5 * Test with artifacts * Test with artifacts 2 * Test with artifacts 3 * Test with artifacts 4 * Test with artifacts 6 * Test with artifacts 7 * Test with artifacts 8 * Test with pypi 1 * Final test * Final test 2 * Final test 3 * Publish Fixes for PyPI (#7) * Test new release publish * Test new release publish 2 * Test new release publish 3 * Test new release publish 4 * Test new release publish 5 * Test new release publish 6 * Test new release publish 7 * Test new release publish 8 * Test new release publish 8 * Test new release publish 9 * Test new release publish 10 * Test new release publish 11 * Test new release publish 12 * Test new release publish 12 * Test new release publish 14 * Test new release publish 15 * Fix Release Issues on PyPI * Add code to handle LIMIT 0 queries to return names and types (DBT compatibility) * Add `metadata_only` to query settings to retrieve only column names * Incorporate PR review comments * Stop accidentially using root logger * Add wheel for lint workflow * Try to make pylint happy * Add pylint extension * Stop removing newlines from SQL queries, pin Pylint version * One more pylint tweak * Add SQL comment tests * First work on storing nested elements * Fix some deprecation, including implicit coercion to float/int for array.array inserts * Lint fix * Refactor session params, promote to beta (#12) * Refactor settings validation and passthrough, promote to Beta status * Tiny cleanup * Quiet linter * Fix non-local test code * Include separate POST data with client.command (#14) * One data attempt * Lint fixes * More lint fixes * Send all SQL as POST data for client.command * Add validation utility for entrypoints * Fix lint * Fix timeout and retry issues (#16) * Only retry queries, not inserts or commands, and use progress headers as keep alive for long running queries * Ensure string data is encoded as utf8 * Lint fixes * Add chinese character test * Decode ClickHouse Error Response to UTF-8 (#19) * Return ClickHouse error messages as decoded UTF-8 * Return ClickHouse error messages as decoded UTF-8 * Don't try to send readonly settings (even for send_progress or compression) (#20) * Make sure we don't send read only settings, remove brotli (seems broken), fix ping * Fix Lint * Don't modify keyword arguments * Add socket options for TCP keep alive to HTTP adapter (#21) * Add socket options for TCP keep alive to HTTP adapter * Tight up filtering of query parameters/settings that are not server settings * Fix lint * Arrow and modest Client method refactoring (#22) * Add arrow integration step 1 * Fix lint * Add socket options for TCP keep alive to HTTP adapter * Tight up filtering of query parameters/settings that are not server settings * Fix lint * modest cleanup and refactoring of client methods * swap abstract method parameters * Fix arrow doc strings * Support for Nested columns (#23) * Checkpoint for nested column support * Nested column support basic functionality * Fix lint * More lint fixes * Low card fix (#25) * Fix issues with empty inserts and numeric LowCardinality types * Checkpoint for nested column support * Nested column support basic functionality * Fix lint * More lint fixes * Fix issues with empty inserts and LowCardinality numeric types * Fix lint * Try to fix tests for ClickHouse 21.8 * Try to fix tests for ClickHouse 21.8 take 2 * Try to fix tests for ClickHouse 21.8 take 3 * Try to fix tests for ClickHouse 21.8 take 5 * Add changelog * Update changelog * Include pip installation (#26) * Format control/JSON/Improvements (#28) * Format control checkpoint * Checkpoint on data format handling * Lint fixes * Lint fixes * Format control checkpoint * Fix lint * Fix lint * Format checkpoint * Fix named tuples * Named tuple tweaks and test * Fix lint * JSON support checkpoint * Enhance JSON library selection and usage, use QueryContext for all queries * Fix lint * Formatting fixes plus more format tests * Fix lint * Fix lint * Expand read format tests * Fix lint, remove spaces from python JSON serialization * format doc checkpoint * Lint fix * Fix SQL comment related issues * Fix comment and sqlalchemy issues * Lint fixes * Update changelog * Fix sql comment issues in DBAPI cursor (#29) * Changelog update * Zero limit fix (#32) * Fix sql comment issues in DBAPI cursor * Fix 0 limit issue. * Arrow inserts (#33) * Arrow table inserts, backtick quote all column and table identifiers * Tweak developer version name * Remove pip local install * Update test requirements * Build cython for lint * Add raw insert test and identifier quote tweak * Fix multibyte utf-8 writes (#34) * added wait_end_of_query in valid_transport_settings (#36) `SELECT name, value, changed, description, type, readonly FROM system.settings` do not return `wait_end_of_query` setting, so it added in HttpClient.valid_transport_settings fixes error handling for errors during queries which use `ON CLUSTER` ClickHouse/ClickHouse#15124 * Test updates (#37) * Update test matrix * First pass at adding cloud tests * Minor updates, put timestamp first in test database * Fix DBI cursor, parameterized query fix and updates (#39) * Fix datetime formatting for parameterization * Fix cursor description problem with commands, support simple format query strings * Sqla 1.14 update (#40) * Fix sqlalchemy 1.4 issue, update superset testing version * fix typo, update changelog * Tweak test requirements * Tweak test requirements * Tweak test requirements (#42) * add query_retries client parameter (#43) * Tweak test requirements * Add query_retries parameter * All the wheels (#45) * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update build and release process checkpoint * Update Version * First cibuildwheel test * Fix yaml * Update Version * Limit test builds * Update Version * Test more versions, test pypi upload * Update Version * Add macos builds * Yaml fixes * Update Version * Tweak build versions * Fix yaml * Update Version * Add non-x86 linux builds * Update Version * Test push only * Conditional git push to facilitate testing * More workflow fixes * Test publish all * Test publish all * Test publish all * Test publish all * Test publish all * Update VERSION * Test publish all * Test publish all * Mac OS build test * Update VERSION * Test Mac OS x86 * Test Mac OS x86 * Test Mac OS x86 * Update VERSION * All builds again with Cython * Update VERSION * All builds again with Cython * Update VERSION * Mac only test * Update VERSION * Mac only Python 3.11 test * Update VERSION * Full Python 3.11 test * Update version * Cleanup for improved builds release * Fix release job * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Release job debugging * Remove invalid python 3.11 * Fix validation errors in Superset connection dialog (#46) * Retry keep alive disconnects (#48) * Append unixtime millis to the test db name when using Cloud (#50) * Append unixtime millis to the test db name when using Cloud * Fix create_db database name * Update ignored test branches Co-authored-by: Geoff Genz <geoff@clickhouse.com> * Add Python code highlighting (#52) * Remove rb (#54) * Remove row_binary format * Append unixtime millis to the test db name when using Cloud (#50) * Append unixtime millis to the test db name when using Cloud * Fix create_db database name * Update ignored test branches Co-authored-by: Geoff Genz <geoff@clickhouse.com> * Cleanup for row_binary removal Co-authored-by: Serge Klochkov <3175289+slvrtrn@users.noreply.github.com> * Valid params fix (#55) * Fix various superset issues * Fix lint * Type fixes (#58) * Fix UInt64 type bug, fix Superset big integers * Fix typos * Fix lint * Release 0.3.2 (#60) * Handle queries that are actually commands, simply and clean up tests * Update version * Fix test ordering * Test fixes * Fix command regex for 'SHOW CREATE' (#62) * Remove exists and explain from commands * SQLAlchemy Fixes (#63) * Fix JSON insert for SQLAlchemy, update related tests * Clean up JSON test configuration * Remove UUID from JSON * SQLAlchemy cleanup to provide better 1.4 compatibility * Update changelog * Update README * Update README.md * Update README.md * Expand tests (#64) * Add additional Python versions to tests * Fix typo * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Tweaking workflow * Remove arrow temporarily * Finalize test matrix * Fix classifier * Version 0.3.5 build Python 3.11 wheels * Fix query_context.updated_copy method, add first examples (#66) * Chunked inserts (#67) * Chunked and compressed inserts * Fix chunk writes * Fix lint * One more chunk fix * Avoid copying when building insert chunks * Fix typo * Fix compression bug * Insert context (#70) * Insert context checkpoint * Pandas rework * Handle internal error while streaming inserts * Add query result in column format * Reusable insert context work * Improve InsertContext, add pandas tests * Fix lint * Update changelog * Use CaseInsensitiveDict for sqla_type_map (#71) * Use CaseInsensitiveDict for sqla_type_map * Fix type * Release 0 4 1 (#72) * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Numpy improvement checkpoint * Add select connect_version to allowing checking from Superset SQLLab * Fix some numpy big int and datetime issues * Ensure little endian numpy types * Clean up settings, minor fixes (#78) * Various settings rework * Fix arrow string setting * Change Pandas NaN handling * Allow setting http client HTTPAdapter, include VERSION in distribution (#79) * Send all recognized transport headers regardless if recognized * update Changelog to follow a common pattern (#81) * Fix common settings bug (#82) * Bug fixes (#83) * Fix JSON inserts using ujson library (#85) * File insert, server side bindings (#87) * Add tool functions for uploading files, support dsn for client creation * Add dsn test and get_client_setting method * Fix dsn host * Add support for server side parameter binding * Fix dbapi issue, update test matrix * Fix lint * Fixes to align with doc * Improve QueryContext usability * Cleanup for contexts and raw_query method * Additional tweaks to match updated documentation * Update test matrix, removed unused docs directory * Update year in LICENSE (#88) * Streaming, C improvements, refactoring (#92) * Spiking around streaming reads * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Add .c files * Native streaming checkpoint * Native streaming checkpoint * Native streaming checkpoint * Tiny cleanup * Streaming updates and refactoring * Fix Python 3.7 issues * Fix versioning glitch * Add fixed string optimizations * Update changelog, uncomment C optimizations * Make sure json allowed for tests. * Make sure json allowed for tests (working version) * Address some PR comments * Noting that 0.5.0 is not yet released * Compression updates, performance improvements (#95) * Spiking around streaming reads * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Streaming reads checkpoint * Add .c files * Native streaming checkpoint * Native streaming checkpoint * Native streaming checkpoint * Tiny cleanup * Streaming updates and refactoring * Fix Python 3.7 issues * Fix versioning glitch * Add fixed string optimizations * Update changelog, uncomment C optimizations * Make sure json allowed for tests. * Make sure json allowed for tests (working version) * Address some PR comments * Remove requests layer * Compression updates (in progress) * Checkpoint for urllib migration * Clean up TLS stuff * Fix settings issue with SQLAlchemy connections, compression updates * Fix certifi requirement * Revert sqlalchemy version for tests * Compression reorganization, QueryResult and Cython buffer cleanup * Client construction cleanup, minor tweaks * Clarify "certifi" package use * Clean up compression config, add DSN test * Fix lint * Throw exception on unsupported compression type * Fix source build issue * Fix source build issue * Still trying to fix builds * Fix ping * Fix handling of None database (#97) * Fix QueryResult "with" statement context (#101) * Fix QueryResult with context * Add unit test for QueryResult with statement * Prep for 0.5.3 patch release * add a PR template (#103) * add pr template * Exclude github templates from tests * Exclude github templates from tests take 2 Co-authored-by: Geoff Genz <geoff@clickhouse.com> * v0.5.4 -- Numpy/Pandas improvements, stream refactoring (#105) * Add C conversion for IPv4 native * Tiny cleanup * Add DateTime to c data_conv * First work on specialized numpy columns * numpy rework checkpoint * numpy rework checkpoint * date calculation checkpoint * date calculation checkpoint * Cleanup checkpoint * query context cleanup * numpy checkpoint * Numpy read checkpoint * context rework checkpoint * format rework checkpoint * format rework checkpoint * progress toward pd/np reads * progress toward pd/np reads * stream cleanup checkpoint * Add C optimization for UUID * Minor cleanup, ready for tests * Fix lint * First work on optimizing None/NULL handling in Cython * Tweak retries/connection resets * Context cleanup checkpoint * Fix lint * Finalize for 0.5.4 release * Fix lint * Fix weird ORDER BY in test * Tweak Cython UUID transform * Add product_name, fix Numpy for Superset * Downgrade Numpy Build for Python 3.7 * Clean up numpy build dependencies * Fix numpy typo * Don't build PyPy 3.7 binary wheels * build debugging * Tweak release/test actions * Fix client_name refs * Update test matrix, try to make cloud tests less flaky * One more attempt at flaky cloud tests * Cleanup changelog * Fix numpy loc card (#110) * Fix numpy lowcardinality exception * Bump version * Fix dataframe insert mutation (#113) * Restore proxy support (#115) * update perf_test, tweak pandas test * Return empty df instead of list (#119) * Add query_tz and column_tzs query arguments (#121) * Add query_tz and column_tzs query arguments * Move get_client docstring, add preliminary support for client_protocol_version * Streaming fixes (#124) * zstd problem debugging * zstd problem debugging * Stream error handling cleanup * Optimize null strings, bug fixes (#125) * zstd problem debugging * zstd problem debugging * Stream error handling cleanup * Optimize nullable strings * Respect use_none=False, add tests * Fix stream error extraction * Make stream error extraction a little tighter * Close urllib3 connections on exit * Close urllib3 connections on exit * Better error message for invalid numpy DateTime64 types (#127) * added column_oriented check to the named_results function (#129) * SSH Tunnel example, support for client_protocol_version (#131) * Better error message for invalid numpy DateTime64 types * tweak test for Python 3.7 * Server host name (#134) * Fix DSN database parsing * Add server_host_name * Fix ssh tunnel example * Update issue templates * Update issue templates * Fix insert test * Fix lint * Panda nulls (#137) * Handle pandas timezones * use_na_values checkpoint * use_na_values checkpoint * use_na_values checkpoint * use_na_values checkpoint * Pandas nulls checkpoint * Pandas nulls checkpoint * Minor cleanup * Fix bad pandas import * Make optional import failures less scary * init * fix * fix * fix * added support for version 19.x * fixes playtest * fixes Host header when server_host_name set * fixes tests for 19.x server version code refactoring * revert back playtest * removed noqa comments * passing HttpClient initial parameters to settings added abiliti to disable http timeouts * pylint checks * pylint fixes * added custom user settings to clickhouse test config --------- Co-authored-by: Geoff Genz <geoff@clickhouse.com> Co-authored-by: guykohen <guy@clickhouse.com> Co-authored-by: guykoh <104740726+guykoh@users.noreply.github.com> Co-authored-by: Michael Harris <mharris@luabase.com> Co-authored-by: Artem Popov <artpop92@gmail.com> Co-authored-by: Serge Klochkov <3175289+slvrtrn@users.noreply.github.com> Co-authored-by: Chun-Sheng, Li <peter279k@gmail.com> Co-authored-by: EugeneTorap <evgenykrutpro@gmail.com> Co-authored-by: Mikhail Shustov <restrry@gmail.com> Co-authored-by: Ivan Blinkov <ivan@clickhouse.com> Co-authored-by: John McCann Cunniff Jr <johncunniff1248@gmail.com> Co-authored-by: Stas <reijnnn@gmail.com> Co-authored-by: Aleksey Astafiev <a.astafyev@crazypanda.ru>
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.
No description provided.