mirror of
https://github.com/hanzo-ds/connect.git
synced 2026-10-30 20:31:24 +00:00
+5
9270ce347a
9270ce347a
* 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` https://github.com/ClickHouse/ClickHouse/issues/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 <[email protected]> * 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 <[email protected]> * Cleanup for row_binary removal Co-authored-by: Serge Klochkov <[email protected]> * 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 <[email protected]> * 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 <[email protected]> Co-authored-by: guykohen <[email protected]> Co-authored-by: guykoh <[email protected]> Co-authored-by: Michael Harris <[email protected]> Co-authored-by: Artem Popov <[email protected]> Co-authored-by: Serge Klochkov <[email protected]> Co-authored-by: Chun-Sheng, Li <[email protected]> Co-authored-by: EugeneTorap <[email protected]> Co-authored-by: Mikhail Shustov <[email protected]> Co-authored-by: Ivan Blinkov <[email protected]> Co-authored-by: John McCann Cunniff Jr <[email protected]> Co-authored-by: Stas <[email protected]> Co-authored-by: Aleksey Astafiev <[email protected]>
20 lines
686 B
Python
20 lines
686 B
Python
import clickhouse_connect
|
|
|
|
|
|
def main():
|
|
print(f'\nClickHouse Connect installed version: {clickhouse_connect.version()}')
|
|
client = clickhouse_connect.get_client(host='play.clickhouse.com',
|
|
username='play',
|
|
password='clickhouse',
|
|
port=443)
|
|
print(f'ClickHouse Play current version and timezone: {client.server_version} ({client.server_tz})')
|
|
result = client.query('SHOW DATABASES')
|
|
print('ClickHouse play Databases:')
|
|
for row in result.result_set:
|
|
print(f' {row[0]}')
|
|
client.close()
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|