The goal of this repo is to allow DataFusion to resolve queries across remote query engines while pushing down as much compute as possible down.
Check out the examples to get a feel for how it works.
Potential use-cases:
The project is in alpha status. Contributions welcome; land a PR = commit access.
This PR adds support for the following FlightSQL features:
I also attempting refactoring according to DataFusion's suggested practice when needing to call async methods within the execute method of a physical plan. As discussed on the prior PR, it is still painful to obtain the schema even when following this method. I do think it is slightly cleaner than try_unfold and can be cleaned up further if we figure out a better way to obtain the expected schema of the flight stream.
I added a Flight SQL example.
cc @devinjdangelo since you showed interest. I do feel FlightSQLExecutor.execute is somewhat complicated, making me wonder if keeping it async is warranted.
The FlightSqlService is rather basic now. It may be possible to add extension points for things like authentication so it becomes usable as a library.
Federation wasn’t working with the Postgres dialect and ConnectorX was crashing on an example using postgres, so this PR adds fixes for both of these issues.
The dialect parameter will allow future extensibility to other dialects for query generation.
This PR simplifies FlightSQL Execution plans by relying on the VirtualExecutionPlan inferred from the sub-logical plan sliced out of the overall query to infer the schema. This avoids the need for block_on and simplifies the code further.
Additionally, I added a new async trait method get_table_schema so SQLExecutors can implement arbitrary logic to return what the overall table schema of each table is.
commitlint.config.js to fix the PullRequestactiondatafusion-federation cratedatafusion-federation crate to remove defaultsI added a SessionStateProvider to allow setting a different SessionState based on the incoming request. This can be used to resolve queries differently, based on auth headers for example.
This project is super cool already, and I loved looking through the optimizer/analyzer code especially. The sqlite-partial example I think is a fantastic proof of concept! Thank you for your work @backkem!
I had some difficulty getting the project to compile / work with rust analyzer initially. Capitalizing all instances of Cargo.toml fixed that.
We can eliminate the need for tokio::block_on in the VirtualExecutionPlan by making SQLExecutor::execute sync. We can still support async code in those execution plans by using methods like futures::stream::unfold to return an async stream from a sync function. See for example here: https://github.com/devinjdangelo/DataWeb/blob/main/webengine/src/web_source.rs#L308.
I tried for a while to improve the CXExecutor to use new_record_batch_iter and then use futures::stream::iter to return that as an async stream. This fails since the returned iterator is not truly an iterator and further is not thread safe, so in general cannot be used within a stream.
I'll open an issue an think more about how we can support true streaming / batched execution with the connector-x.
hello all
I have test a sql like
SELECT t.TrackId, t.Name AS TrackName, a.Title AS AlbumTitle, ar.Name AS ArtistName FROM Track t JOIN Album a ON t.AlbumId = a.AlbumId JOIN Artist ar ON a.ArtistId = ar.ArtistId WHERE t.TrackId > 375 order by t.TrackId limit 10
I have get logic plan
where t.trackid > Int64 does not pushdown?
How to do this optimization?
Updates the requirements on tonic to permit the latest version.
Sourced from tonic's changelog.
0.11.0 (2024-02-08)
BREAKING CHANGES:
- Removed
NamedServicefrom thetransportmodule, please import it viatonic::server::NamedService.- MSRV bumped to
1.70.Features
- Added
zstdcompression support.- Added connection timeout for
connecto_with_connector_lazy.- Upgrade rustls to
v0.22- Feature gate server implementation for
tonic-reflection.0.10.2 (2023-09-28)
Bug Fixes
0.10.1 (2023-09-21)
Bug Fixes
0.10.0 (2023-09-08)
Bug Fixes
- codegen: Use stream type from codegen mod (#1446) (60d776b)
- examples: Use https scheme when using tls (#1466) (388b177)
- tls: Don't use tls w/
httpscheme (#1454) (95e81f5)Features
- build: Add optional default unimplemented stubs (#1344) (aff1daf)
- core: amortize many ready messages into fewer, larger buffers (#1423) (76eedc1)
- tonic-types: add ability to extract rich error details from
google.rpc.Status(#1430) (5fd635a)- transport: Add
Router::into_router(#1442) (ea06a1b)- transport: Expose TcpConnectInfo fields (#1449) (74b079c)
- web: Add
GrpcWebClientService(#1472) (dc29c17)
... (truncated)
ea8cd3f chore: prepare v0.11.0 release (#1626)209fcbe chore(ci): fix udeps warning errors (#1627)4263f87 feat(tls): Implement AsMut for Certificate (#1621)23106dd chore(tls): Update to rustls 0.22.0 (#1509)2a2809b Update Cargo.toml (#1622)4227468 Update helloworld-tutorial.md - Fix Clippy lints (#1617)4859a73 Add connection timeout to connect_with_connector_lazy (#1619)227b169 web: fix invalid bit header error message (#1618)c30cb78 chore(reflection): Make prost-types dependency optional (#1610)f714f42 Repair rust-analyzer for tonic files (#1604)You can trigger a rebase of this PR by commenting @dependabot rebase.
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebase will rebase this PR@dependabot recreate will recreate this PR, overwriting any edits that have been made to it@dependabot merge will merge this PR after your CI passes on it@dependabot squash and merge will squash and merge this PR after your CI passes on it@dependabot cancel merge will cancel a previously requested merge and block automerging@dependabot reopen will reopen this PR if it is closed@dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency@dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)Updates the requirements on datafusion to permit the latest version.
Sourced from datafusion's changelog.
#!/bin/bash
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
some issues are just documentation
add-sections={"documentation":{"prefix":"Documentation updates:","labels":["documentation"]},"performance":{"prefix":"Performance improvements:","labels":["performance"]}}
uncomment to not show PRs. TBD if we shown them or not.
#pull-requests=false
so that the component is shown associated with the issue
issue-line-labels=sql exclude-labels=development-process,invalid breaking-labels=api change
You can trigger a rebase of this PR by commenting @dependabot rebase.
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebase will rebase this PR@dependabot recreate will recreate this PR, overwriting any edits that have been made to it@dependabot merge will merge this PR after your CI passes on it@dependabot squash and merge will squash and merge this PR after your CI passes on it@dependabot cancel merge will cancel a previously requested merge and block automerging@dependabot reopen will reopen this PR if it is closed@dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency@dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)Updates the requirements on arrow to permit the latest version.
Sourced from arrow's changelog.
50.0.0 (2024-01-08)
Breaking changes:
- Make regexp_match take scalar pattern and flag #5245 [arrow] (viirya)
- Use Vec in ColumnReader (#5177) #5193 [parquet] (tustvold)
- Remove SIMD Feature #5184 [arrow] (tustvold)
- Use Total Ordering for Aggregates and Refactor for Better Auto-Vectorization #5100 [arrow] (jhorstmann)
- Allow the
zipcompute function to operator onScalarvalues viaDatum#5086 [arrow] (Nathan-Fenner)- Improve C Data Interface and Add Integration Testing Entrypoints #5080 [arrow] (pitrou)
- Parquet: read/write f16 for Arrow #5003 [parquet] (Jefffrey)
Implemented enhancements:
- Support get offsets or blocks info from arrow file. #5252 [arrow]
- Make regexp_match take scalar pattern and flag #5246 [arrow]
- Cannot access pen state website on arrow-row #5238 [arrow]
- RecordBatch with_schema's error message is hard to read #5227 [arrow]
- Support cast between StructArray. #5219 [arrow]
- Remove nightly-only simd feature and related code in ArrowNumericType #5185 [arrow]
- Use Vec instead of Slice in ColumnReader #5177 [parquet]
- Request to Memmap Arrow IPC files on disk #5153 [arrow]
- GenericColumnReader::read_records Yields Truncated Records #5150 [parquet]
- Nested Schema Projection #5148 [parquet] [arrow]
- Support specifying
quoteandescapein CsvWriterBuilder#5146 [arrow]- Support casting of Float16 with other numeric types #5138 [arrow]
- Parquet: read parquet metadata with page index in async and with size hints #5129 [parquet]
- Cast from floating/timestamp to timestamp/floating #5122 [arrow]
- Support Casting List To/From LargeList in Cast Kernel #5113 [arrow]
- Expose a path for converting
bytes::Bytesintoarrow_buffer::Bufferwithout copy #5104 [arrow]- API inconsistency of ListBuilder make it hard to use as nested builder #5098 [arrow]
- Parquet: don't truncate min/max statistics for float16 and decimal when writing file #5075 [parquet]
- Parquet: derive boundary order when writing columns #5074 [parquet]
- Support new Arrow PyCapsule Interface for Python FFI #5067 [arrow]
48.0.1arrow patch release #5050 [parquet] [arrow]- Binary columns do not receive truncated statistics #5037 [parquet]
- Re-evaluate Explicit SIMD Aggregations #5032 [arrow]
- Min/Max Kernels Should Use Total Ordering #5031 [arrow]
- Allow
zipcompute kernel to takeScalar/Datum#5011 [arrow]- Add Float16/Half-float logical type to Parquet #4986 [parquet]
- feat: cast (Large)List to FixedSizeList #5081 [arrow] (wjones127)
- Update Parquet Encoding Documentation #5051 [parquet]
Fixed bugs:
... (truncated)
db81108 Prepare arrow 50.0.0 (#5291)ca14f6f Update parquet object_store dependency to 0.9.0 (#5290)e831d5c Update proc-macro2 requirement from =1.0.75 to =1.0.76 (#5289)a61e824 Enable JS tests again (#5287)cb16050 Prepare object_store 0.9.0 (#5224) (#5285)2e86650 Print container logs in CI (#5286)f7101ec Support S3 Express One Zone (#5268)2f5dcdf object_store: full HTTP range support (#5222)cf61bb8 DynamoDB ConditionalPut (#5247)2f383e7 feat(object_store): Azure url signing (#5259)You can trigger a rebase of this PR by commenting @dependabot rebase.
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebase will rebase this PR@dependabot recreate will recreate this PR, overwriting any edits that have been made to it@dependabot merge will merge this PR after your CI passes on it@dependabot squash and merge will squash and merge this PR after your CI passes on it@dependabot cancel merge will cancel a previously requested merge and block automerging@dependabot reopen will reopen this PR if it is closed@dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency@dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)Updates the requirements on arrow-flight to permit the latest version.
Sourced from arrow-flight's changelog.
50.0.0 (2024-01-08)
Breaking changes:
- Make regexp_match take scalar pattern and flag #5245 [arrow] (viirya)
- Use Vec in ColumnReader (#5177) #5193 [parquet] (tustvold)
- Remove SIMD Feature #5184 [arrow] (tustvold)
- Use Total Ordering for Aggregates and Refactor for Better Auto-Vectorization #5100 [arrow] (jhorstmann)
- Allow the
zipcompute function to operator onScalarvalues viaDatum#5086 [arrow] (Nathan-Fenner)- Improve C Data Interface and Add Integration Testing Entrypoints #5080 [arrow] (pitrou)
- Parquet: read/write f16 for Arrow #5003 [parquet] (Jefffrey)
Implemented enhancements:
- Support get offsets or blocks info from arrow file. #5252 [arrow]
- Make regexp_match take scalar pattern and flag #5246 [arrow]
- Cannot access pen state website on arrow-row #5238 [arrow]
- RecordBatch with_schema's error message is hard to read #5227 [arrow]
- Support cast between StructArray. #5219 [arrow]
- Remove nightly-only simd feature and related code in ArrowNumericType #5185 [arrow]
- Use Vec instead of Slice in ColumnReader #5177 [parquet]
- Request to Memmap Arrow IPC files on disk #5153 [arrow]
- GenericColumnReader::read_records Yields Truncated Records #5150 [parquet]
- Nested Schema Projection #5148 [parquet] [arrow]
- Support specifying
quoteandescapein CsvWriterBuilder#5146 [arrow]- Support casting of Float16 with other numeric types #5138 [arrow]
- Parquet: read parquet metadata with page index in async and with size hints #5129 [parquet]
- Cast from floating/timestamp to timestamp/floating #5122 [arrow]
- Support Casting List To/From LargeList in Cast Kernel #5113 [arrow]
- Expose a path for converting
bytes::Bytesintoarrow_buffer::Bufferwithout copy #5104 [arrow]- API inconsistency of ListBuilder make it hard to use as nested builder #5098 [arrow]
- Parquet: don't truncate min/max statistics for float16 and decimal when writing file #5075 [parquet]
- Parquet: derive boundary order when writing columns #5074 [parquet]
- Support new Arrow PyCapsule Interface for Python FFI #5067 [arrow]
48.0.1arrow patch release #5050 [parquet] [arrow]- Binary columns do not receive truncated statistics #5037 [parquet]
- Re-evaluate Explicit SIMD Aggregations #5032 [arrow]
- Min/Max Kernels Should Use Total Ordering #5031 [arrow]
- Allow
zipcompute kernel to takeScalar/Datum#5011 [arrow]- Add Float16/Half-float logical type to Parquet #4986 [parquet]
- feat: cast (Large)List to FixedSizeList #5081 [arrow] (wjones127)
- Update Parquet Encoding Documentation #5051 [parquet]
Fixed bugs:
... (truncated)
db81108 Prepare arrow 50.0.0 (#5291)ca14f6f Update parquet object_store dependency to 0.9.0 (#5290)e831d5c Update proc-macro2 requirement from =1.0.75 to =1.0.76 (#5289)a61e824 Enable JS tests again (#5287)cb16050 Prepare object_store 0.9.0 (#5224) (#5285)2e86650 Print container logs in CI (#5286)f7101ec Support S3 Express One Zone (#5268)2f5dcdf object_store: full HTTP range support (#5222)cf61bb8 DynamoDB ConditionalPut (#5247)2f383e7 feat(object_store): Azure url signing (#5259)You can trigger a rebase of this PR by commenting @dependabot rebase.
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebase will rebase this PR@dependabot recreate will recreate this PR, overwriting any edits that have been made to it@dependabot merge will merge this PR after your CI passes on it@dependabot squash and merge will squash and merge this PR after your CI passes on it@dependabot cancel merge will cancel a previously requested merge and block automerging@dependabot reopen will reopen this PR if it is closed@dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency@dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)Updates the requirements on datafusion-substrait to permit the latest version.
Sourced from datafusion-substrait's changelog.
#!/bin/bash
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
distributed with this work for additional information
regarding copyright ownership. The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
some issues are just documentation
add-sections={"documentation":{"prefix":"Documentation updates:","labels":["documentation"]},"performance":{"prefix":"Performance improvements:","labels":["performance"]}}
uncomment to not show PRs. TBD if we shown them or not.
#pull-requests=false
so that the component is shown associated with the issue
issue-line-labels=sql exclude-labels=development-process,invalid breaking-labels=api change
You can trigger a rebase of this PR by commenting @dependabot rebase.
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebase will rebase this PR@dependabot recreate will recreate this PR, overwriting any edits that have been made to it@dependabot merge will merge this PR after your CI passes on it@dependabot squash and merge will squash and merge this PR after your CI passes on it@dependabot cancel merge will cancel a previously requested merge and block automerging@dependabot reopen will reopen this PR if it is closed@dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually@dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency@dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)Ballista: Distributed Compute Platform Overview Ballista is a distributed compute platform primarily implemented in Rust, powered by Apache Arrow. It
Contents Introduction Installation Usage License Introduction Rust implementation of levensthein distance (https://en.wikipedia.org/wiki/Levenshtein_d
descent Toy library for neural networks in Rust using Vulkan compute shaders. Features Multi-dimensional arrays backed by Vulkan device memory Use Rus
Mars Mars (ma-rs) is an blazingly fast rust machine learning library. Simple and Powerful! ?? ?? Contribution: Feel free to build this project. This i
nannou-sand-simulation Learning nannou, an open-source creative-coding toolkit for Rust, by implementing a visualization for a simulation of sand fall
DataFusion is an extensible query execution framework, written in Rust, that uses Apache Arrow as its in-memory format.
xh is a friendly and fast tool for sending HTTP requests. It reimplements as much as possible of HTTPie's excellent design, with a focus on improved performance
Purpose of this program I've made this web scraper so you can use it to get the holder's amount from BSCscan and it will upload for you in JSON format
log4j_interpreter A Rust library for evaluating log4j substitution queries in order to determine whether or not malicious queries may exist. Limitatio
Grsql is a great tool to allow you set up your remote sqlite database as service and CRUD (create/ read/ update/ delete) it using gRPC. Why Create Thi
box This is an experimental repository to perform a proof of concept replacement of the Apache Spark executor for Arc with Apache DataFusion. This is
CloudLLM CloudLLM is a Rust library designed to seamlessly bridge applications with remote Language Learning Models (LLMs) across various platforms. W
FileQL - File Query Language FileQL is a tool that allow you to run SQL-like query on local files instead of database files using the GitQL SDK. Sampl
ClangQL - Clang AST Query Language ClangQL is a tool that allow you to run SQL-like query on C/C++ Code instead of database files using the GitQL SDK.
qlcache An object-relational in-memory cache, supports queries with an SQL-like query language. Warning This is a rather low-level library, and only p
typed-qb: a compile-time typed "query builder" typed-qb is a compile-time, typed, query builder. The goal of this crate is to explore the gap between
This Rust library uses the actual PostgreSQL server source to parse SQL queries and return the internal PostgreSQL parse tree.
joker_query The joker_query is most sugared query builder of Rust, with joker_query can implement most complex queries with sugar syntax Features − (O
ES Resolve JavaScript/TypeScript module resolution in Rust Installation cargo add es_resolve Get Started use std::path::{Path, PathBuf}; use es_resolv
cargo-config2 Load and resolve Cargo configuration. This library is intended to accurately emulate the actual behavior of Cargo configuration, for exa
2.3k Jan 3, 2023
2 Feb 21, 2022
71 Dec 16, 2022
3 Dec 25, 2022
3 Dec 20, 2022
2.9k Jan 2, 2023
3.4k Jan 6, 2023
3 Jul 7, 2022
22 Nov 7, 2022
33 Dec 16, 2022
1 Nov 26, 2021
4 Oct 13, 2023
39 Mar 12, 2024
3 Nov 14, 2021
3 Jan 22, 2022
37 Dec 18, 2022
8 Dec 13, 2023
2 Oct 12, 2022
6 Jan 10, 2023