Conductor is a cutting-edge, open-source GraphQL Gateway, fully compliant with the GraphQL specification and designed to supercharge any API with powerful features and proxy flows.
Home Page: https://the-guild.dev/graphql/gateway
License: MIT License
I think it might be better to address #222 first, and let the config primitive to handle things like compilation and error handling.
requires #4 to be done before we can do such thing
requires #8
Related #16
This plugin can be used to achieve regional load balancing, @dotansimha came up with how the usage might look like:
server:
port: 9000
sources:
- id: my-api
type: graphql
config:
endpoint: https://us.myapi.com/ # this is the default
endpoints:
- path: /graphql
from: my-api
plugins:
- type: upstream-endpoint-match
config:
value: {{ headers["CF-Geo-Location"] }}
set:
- match: US-America
value: https://us.myapi.com/
- match: EU
value: https://eu.myapi.com/We should be compatible by default, I think even when the upstream isn't (we can wrap responses)
When running cargo run ../../test_config/config.yaml while cd crates/conductor, and having my config.yaml as follows:
plugins:
- type: persisted_operations
config:
store:
source: file
path: ./test_config/persisted_operations_store.jsonIt breaks with the error:
thread 'main' panicked at 'Failed to read file:
Os{ code: 2, kind: NotFound, message: "No such file or directory" }', crates/config/src/serde_utils.rs:22:61Because, it was unable to find the ./test_config/persisted_operations_store.json path since it started the relative path from the binary's pwd, instead of the config's parent directory, which should be the expected behavior.
We should add a test that gets an endpoint and runs some sanity checks against it.
This way we can find critical runtime issues (wasm might fail at runtime, actix-web might throw a slient error in case of invalid Data, etc..).
The content you are editing has changed. Please copy your edits and refresh the page.
endpoints:
- path: /graphql
from: countries
plugins:
- type: graphiql
- type: http_getWhen accessing through browser, I get an error about invalid GraphQL extraction, and not GraphiQL 😢
This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.
These updates are currently rate-limited. Click on a checkbox below to force their creation now.
These updates have all been created already. Click a checkbox below to force a retry/rebase of any.
These are blocked by an existing closed PR and will not be recreated unless you click a checkbox below.
async-graphql, async-graphql-actix-web)Cargo.toml
tokio 1.37.0futures 0.3.30serde 1.0.197serde_json 1.0.115tracing 0.1.40http 0.2.12http-body 0.4.6bytes 1.6.0async-trait 0.1.79anyhow 1.0.81reqwest 0.11.27thiserror 1.0.58reqwest-middleware 0.2.5tracing-subscriber 0.3.18base64 0.21.7schemars 0.8.16minitrace 0.6.4bin/cloudflare_worker/Cargo.toml
tracing-web 0.1.3time 0.3.34console_error_panic_hook 0.1.7bin/conductor/Cargo.toml
actix-web 4.5.1futures-util 0.3.30ulid 1.1.2openssl 0.10libs/benches/Cargo.toml
criterion 0.5.1hyper 0.14.28actix-web 4.5.1url 2.5.0libs/common/Cargo.toml
graphql-tools 0.2.5mime 0.3.17url 2.5.0querystring 1.1.0once_cell 1.19.0lazy_static 1.4.0libs/config/Cargo.toml
serde_yaml 0.9.33regex 1.10.4humantime-serde 1.1.1http-serde 1.1.3libs/e2e_tests/Cargo.toml
httpmock 0.7.0lazy_static 1.4.0libs/engine/Cargo.toml
ureq 2.9.6humantime 2.1.0libs/federation_query_planner/Cargo.toml
linked-hash-map 0.5.6lazy_static 1.4.0async-graphql 7.0.1insta 1.38.0criterion 0.5.1libs/logger/Cargo.toml
atty 0.2.14tracing-web 0.1.3libs/minitrace_reqwest/Cargo.toml
task-local-extensions 0.1.4libs/napi/Cargo.toml
napi 2.16.1napi-derive 2.16.1actix-web 4.5.1napi-build 2.1.2libs/smoke_tests/Cargo.toml
insta 1.38.0lazy_static 1.4.0serial_test 3.0.0libs/tracing/Cargo.toml
opentelemetry 0.22.0opentelemetry_sdk 0.22.1task-local-extensions 0.1.4rand 0.8.5libs/wasm_polyfills/Cargo.toml
wasm-bindgen-futures 0.4.42send_wrapper 0.6plugins/cors/Cargo.toml
plugins/disable_introspection/Cargo.toml
plugins/graphiql/Cargo.toml
plugins/graphql_validation/Cargo.toml
plugins/http_get/Cargo.toml
urlencoding 2.1.3plugins/jwt_auth/Cargo.toml
jsonwebtoken 9.3.0humantime-serde 1.1.1cookie 0.18.1web-time 1.1.0lazy_static 1.4.0plugins/match_content_type/Cargo.toml
plugins/telemetry/Cargo.toml
humantime-serde 1.1.1opentelemetry 0.22.0opentelemetry_sdk 0.22.1opentelemetry-http 0.11.0rmp-serde 1.1.2web-time 1.1.0opentelemetry-otlp 0.15.0opentelemetry-zipkin 0.20.0opentelemetry-otlp 0.15.0minitrace-datadog 0.6.4opentelemetry-zipkin 0.20.0minitrace-jaeger 0.6.4opentelemetry-http 0.11.0futures 0.3plugins/trusted_documents/Cargo.toml
plugins/vrl/Cargo.toml
tests/test-server/Cargo.toml
async-graphql 7.0.1async-graphql-actix-web 7.0.1actix-web 4.5.1
libs/smoke_tests/docker-compose.yaml
quay.io/keycloak/keycloak 24.0.2jaegertracing/all-in-one 1.55
bin/conductor/docker/Dockerfile
rust 1.77.0debian 12.5
.github/actions/setup/action.yml
Swatinem/rust-cache v2.github/workflows/benchmark.yaml
actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions/upload-artifact v4actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11JarvusInnovations/background-action v1JarvusInnovations/background-action v1ubuntu 22.04ubuntu 22.04.github/workflows/ci.yaml
actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions/setup-node v4actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11JarvusInnovations/background-action v1JarvusInnovations/background-action v1JarvusInnovations/background-action v1actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11JarvusInnovations/background-action v1JarvusInnovations/background-action v1actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions-rs/clippy-check v1actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11thollander/actions-comment-pull-request v2ubuntu 22.04ubuntu 22.04ubuntu 22.04ubuntu 22.04ubuntu 22.04.github/workflows/release.yaml
actions/checkout v4docker/setup-qemu-action v3docker/setup-buildx-action v3docker/login-action v3docker/bake-action v4marocchino/sticky-pull-request-comment v2softprops/action-gh-release v2actions/checkout v4actions/upload-artifact v4montudor/action-zip v1svenstaro/upload-release-action v2actions/checkout v4actions/upload-artifact v4svenstaro/upload-release-action v2actions/checkout v4houseabsolute/actions-rust-cross v0actions/upload-artifact v4svenstaro/upload-release-action v2ubuntu 22.04ubuntu 22.04ubuntu 22.04.github/workflows/website.yaml
actions/checkout v4actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11actions/checkout v4@b4ffde65f46336ab88eb53be808477a3936bae11ubuntu 22.04
bin/cloudflare_worker/package.json
wrangler 3.48.0bin/npm/package.json
@graphql-conductor/lib 0.0.4-alpha.0libs/napi/npm/darwin-arm64/package.json
node >= 10libs/napi/npm/darwin-x64/package.json
node >= 10libs/napi/npm/linux-x64-gnu/package.json
node >= 10libs/napi/package.json
@napi-rs/cli ^2.16.5ava ^5.1.1node >= 10tests/graphql-over-http/package.json
graphql-http 1.22.0node >=20pnpm >=8pnpm 8.15.6website/package.json
@monaco-editor/react 4.6.0@theguild/components 6.4.1@theguild/prettier-config 2.0.6clsx 2.1.0graphql 16.8.1json-to-pretty-yaml 1.2.2next 14.1.4next-sitemap 4.2.3next-themes 0.3.0nextra 3.0.0-alpha.22prettier 3.2.5prettier-plugin-tailwindcss 0.5.13react 18.2.0react-dom 18.2.0react-icons 5.0.1@theguild/tailwind-config 0.3.2@types/json-schema 7.0.15@types/json-to-pretty-yaml 1.2.1@types/node 20.11.30@types/react 18.2.74typescript 5.4.4node >=20pnpm >=8pnpm 8.15.6
Could be useful: https://logankeenan.com/posts/rust-axum-and-cloudflare-workers/
as a follow up to #260
Not sure if it should be under the source, or as a plugin.
Given a schema like this, federation has its own internal parts that they call “specs”
schema
@link(url: "https://specs.apollo.dev/link/v1.0")
@link(url: "https://specs.apollo.dev/join/v0.3", for: EXECUTION) {
query: Query
}we should detect these and inform users if we support it or not, specs can be referenced from http://specs.apollo.dev.
Not sure where this needs to be implemented. As we might want to allow users to easily do mapping of headers, and other request parameters.
I was thinking of a plugin like this maybe?
plugins:
- type: mapping_transform
config:
- from: {request.headers['X-Request-ID']}
to: {upstreamRequest.headers['X-Request-ID']}End-user sends query { something @live { ... }}, we send to the upstream query { something { ... }} and start polling for that specific request.
We can do something like that:
enum ConfigCondition<Context, Alternative> {
Primitive(Alternative),
Condition(Context)
}
struct SomeConfig {
do_something: Option<ConfigCondition<ConditionContext, bool>>
}
In this way we can easily enable configurations like the following:
do_something: true
do_something: false # pass primitive value
do_something: "%request.headers.some-header" == "my-value" # Context will represent the available data on %💡 This page's content and the list of tasks is synced automatically from The Guild's Notion. We are open sourcing our roadmap and tasks because we wish to build tools in the public, and allow developers to take part in the process of shaping the future of this library.
🚀 Feel free to share your thoughts, feedback and ideas with us.
👍 If you wish to show interest and help us prioritize tasks, use the 👍 on the issue.
follow up for #99
Our goal is to allow implementing plugins at the endpoint level. This will allow us to implement caching, persisted operations and so on.
requires #8
similar to what's available in GraphQL-Armor.
In general, we should aim for a panic-free flow during a request execution.
requires #8
The goal is to allow running persisted operations by using a hash, and then forward the original query to the upstream.
The content you are editing has changed. Please copy your edits and refresh the page.
We can add configuration for the source level, not just endpoint.
Some ideas:
And in the future:
Ideally, we want a graphql source to be able to load introspection for the source.
These are the options we need:
(default)
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
or
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
introspection:
from: source
or, with custom headers:
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
introspection:
from: source
headers:
X-Admin-Key: 123
or, with polling:
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
introspection:
from: source
polling_interval: 60
remote:
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
introspection:
from: json
location: https://countries.trevorblades.com/introspection.json
local:
sources:
- id: countries
type: graphql
config:
endpoint: https://countries.trevorblades.com/
introspection:
from: json
location: /file-system/introspection.json
The content you are editing has changed. Please copy your edits and refresh the page.
Ideally, a user would like to test/use Conductor also in development, to make sure everything works correctly.
To do that, we might want to consider to run it easily using things like npx?
Related to: #149
The idea is to allow something like that:
plugins:
- type: graphiql
enabled: ${GRAPHIQL_ENABLED}
config: { ... }requires #8
The goal is to create a new source that allows us to load Supergraph as a source.
This comes with a set of other requirements:
The content you are editing has changed. Please copy your edits and refresh the page.
The content you are editing has changed. Please copy your edits and refresh the page.
I wish to add a top-level cache configuration, so we can later use it for implementing plugins and other useful core things.
For example:
caches:
- type: redis
id: my-cache
config:
connection_string: "..."And then the core of Conductor can use that, or plugins:
plugins:
- type: response_cache
config:
cache: my-cacheThis way we can also assume that conductor cache is a Key->Value, and users can use a cache that matches their env (D1/KV for CF, Redis/something else for other servers).
Related:
Some headers, like X-Request-Id, can be automatically passed through to the upstream.
We should explore what are other plugins that needs passthrough.
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.