Reproduction:

Create a stitching project, then publish the same schema twice but with a different service name.

yarn workspace @graphql-hive/cli start -- schema:publish --service fooooo /Users/laurinquast/Projects/graphql-hive-2/packages/libraries/cli/examples/single.graphql

yarn workspace @graphql-hive/cli start -- schema:publish --service brrrrrt /Users/laurinquast/Projects/graphql-hive-2/packages/libraries/cli/examples/single.graphql

On the second publish the output you receive is

No changes. Skipping.

Afterward, the service brrrrrt is not showing up on the schema versions in the web app.

It seems like it is impossible to publish the same schema with a different service name and it is simply skipped/not inserted into the database

Related: #147

When being rate-limited we should send a slack message to notify the users.

When an org is selected, and then Hive logo is pressed, it should take the user to the main page of the org, and not to the user's default personal org.

• Give instructions next to errors: when the service URL is missing, show an error and tell how to pass it to the CLI, one example.
• When publishing the schema, show a link to the new version.
• When checking the initial schema, say it’s initial.
• When the token is missing, explain how to create and apply a token.
• #207

Currently, it is only possible to integrate notifications/alerts into Slack by using a webhook proxy. It might make sense to have a first-class integration for Discord.

TypeError: Cannot read properties of undefined (reading 'displayName')

# src/components/v2/activities.tsx (157:18) @ getActivity
  155 | content: (
  156 |   <>
> 157 |     {user.displayName} created {targetLink} target in {projectLink} project
      |          ^
  158 |   </>
  159 | ),
  160 | icon: <PlusIcon className="h-4 w-4" />,

Reproduction:

1. Create an organization or visit an existing one that uses users.displayName in one of its activities
2. Click the "Settings" tab
3. Click the "Overview" tab
4. Error

related: #179

The personal organization seems to be unused in most cases, and people prefer to create orgs for companies/groups.
Maybe we should show the org creation wizard when there are no orgs at all for the user?

@kamilkisiela @B2o5T what do you think?

integration-tests/testkit/flow.ts
  0:0  error  Parsing error: "parserOptions.project" has been set for @typescript-eslint/parser.
The file does not match your project config: integration-tests/testkit/flow.ts.
The file must be included in at least one of the projects provided

Reproduction

1. Touch any of *.spec.ts files in integration-tests directory.
2. Try to git commit the change
3. Error

The next version of Apollo Gateway is Apollo Router with description here: https://www.apollographql.com/docs/router

Need a plugin for integrate between Hive and Apollo Router.

Regards.

We have pre-launch users that got a special offer to remain on Hobby with custom limits. These users also have custom value for their retention. The Subscription page shows the retention of the plan and not the actual.

For example:

I think it's missing in the new design. We used to have Monaco Editor (in diff mode) and a simple view of changes as a list.

All versions of the CLI older than v0.9.1 does not support SchemaPublishMissingServiceError and the resolver of Mutation.schemaPublish detects if a fragment of SchemaPublishMissingServiceError is used in the DocumentNode.

Let's drop this logic in 30d after deploying #139 to prod.

Hello, when running hive schema:publish /path/to/schema I get the following error:

 ❯ hive schema:publish
 ›   ModuleLoadError: [MODULE_NOT_FOUND] require failed to load /path/to/project/node_modules/@graphql-hive/cli/dist/commands/schema/publish.js
 ›   Code: MODULE_NOT_FOUND

 ❯ hive --version
@graphql-hive/cli/0.14.6 darwin-arm64 node-v16.13.0

I have no clue what is happening here... I have installed "@graphql-hive/cli": "^0.14.6" as a dev dependency using Yarn. Ran yarn install and even yarn upgrade to see if that changes anything but no. Also tried version 0.13.0 but that didn't make a difference.

The file mentioned in the error exists on disk and looks complete enough.

It's also not only for schema:publish but for any command, I can execute hive schema en see the help for the available commands but any actual command is throwing a similar error with the exception for hive update and hive help which seem to function correctly.

Any clue what I'm missing here?

When publishing the subgraph (applies to GraphQL API as well) the service url in empty.

$ hive schema:publish --service users --url http://localhost:6969 supergraph.graphql

We're thinking of a read-only demo with public access or a walkthrough video (at first and later on maybe a video per feature) that we could put on the landing page and in docs.

Sentry Issue: GRAPHQL-HIVE-CY

GraphQLError: Resource not found.
  File "/app/node_modules/@hive/server/index.js", line 213137, in addEventId
    return new import_graphql33.GraphQLError(err.message, err.nodes, err.source, err.positions, err.path, void 0, __spreadProps(__spreadValu {snip}
  File "/app/node_modules/@hive/server/index.js", line 213273, in <anonymous>
    return addEventId(err, eventId);
  File "/app/node_modules/@hive/server/index.js", in Array.map
    #!/usr/bin/env node
  File "/app/node_modules/@hive/server/index.js", line 213252, in <anonymous>
    const errors = (_a4 = result.errors) === null || _a4 === void 0 ? void 0 : _a4.map((err) => {
  File "/app/node_modules/@hive/server/index.js", line 169282, in Hub1.withScope
    callback(scope);
...
(5 additional frame(s) were not displayed)

it's there to tell "hey this schema version is valid, the breaking changes are expected"

Hi @charlypoly

I am facing an error while publishing the schema. I have followed all the steps and then I have generated the token as well. Finally, I am running the command "hive schema: publish".mesh/schema.graphql". Basically, I have already completed the GraphQL Mesh. Now I tried "Hive" for the CI/CD pipeline. I am facing the below error. Just update me if I am doing anything wrong.

Error:

vijayakumar@Vijayakumars-MacBook-Pro postgres-geodb % hive schema:publish ".mesh/schema.graphql" --force
(node:35108) ExperimentalWarning: stream/web is an experimental feature. This feature could change at any time
(Use node --trace-warnings ... to show where the warning was created)
✖ Failed to publish schema
› Error: Unexpected error.
› Reference: 6e4f3f1e-b58c-419c-9580-72c10be689fa

At the moment we don't have a notification for that, it's only visible through the UI.
We need to find a way to let the users know about this, since their data is being dropped.

Use toDate(expires_at) in the SummingMergeTree + add it to ORDER BY.
Do it with MODIFY.

Think about:

• running OPTIMIZE to delete expired rows.
• old rows without expires_at column

Wrap pages with <ProjectLayout/> and <TargetLayout/> components

There is no information on the shape of the Webhook payload in the docs: https://docs.graphql-hive.com/features/alerts-notifications#adding-a-channel

In addition to that, it could be helpful to have some kind of preview and "send test event" on the "Adding a Channel Page" and "Available Channels" table.

Only nodes[0] :)

Currently, error masking is only enabled in staging and production environments. It should also be enabled in development mode, so the behavior is as close as possible.

A failed webhook delivery is swallowed by the service and only logged to internal logs.

In addition, there should be a UI and/or email notification for users with the information that the delivery failed and also a reason on why the delivery failed (HTTP timeout, not found, etc.).

The graphql-client-vesion has a typo because of reasons mentioned in #139.

Let's revert the #139 after we close #140.

It should be available for all kind of projects

Try to explain the most popular issues with this structure:

• Symptom
• Diagnosis
• Solution

On the Target page: add setup instructions in the existing placeholder (”Schema”/”Operations” tabs)

• tabs with languages (CLI/Node.js/Ruby/PHP) with CLI first
• if user selected Federation, we should display proper snippet

The error looks fine

Context #187

The app should use strict mode.

Rules to enable

• useUnknownInCatchVariables (#82)
• noImplicitThis (#85)
• noImplicitAny (#94)
• strictPropertyInitialization
• strictFunctionTypes (#182)
• strictBindCallApply (#85)
• strictNullChecks
• alwaysStrict (#85)

• Projects list → placeholder should be informative with a CTA under
• #208
• #232