@resolution/confluence-api-client NPM

confluence-api-client

Atlassian Confluence API client based on OpenAPI schema.

Compatible with the following request/fetch Atlassian libraries:

Installation

Note that this package requires zod as a peer dependency.

Install using npm

npm add @resolution/confluence-api-client zod

Or using yarn

yarn add @resolution/confluence-api-client zod

Usage

Connect

Inside Connect iframe:

import { ConfluenceApiV1, ConfluenceApiV2 } from '@resolution/confluence-api-client';

async function test() {
  const confluenceApiV1 = new ConfluenceApiV1({ AP: window.AP });
  const currentUser = await confluenceApiV1.users.getCurrentUser();
  console.log("Current User:", currentUser);

  const confluenceApiV2 = new ConfluenceApiV2({ AP: window.AP });
  const spaces = await confluenceApiV2.space.getSpaces({ keys: ["SPACE"] });
  console.log("Space:", spaces.results[0]);
}

test().catch(console.error);

On the server side using atlassian-connect-express:

import atlassianConnectExpress from 'atlassian-connect-express';
import { ConfluenceApiV1, ConfluenceApiV2 } from '@resolution/confluence-api-client';

const app = express();
const ace = atlassianConnectExpress(app);

app.use(ace.authenticate());

app.get('/current-user', async () => {
  const confluenceApiV1 = new ConfluenceApiV1({
    ace,
    clientKey: req.context.clientKey,
    userAccountId: req.context.userAccountId
  });
  const currentUser = await confluenceApiV1.users.getCurrentUser();
  res.status(200).send(currentUser);
});

app.get('/spaces/:spaceKey', async () => {
  const confluenceApiV2 = new ConfluenceApiV2({ ace, clientKey: req.context.clientKey });
  const spaces = await confluenceApiV2.space.getSpaces({ keys: [req.params.spaceKey] });
  res.status(200).send(spaces.results[0]);
});

app.listen(3000);

Forge

Inside Forge iframe:

import { ConfluenceApiV1, ConfluenceApiV2 } from '@resolution/confluence-api-client';
import { requestConfluence } from '@forge/bridge';

async function test() {
  const confluenceApiV1 = new ConfluenceApiV1({ requestConfluence });
  const currentUser = await confluenceApiV1.users.getCurrentUser();
  console.log("Current User:", currentUser);

  const confluenceApiV2 = new ConfluenceApiV2({ requestConfluence });
  const spaces = await confluenceApiV2.space.getSpaces({ keys: ["SPACE"] });
  console.log("Space:", spaces.results[0]);
}

test().catch(console.error);

On the server side using @forge/api:

import { ConfluenceApiV1, ConfluenceApiV2 } from '@resolution/confluence-api-client';
import * as forgeApi from '@forge/api';

export const handler = async ({ payload }) => {
    const confluenceApiV1 = new ConfluenceApiV1({ forgeApi });
    const currentUser = await confluenceApiV1.users.getCurrentUser();

    const confluenceApiV2 = new ConfluenceApiV2({ forgeApi });
    const spaces = await confluenceApiV2.space.getSpaces({ keys: [payload.spaceKey] });

    return { currentUser, space: spaces.results[0] };
}

External fetch on the server side:

import { ConfluenceApiV1, ConfluenceApiV2 } from '@resolution/confluence-api-client';
import { fetch } from "@forge/api";

const baseUrl = "https://your-jira-instance.atlassian.net";

export const handler = async ({ payload }) => {
    const confluenceApiV1 = new ConfluenceApiV1({ fetch, baseUrl });
    const currentUser = await confluenceApiV1.users.getCurrentUser();

    const confluenceApiV2 = new ConfluenceApiV2({ fetch, baseUrl });
    const spaces = await confluenceApiV2.space.getSpaces({ keys: [payload.spaceKey] });

    return { currentUser, space: spaces.results[0] };
}

Impersonalization

When using API on the server side, you can impersonalize the API client. This is useful when you need to access API resources on behalf of a different user or app.

`atlassian-connect-express`

By default, atlassian-connect-express makes requests on behalf of the app. To impersonalize the API client, you need to provide the userAccountId to the API client:

// ...
app.get('/current-user', async () => {
  const confluenceApiV1 = new ConfluenceApiV1({
    ace,
    clientKey: req.context.clientKey,
    userAccountId: req.context.userAccountId
  });
  const currentUser = await confluenceApiV1.users.getCurrentUser();
  res.status(200).send(currentUser);
});
// ...

This can also be achieved by calling asUser method on the API client:

const confluenceApiV1 = new ConfluenceApiV1({
  ace,
  clientKey: req.context.clientKey
});
const currentUser = await confluenceApiV1.asUser(req.context.userAccountId).users.getCurrentUser();

`@forge/api`

By default, @forge/api makes requests on behalf of the current user. To perform requests on behalf of the app, you need to provide the asApp to the API client:

// ...
export const handler = async ({ payload }) => {
  const confluenceApiV2 = new ConfluenceApiV2({ forgeApi, asApp: true });
  const spaces = await confluenceApiV2.space.getSpaces({ keys: [payload.spaceKey] });
  return spaces.results[0];
};

This can also be achieved by calling asApp method on the API client:

const confluenceApiV2 = new ConfluenceApiV2({ forgeApi });
const spaces = await confluenceApiV2.asApp().space.getSpaces({ keys: ["SPACE"] });

Error handling

All API methods return a promise that resolves to the response data or rejects with an ApiError instance.

ApiError class:

export class ApiError extends Error {
  readonly message: string;
  readonly name: "ApiError";
  readonly url: URL;
  readonly request: CommonHttpClientFetchRequest | undefined;
  readonly response: CommonHttpClientFetchResponse | undefined;
  readonly options: CommonHttpClientOptions | undefined;
}

Checking response status example:

try {
  const spaces = await confluenceApiV2.space.getSpaces({ keys: ["SPACE"] });
  console.log("Space:", spaces.results[0]);
} catch (error) {
  if (error instanceof ApiError && error.response?.status === 404) {
    console.log("Space not found.");
  } else {
    console.error(error);
  }
}