Improve "TEAMS" Dashboard Tab: Optimize Metrics Retrieval Based on GitHub Enterprise Type

[liguori]

This pull request adds support for distinguishing between "Full GitHub Enterprises" and "Copilot Business Only" enterprises, enabling more accurate team retrieval and metrics for each enterprise type. The change introduces a new enterpriseType option throughout the configuration, API, and model layers, updates documentation, and provides comprehensive tests for the new logic.

Enterprise Type Support and API Logic:

• Added NUXT_PUBLIC_ENTERPRISE_TYPE environment variable and propagated its usage throughout the codebase to distinguish between full and copilot-only enterprise types, affecting how teams and metrics are retrieved. (.env, nuxt.config.ts, [1] [2]
• Updated Options model to include enterpriseType, with serialization, deserialization, and merging logic, and modified API URL construction to support both enterprise types for team and enterprise scopes. (app/model/Options.ts, [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14]

Backend Team Retrieval Logic:

• Enhanced the teams API handler to use GraphQL for enumerating organizations and fetching teams for full enterprises, while maintaining existing logic for copilot-only enterprises and organizations. (server/api/teams.ts, [1] [2] [3] [4]

Documentation Updates:

• Updated deployment and README documentation to explain the new enterprise type configuration, usage, and behavioral differences for team retrieval and metrics. (DEPLOYMENT.md, [1]; README.md, [2]

Testing:

• Added comprehensive tests for Options enterprise type handling, including serialization, merging, API URL logic, and error cases. (tests/enterprise-type.spec.ts, tests/enterprise-type.spec.tsR1-R122)

@karpikpl as discussed this is a first implementation created with the help of Coding Agent and tested by myself in a Codespaces, should start to address #271, please verify it.

NOTE: A new scope is required for the GitHub Token (read:org)

[Copilot]

Pull Request Overview

This PR introduces support for distinguishing between "Full GitHub Enterprises" and "Copilot Business Only" enterprises to optimize team retrieval and metrics access. The implementation adds a new enterpriseType configuration option that affects how teams are fetched - using GraphQL to enumerate organizations for full enterprises versus using enterprise-level APIs for copilot-only enterprises.

Key changes include:

• Added enterpriseType configuration with values full or copilot-only (defaulting to copilot-only for backward compatibility)
• Enhanced team retrieval logic to use GraphQL for full enterprises and existing enterprise APIs for copilot-only
• Updated API URL construction to handle different enterprise types appropriately

Reviewed Changes

Copilot reviewed 7 out of 8 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
tests/enterprise-type.spec.ts	Comprehensive test suite covering enterprise type handling, serialization, and API URL generation
server/api/teams.ts	Enhanced teams API with GraphQL support for full enterprises and organization enumeration
nuxt.config.ts	Added default enterpriseType configuration
app/model/Options.ts	Extended Options class with enterprise type support and API URL logic
README.md	Updated documentation explaining enterprise type differences
DEPLOYMENT.md	Added deployment examples for both enterprise types
.env	Added environment variable configuration for enterprise type

[Copilot]

Direct string interpolation in GraphQL query creates a potential injection vulnerability. The enterprise slug should be properly escaped or parameterized to prevent GraphQL injection attacks.

Suggested change

	query: `query { enterprise(slug: "${options.githubEnt}") { organizations(first: 100) { nodes { login name url } } } }`
	query: `query($slug: String!) { enterprise(slug: $slug) { organizations(first: 100) { nodes { login name url } } } }`,
	variables: { slug: options.githubEnt }

[Copilot]

The hardcoded limit of 100 organizations could be problematic for large enterprises. Consider implementing pagination or making this configurable to handle enterprises with more than 100 organizations.

Suggested change

	const graphqlQuery = {
	query: `query { enterprise(slug: "${options.githubEnt}") { organizations(first: 100) { nodes { login name url } } } }`
	}
	const graphqlResponse = await $fetch<GraphQLResponse>('https://api.github.com/graphql', {
	method: 'POST',
	headers: event.context.headers,
	body: JSON.stringify(graphqlQuery)
	})
	const data = res._data as GitHubTeam[]
	for (const t of data) {
	const name: string = t.name
	const slug: string = t.slug
	const description: string = t.description || ''
	if (name && slug) allTeams.push({ name, slug, description })
	const organizations = graphqlResponse.data.enterprise.organizations.nodes
	logger.info(`Found ${organizations.length} organizations in enterprise`)
	// Paginate through all organizations in the enterprise
	let organizations: GraphQLOrganization[] = [];
	let hasNextPage = true;
	let endCursor: string | null = null;
	while (hasNextPage) {
	const graphqlQuery = {
	query: `query {
	enterprise(slug: "${options.githubEnt}") {
	organizations(first: 100${endCursor ? `, after: "${endCursor}"` : ''}) {
	nodes { login name url }
	pageInfo { hasNextPage endCursor }
	}
	}
	}`
	};
	const graphqlResponse = await $fetch<any>('https://api.github.com/graphql', {
	method: 'POST',
	headers: event.context.headers,
	body: JSON.stringify(graphqlQuery)
	});
	const orgs = graphqlResponse.data.enterprise.organizations.nodes;
	organizations = organizations.concat(orgs);
	const pageInfo = graphqlResponse.data.enterprise.organizations.pageInfo;
	hasNextPage = pageInfo.hasNextPage;
	endCursor = pageInfo.endCursor;
	}
	logger.info(`Found ${organizations.length} organizations in enterprise`);

[Copilot]

The team slug parsing logic uses teamParts.length >= 2 but only extracts the first two parts. If a team name contains ' - ', this could lead to incorrect parsing. Consider using a more robust parsing approach like split(' - ', 2) or joining remaining parts for the team name.

Suggested change

	const teamParts = this.githubTeam.split(' - ');
	if (teamParts.length >= 2) {
	const orgName = teamParts[0];
	const teamName = teamParts[1];
	const separator = ' - ';
	const sepIndex = this.githubTeam.indexOf(separator);
	if (sepIndex > 0) {
	const orgName = this.githubTeam.substring(0, sepIndex);
	const teamName = this.githubTeam.substring(sepIndex + separator.length);