Use Repo type for repo field and fully populate

For consistency with other dataclasses declared in github.py,
use the Repo type for a Codespace's repo attribute.
At the time of writing, the organisation codespaces API endpoint
does not provide all the data required for this type, and since
we are retreiving info on codespaces non-Tech team repos/orgs,
extra API calls are required to get the required information
(teams, repos, repo metadata)

INSTALL.md

@@ -25,9 +25,9 @@ Code scanning alerts, Dependabot alerts, Metadata, Pull requests and Repository
 
 The `GITHUB_OS_TOKEN` is a fine-grained GitHub personal access token that is used for authenticating with the GitHub REST API.
 It is assigned to a single organisation and should have the following *read-only* permissions:
-* organisation permissions: codespaces
+* organisation permissions: Organisation codespaces and Members
 * *all repositories* owned by the organisation with the following permissions:
-Codespaces and Metadata
+Codespaces, Metadata, and Repository security advisories
 
 ## Disable checks
 Dokku performs health checks on apps during deploy by sending requests to port 80.

metrics/github/client.py

@@ -55,7 +55,7 @@ def rest_query(self, path, **variables):
             # unlike the PRs/Issues/Commits GitHub API endpoints
             # the codespaces endpoint returns a dict containing a
             # count of codespaces, and a list thereof (what we want)
-            if "codespaces" in path and isinstance(data, dict):
+            elif "codespaces" in path and isinstance(data, dict):
                 yield from data["codespaces"]
             else:
                 raise RuntimeError("Unexpected response format:", data)

metrics/github/github.py

@@ -110,26 +110,36 @@ def from_dict(cls, data, repo):
 
 @dataclass(frozen=True)
 class Codespace:
-    org: str
-    # The Repo type requires fields neither returned by the codespaces
-    # endpoint, nor required for codespaces metrics so str for repo name
-    repo: str
+    repo: Repo
     user: str
     created_at: datetime.datetime
     last_used_at: datetime.datetime
 
     @classmethod
-    def from_dict(cls, **kwargs):
-        return cls(**kwargs)
+    def from_dict(cls, repo, **kwargs):
+        if "repo_name" in kwargs:
+            del kwargs["repo_name"]
+        return cls(repo, **kwargs)
 
 
 def codespaces(org):
+    org_teams = teams(org)
+    ownership = _repo_owners(org, org_teams)
+    repos = {
+        r["name"]: Repo.from_dict(r, org=org, team=ownership.get(r["name"], None))
+        for r in query.repos(org)
+    }
+
     return [
-        Codespace.from_dict(**({"org": org} | codespace))
-        for codespace in query.codespaces(org)
+        Codespace.from_dict(repo=repos[c["repo_name"]], **c)
+        for c in query.codespaces(org)
     ]
 
 
+def teams(org):
+    return [t["name"] for t in query.teams(org)]
+
+
 def tech_prs():
     tech_team_members = _tech_team_members()
     return [
@@ -165,8 +175,8 @@ def _get_repos():
     return repos
 
 
-def _repo_owners(org):
-    return {repo: team for team in _TECH_TEAMS for repo in query.team_repos(org, team)}
+def _repo_owners(org, teams=_TECH_TEAMS):
+    return {repo: team for team in teams for repo in query.team_repos(org, team)}
 
 
 def _tech_team_members():

metrics/github/metrics.py

@@ -81,8 +81,8 @@ def get_codespaces_metrics(codespaces):
     return [
         {
             "created_at": c.created_at,
-            "organisation": c.org,
-            "repo": c.repo,
+            "organisation": c.repo.org,
+            "repo": c.repo.name,
             "user": c.user,
             "last_used_at": c.last_used_at,
         }

metrics/github/query.py

@@ -45,6 +45,27 @@ def team_members(org, team):
         yield member["login"]
 
 
+def teams(org):
+    query = """
+    query teams($cursor:String, $org: String!) {
+      organization(login: $org) {
+        teams(first:100, after: $cursor) {
+          nodes {
+            name
+          }
+          pageInfo{
+            endCursor
+            hasNextPage
+          }
+        }
+      }
+    }
+    """
+    return maybe_truncate(
+        _client().graphql_query(query, path=["organization", "teams"], org=org)
+    )
+
+
 def vulnerabilities(org, repo):
     query = """
     query vulnerabilities($cursor: String, $org: String!, $repo: String!) {
@@ -146,7 +167,7 @@ def codespaces(org):
     for codespace in codespaces:
         yield {
             "user": codespace["owner"]["login"],
-            "repo": codespace["repository"]["name"],
+            "repo_name": codespace["repository"]["name"],
             "created_at": codespace["created_at"],
             "last_used_at": codespace["last_used_at"],
         }

tests/metrics/github/test_github.py

@@ -42,14 +42,36 @@ def test_codespaces(patch):
             "opensafely": [
                 {
                     "user": "testuser",
-                    "repo": "testrepo",
+                    "repo_name": "testrepo",
+                    "created_at": datetime.datetime.now().isoformat(),
+                    "last_used_at": datetime.datetime.now().isoformat(),
+                },
+                {
+                    "user": "testuser",
+                    "repo_name": "testrepo2",
                     "created_at": datetime.datetime.now().isoformat(),
                     "last_used_at": datetime.datetime.now().isoformat(),
                 },
             ]
         },
     )
-    assert len(github.codespaces("opensafely")) == 1
+    patch("teams", {"opensafely": [{"name": "testteam"}]})
+    # Expect that at least one repo will not be assigned to a team
+    patch("team_repos", {"opensafely": {"testteam": ["testrepo"]}})
+    # Expect that there will be repositories without codespaces
+    patch(
+        "repos",
+        {
+            "opensafely": [
+                repo_data("testrepo"),
+                repo_data("testrepo2"),
+                repo_data("testrepo3"),
+            ]
+        },
+    )
+
+    # All codespaces for repos in this org, regardless of team should be counted
+    assert len(github.codespaces("opensafely")) == 2
 
 
 def test_includes_tech_owned_repos(patch):