fix: expand Gerrit channel URLs in health score v2 maintainer and development pipes (IN-1229)

services/libs/tinybird/datasources/repos_channels_ds.datasource

@@ -0,0 +1,18 @@
+DESCRIPTION >
+    - Materialized channel expansion for all repositories. Populated by `repos_channels_copy.pipe`
+    (daily, before health score pipes run).
+    - Each non-Gerrit repo produces one row (channel = url). Each Gerrit repo produces two rows:
+    the plain URL and the /q/project: variant used by Gerrit activity ingestion.
+    - `isGerrit` and `isExcluded` are denormalized from `repositories`/`integrations` so health
+    score pipes can classify repos without re-joining those tables.
+    - Sorting key is `channel` because all consumers join on `channel`; `repoUrl` lookups are
+    full-scans but the table is small enough that this is fine.
+
+SCHEMA >
+    `repoUrl` String,
+    `channel` String,
+    `isGerrit` UInt8,
+    `isExcluded` UInt8
+
+ENGINE MergeTree
+ENGINE_SORTING_KEY channel

services/libs/tinybird/pipes/health_score_v2.pipe

@@ -61,7 +61,12 @@ SQL >
                 ) AS coveredCategoryWeight,
                 l.lifecycleLabelV2 AS lifecycleLabelV2,
                 i.impactScoreRaw AS impactScoreRaw
-            FROM (SELECT DISTINCT url AS repoUrl FROM repositories WHERE deletedAt IS NULL) AS base
+            FROM
+                (
+                    SELECT DISTINCT url AS repoUrl
+                    FROM repositories FINAL
+                    WHERE deletedAt IS NULL AND excluded = false
+                ) AS base
             LEFT JOIN health_score_v2_maintainer_ds AS m ON m.repoUrl = base.repoUrl
             LEFT JOIN health_score_v2_security_ds AS s ON s.repoUrl = base.repoUrl
             LEFT JOIN health_score_v2_development_ds AS d ON d.repoUrl = base.repoUrl

services/libs/tinybird/pipes/health_score_v2_development.pipe

@@ -10,6 +10,9 @@ DESCRIPTION >
     (most repos) having lastCommitAt from GitHub enrichment — the fallback covers the vast majority.
     - Split out of health_score_v2.pipe into its own copy pipe (2026-07-22) — see
     health_score_v2_maintainer.pipe description for why.
+    - All joins against activityRelations, pull_requests_analyzed, and issues_analyzed route through
+    `repos_channels_ds` (populated by repos_channels_copy.pipe), which expands each Gerrit repo URL
+    to both its plain URL and /q/project: channel variant.
 
 NODE health_score_v2_development_calc
 SQL >
@@ -147,7 +150,12 @@ SQL >
                     FROM
                         (
                             SELECT base.url AS repoUrl, rc.lastCommitAt AS lastCommitAt
-                            FROM (SELECT DISTINCT url FROM repositories WHERE deletedAt IS NULL) AS base
+                            FROM
+                                (
+                                    SELECT DISTINCT url
+                                    FROM repositories FINAL
+                                    WHERE deletedAt IS NULL AND archived = false AND excluded = false
+                                ) AS base
                             LEFT JOIN
                                 (
                                     SELECT url, argMax(lastCommitAt, updatedAt) AS lastCommitAt
@@ -192,37 +200,46 @@ SQL >
                     LEFT JOIN
                         (
                             SELECT
-                                channel AS repoUrl,
-                                countIf(timestamp > now() - INTERVAL 6 MONTH) AS commitsLast6m
-                            FROM activityRelations_deduplicated_cleaned_bucket_union
-                            WHERE type = 'authored-commit'
-                            GROUP BY channel
+                                ch.repoUrl AS repoUrl,
+                                countIf(ar.timestamp > now() - INTERVAL 6 MONTH) AS commitsLast6m
+                            FROM activityRelations_deduplicated_cleaned_bucket_union ar
+                            INNER JOIN repos_channels_ds ch ON ar.channel = ch.channel
+                            WHERE ar.type = 'authored-commit'
+                            GROUP BY ch.repoUrl
                         ) AS c
                         ON c.repoUrl = rp.repoUrl
                     LEFT JOIN
                         (
                             SELECT
-                                channel AS repoUrl,
-                                countIf(closedAt IS NOT NULL) AS closed12m,
+                                ch.repoUrl AS repoUrl,
+                                countIf(ia.closedAt IS NOT NULL) AS closed12m,
                                 count() AS opened12m,
                                 quantileIf(0.5)
-                                (closedInSeconds, closedAt > now() - INTERVAL 12 MONTH) AS medianCloseS
-                            FROM issues_analyzed
-                            WHERE openedAt > now() - INTERVAL 12 MONTH
-                            GROUP BY channel
+                                (
+                                    ia.closedInSeconds, ia.closedAt > now() - INTERVAL 12 MONTH
+                                ) AS medianCloseS
+                            FROM issues_analyzed ia
+                            INNER JOIN repos_channels_ds ch ON ia.channel = ch.channel
+                            WHERE ia.openedAt > now() - INTERVAL 12 MONTH
+                            GROUP BY ch.repoUrl
                         ) AS ist
                         ON ist.repoUrl = rp.repoUrl
                     LEFT JOIN
                         (
                             SELECT
-                                channel AS repoUrl,
-                                countIf(mergedAt IS NOT NULL) AS merged12m,
-                                countIf(closedAt IS NOT NULL AND mergedAt IS NULL) AS closedUnmerged12m,
+                                ch.repoUrl AS repoUrl,
+                                countIf(pra.mergedAt IS NOT NULL) AS merged12m,
+                                countIf(
+                                    pra.closedAt IS NOT NULL AND pra.mergedAt IS NULL
+                                ) AS closedUnmerged12m,
                                 quantileIf(0.5)
-                                (mergedInSeconds, mergedAt > now() - INTERVAL 12 MONTH) AS medianMergeS
-                            FROM pull_requests_analyzed
-                            WHERE openedAt > now() - INTERVAL 12 MONTH
-                            GROUP BY channel
+                                (
+                                    pra.mergedInSeconds, pra.mergedAt > now() - INTERVAL 12 MONTH
+                                ) AS medianMergeS
+                            FROM pull_requests_analyzed pra
+                            INNER JOIN repos_channels_ds ch ON pra.channel = ch.channel
+                            WHERE pra.openedAt > now() - INTERVAL 12 MONTH
+                            GROUP BY ch.repoUrl
                         ) AS prs
                         ON prs.repoUrl = rp.repoUrl
                 )

services/libs/tinybird/pipes/health_score_v2_impact.pipe

@@ -11,7 +11,10 @@ DESCRIPTION >
 NODE health_score_v2_impact_calc
 SQL >
     SELECT base.url AS repoUrl, max(toFloat64OrNull(pk.impact)) * 100 AS impactScoreRaw
-    FROM (SELECT DISTINCT url FROM repositories WHERE deletedAt IS NULL) AS base
+    FROM
+        (
+            SELECT DISTINCT url FROM repositories FINAL WHERE deletedAt IS NULL AND excluded = false
+        ) AS base
     INNER JOIN repos r ON r.url = base.url
     INNER JOIN packageRepos pr ON pr.repoId = r.id
     INNER JOIN ossPackages_enriched_ds pk ON pk.id = pr.packageId

services/libs/tinybird/pipes/health_score_v2_lifecycle.pipe

@@ -105,7 +105,12 @@ SQL >
     FROM
         (
             SELECT base.url AS url, base.archived AS archived, rc.lastCommitAt AS lastCommitAt
-            FROM (SELECT DISTINCT url, archived FROM repositories WHERE deletedAt IS NULL) AS base
+            FROM
+                (
+                    SELECT DISTINCT url, archived
+                    FROM repositories FINAL
+                    WHERE deletedAt IS NULL AND excluded = false
+                ) AS base
             LEFT JOIN
                 (
                     SELECT url, argMax(lastCommitAt, updatedAt) AS lastCommitAt FROM repos GROUP BY url
@@ -115,68 +120,76 @@ SQL >
     LEFT JOIN
         (
             SELECT
-                channel AS repoUrl,
-                countIf(timestamp > now() - INTERVAL 6 MONTH) AS commitsLast6m,
+                ch.repoUrl AS repoUrl,
+                countIf(ar.timestamp > now() - INTERVAL 6 MONTH) AS commitsLast6m,
                 countIf(
-                    timestamp <= now() - INTERVAL 6 MONTH AND timestamp > now() - INTERVAL 12 MONTH
+                    ar.timestamp <= now() - INTERVAL 6 MONTH
+                    AND ar.timestamp > now() - INTERVAL 12 MONTH
                 ) AS commitsPrior6m
-            FROM activityRelations_deduplicated_cleaned_bucket_union
-            WHERE type = 'authored-commit'
-            GROUP BY channel
+            FROM activityRelations_deduplicated_cleaned_bucket_union ar
+            INNER JOIN repos_channels_ds ch ON ar.channel = ch.channel
+            WHERE ar.type = 'authored-commit'
+            GROUP BY ch.repoUrl
         ) AS c
         ON c.repoUrl = r.url
     LEFT JOIN
         (
-            SELECT channel AS repoUrl, countIf(closedAt IS NULL) AS issuesOpenNow
-            FROM issues_analyzed
-            GROUP BY channel
+            SELECT ch.repoUrl AS repoUrl, countIf(ia.closedAt IS NULL) AS issuesOpenNow
+            FROM issues_analyzed ia
+            INNER JOIN repos_channels_ds ch ON ia.channel = ch.channel
+            GROUP BY ch.repoUrl
         ) AS n
         ON n.repoUrl = r.url
     LEFT JOIN
         (
             SELECT
-                channel AS repoUrl,
-                countIf(openedAt > now() - INTERVAL 6 MONTH) AS issuesOpenedLast6m,
+                ch.repoUrl AS repoUrl,
+                countIf(ia.openedAt > now() - INTERVAL 6 MONTH) AS issuesOpenedLast6m,
                 countIf(
-                    openedAt <= now() - INTERVAL 6 MONTH AND openedAt > now() - INTERVAL 12 MONTH
+                    ia.openedAt <= now() - INTERVAL 6 MONTH AND ia.openedAt > now() - INTERVAL 12 MONTH
                 ) AS issuesOpenedPrior6m,
-                countIf(openedAt > now() - INTERVAL 18 MONTH) AS issuesInWindow18m,
+                countIf(ia.openedAt > now() - INTERVAL 18 MONTH) AS issuesInWindow18m,
                 countIf(
-                    openedAt > now() - INTERVAL 18 MONTH
-                    AND respondedInSeconds IS NULL
-                    AND closedAt IS NULL
-                    AND openedAt <= now() - INTERVAL 90 DAY
+                    ia.openedAt > now() - INTERVAL 18 MONTH
+                    AND ia.respondedInSeconds IS NULL
+                    AND ia.closedAt IS NULL
+                    AND ia.openedAt <= now() - INTERVAL 90 DAY
                 ) AS unansweredAged90d
-            FROM issues_analyzed
-            GROUP BY channel
+            FROM issues_analyzed ia
+            INNER JOIN repos_channels_ds ch ON ia.channel = ch.channel
+            GROUP BY ch.repoUrl
         ) AS w
         ON w.repoUrl = r.url
     LEFT JOIN
         (
-            SELECT channel AS repoUrl, countIf(openedAt > now() - INTERVAL 18 MONTH) AS prsInWindow18m
-            FROM pull_requests_analyzed
-            GROUP BY channel
+            SELECT
+                ch.repoUrl AS repoUrl,
+                countIf(pra.openedAt > now() - INTERVAL 18 MONTH) AS prsInWindow18m
+            FROM pull_requests_analyzed pra
+            INNER JOIN repos_channels_ds ch ON pra.channel = ch.channel
+            GROUP BY ch.repoUrl
         ) AS p
         ON p.repoUrl = r.url
     LEFT JOIN
         (
             SELECT
-                channel AS repoUrl,
+                ch.repoUrl AS repoUrl,
                 countIf(
-                    openedAt > now() - INTERVAL 18 MONTH
-                    AND reviewedAt IS NULL
-                    AND approvedAt IS NULL
-                    AND closedAt IS NULL
-                    AND openedAt <= now() - INTERVAL 90 DAY
+                    pra.openedAt > now() - INTERVAL 18 MONTH
+                    AND pra.reviewedAt IS NULL
+                    AND pra.approvedAt IS NULL
+                    AND pra.closedAt IS NULL
+                    AND pra.openedAt <= now() - INTERVAL 90 DAY
                 ) AS unansweredAged90d
-            FROM pull_requests_analyzed
-            GROUP BY channel
+            FROM pull_requests_analyzed pra
+            INNER JOIN repos_channels_ds ch ON pra.channel = ch.channel
+            GROUP BY ch.repoUrl
         ) AS pw
         ON pw.repoUrl = r.url
     LEFT JOIN
         (
             SELECT repoUrl, countIf(status = 'OPEN' AND severity = 'CRITICAL') AS openCriticals
-            FROM vulnerabilities
+            FROM vulnerabilities FINAL
             GROUP BY repoUrl
         ) AS v
         ON v.repoUrl = r.url