Use previous org repo stats for star repair

Author: ciembor

docs/crawl-jobs.md

@@ -157,7 +157,7 @@ Monthly ranking resume behavior:
 - `--use-stars-diff` uses previous stored repository star observations instead
   of fetching monthly star history again from the source API.
 - `--refresh-organization-stars` repairs organization repository monthly star
-  deltas from the previous stored repository star observations and recomputes
+  deltas from the previous stored monthly repository stats and recomputes
   organization star totals from those repository rows.
 
 Package ranking resume behavior:

lib/polish_open_source_rank/contexts/ranking/application/monthly_source_metric_backfill.rb

@@ -146,12 +146,12 @@ def refresh_organization_repository_stars_for_source(period, source)
             work = BACKFILL_WORK.fetch(:organization_repository_stars)
             return if completed_subject_ids(period, source, work).include?(source.platform)
 
-            log(source, 'refreshing organization repository stars from stored observations')
+            log(source, 'refreshing organization repository stars from previous monthly stats')
             record_work_event(
               period,
               work.merge(platform: source.platform, subject_id: source.platform, subject_label: source.platform)
             ) do
-              store.refresh_organization_repository_star_deltas_from_observations(
+              store.refresh_organization_repository_star_deltas_from_previous_stats(
                 period,
                 platform: source.platform
               )

lib/polish_open_source_rank/contexts/ranking/infrastructure/sqlite/monthly_snapshot_store.rb

@@ -116,8 +116,8 @@ def organization_stats_for_period(period, platform:)
               snapshot_repository.organization_stats_for_period(period, platform)
             end
 
-            def refresh_organization_repository_star_deltas_from_observations(period, platform:)
-              snapshot_repository.refresh_organization_repository_star_deltas_from_observations(
+            def refresh_organization_repository_star_deltas_from_previous_stats(period, platform:)
+              snapshot_repository.refresh_organization_repository_star_deltas_from_previous_stats(
                 period,
                 platform: platform
               )

lib/polish_open_source_rank/contexts/ranking/infrastructure/sqlite/sqlite_snapshot_repository.rb

@@ -11,13 +11,13 @@ class SQLiteSnapshotRepository
               SET monthly_stars_delta = COALESCE((
                     SELECT MAX(
                              organization_repository_monthly_stats.stargazers_count -
-                               previous_observations.stargazers_count,
+                               previous_stats.stargazers_count,
                              0
                            )
-                    FROM organization_repository_star_observations previous_observations
-                    WHERE previous_observations.period_start = ?
-                      AND previous_observations.platform = organization_repository_monthly_stats.platform
-                      AND previous_observations.repository_github_id =
+                    FROM organization_repository_monthly_stats previous_stats
+                    WHERE previous_stats.period_start = ?
+                      AND previous_stats.platform = organization_repository_monthly_stats.platform
+                      AND previous_stats.repository_github_id =
                           organization_repository_monthly_stats.repository_github_id
                   ), 0),
                   updated_at = ?
@@ -177,7 +177,7 @@ def organization_stats_for_period(period, platform)
               SQL
             end
 
-            def refresh_organization_repository_star_deltas_from_observations(period, platform:)
+            def refresh_organization_repository_star_deltas_from_previous_stats(period, platform:)
               database.execute(
                 REFRESH_ORGANIZATION_REPOSITORY_STAR_DELTAS_SQL,
                 [previous_period_start(period), timestamp, period.start_date.to_s, platform]

spec/polish_open_source_rank/contexts/ranking/application/monthly_source_metric_backfill_spec.rb

@@ -151,7 +151,7 @@
     )
   end
 
-  it 'refreshes organization repository stars from stored observations' do
+  it 'refreshes organization repository stars from previous monthly stats' do
     source = instance_double(GitHubSource, platform: 'github', supports_organizations?: true)
     allow(work_events).to receive(:successful_subject_ids).with(
       {
@@ -162,13 +162,13 @@
         platform: 'github'
       }
     ).and_return(Set.new)
-    allow(store).to receive(:refresh_organization_repository_star_deltas_from_observations)
+    allow(store).to receive(:refresh_organization_repository_star_deltas_from_previous_stats)
     allow(store).to receive(:refresh_organization_repository_metrics)
 
     described_class.new(store: store, sources: [source], logger: logger, work_events: work_events)
                    .call(period, refresh_organization_stars: true)
 
-    expect(store).to have_received(:refresh_organization_repository_star_deltas_from_observations)
+    expect(store).to have_received(:refresh_organization_repository_star_deltas_from_previous_stats)
       .with(period, platform: 'github')
     expect(store).to have_received(:refresh_organization_repository_metrics).with(period, platform: 'github')
   end
@@ -184,13 +184,13 @@
         platform: 'github'
       }
     ).and_return(Set['github'])
-    allow(store).to receive(:refresh_organization_repository_star_deltas_from_observations)
+    allow(store).to receive(:refresh_organization_repository_star_deltas_from_previous_stats)
     allow(store).to receive(:refresh_organization_repository_metrics)
 
     described_class.new(store: store, sources: [source], logger: logger, work_events: work_events)
                    .call(period, refresh_organization_stars: true)
 
-    expect(store).not_to have_received(:refresh_organization_repository_star_deltas_from_observations)
+    expect(store).not_to have_received(:refresh_organization_repository_star_deltas_from_previous_stats)
     expect(store).not_to have_received(:refresh_organization_repository_metrics)
   end
 

spec/polish_open_source_rank/contexts/ranking/infrastructure/sqlite/monthly_snapshot_store_spec.rb

@@ -62,7 +62,7 @@ def exercise_monthly_snapshot_store_contract
     allow(snapshot_repository).to receive(:record_organization_snapshot).with(organization_snapshot)
     allow(snapshot_repository).to receive(:record_organization_repository_snapshot)
       .with(organization_repository_snapshot)
-    allow(snapshot_repository).to receive(:refresh_organization_repository_star_deltas_from_observations)
+    allow(snapshot_repository).to receive(:refresh_organization_repository_star_deltas_from_previous_stats)
       .with(period, platform: 'gitlab')
     allow(snapshot_repository).to receive(:refresh_organization_repository_metrics)
       .with(period, platform: 'gitlab')
@@ -100,7 +100,7 @@ def exercise_monthly_snapshot_store_contract
     snapshot_store.record_repository_snapshot(repository_snapshot)
     snapshot_store.record_organization_snapshot(organization_snapshot)
     snapshot_store.record_organization_repository_snapshot(organization_repository_snapshot)
-    snapshot_store.refresh_organization_repository_star_deltas_from_observations(period, platform: 'gitlab')
+    snapshot_store.refresh_organization_repository_star_deltas_from_previous_stats(period, platform: 'gitlab')
     snapshot_store.refresh_organization_repository_metrics(period, platform: 'gitlab')
     snapshot_store.prune_rankings(period)
   end

spec/polish_open_source_rank/contexts/ranking/infrastructure/sqlite/sqlite_snapshot_repository_spec.rb

@@ -126,7 +126,7 @@
     repository.record_organization_repository_snapshot(
       organization_repository_snapshot(source_id: 201, name: 'docs', stars: 10, delta: 0)
     )
-    repository.refresh_organization_repository_star_deltas_from_observations(period, platform: 'github')
+    repository.refresh_organization_repository_star_deltas_from_previous_stats(period, platform: 'github')
     repository.refresh_organization_repository_metrics(period, platform: 'github')
 
     expect(row('organization_monthly_stats')).to include(