feat(tap-uk-ees): fetch all historical releases, not just latest

Add get_all_release_ids() to paginate /publications/{slug}/releases and
iterate over every release in get_records(). Add latest_only config flag
(default false) to restore single-release behaviour for daily runs.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

pipeline/plugins/extractors/tap-uk-ees/tap_uk_ees/tap.py

@@ -31,6 +31,24 @@ def get_content_release_id(publication_slug: str) -> str:
     return resp.json()["id"]
 
 
+def get_all_release_ids(publication_slug: str) -> list[str]:
+    """Return all release IDs for a publication, newest first."""
+    url = f"{CONTENT_API_BASE}/publications/{publication_slug}/releases"
+    ids = []
+    page = 1
+    while True:
+        resp = requests.get(url, params={"page": page, "pageSize": 20}, timeout=TIMEOUT)
+        resp.raise_for_status()
+        data = resp.json()
+        for r in data.get("results", []):
+            ids.append(r["id"])
+        paging = data.get("paging", {})
+        if page >= paging.get("totalPages", 1):
+            break
+        page += 1
+    return ids
+
+
 def download_release_zip(release_id: str) -> zipfile.ZipFile:
     """Download all data files for a release as a ZIP."""
     url = f"{CONTENT_API_BASE}/releases/{release_id}/files"
@@ -58,11 +76,19 @@ class EESDatasetStream(Stream):
     def get_records(self, context):
         import pandas as pd
 
-        release_id = get_content_release_id(self._publication_slug)
+        latest_only = self.config.get("latest_only", False)
+        if latest_only:
+            release_ids = [get_content_release_id(self._publication_slug)]
+        else:
+            release_ids = get_all_release_ids(self._publication_slug)
+
         self.logger.info(
-            "Downloading release %s for %s",
-            release_id,
-            self._publication_slug,
+            "Found %d release(s) for %s", len(release_ids), self._publication_slug
+        )
+
+        for release_id in release_ids:
+            self.logger.info(
+                "Downloading release %s for %s", release_id, self._publication_slug
             )
             zf = download_release_zip(release_id)
 
@@ -75,12 +101,13 @@ class EESDatasetStream(Stream):
                     break
 
             if not target:
-            self.logger.error(
-                "File matching '%s' not found in ZIP. Available: %s",
+                self.logger.warning(
+                    "File matching '%s' not found in release %s. Available: %s",
                     self._target_filename,
+                    release_id,
                     [n for n in all_files if n.endswith(".csv")],
                 )
-            return
+                continue
 
             self.logger.info("Reading %s from ZIP", target)
             with zf.open(target) as f:
@@ -103,7 +130,7 @@ class EESDatasetStream(Stream):
             if urn_col in df.columns:
                 df = df[df[urn_col].notna() & (df[urn_col] != "")]
 
-        self.logger.info("Emitting %d school-level rows", len(df))
+            self.logger.info("Emitting %d school-level rows from release %s", len(df), release_id)
 
             for _, row in df.iterrows():
                 record = row.to_dict()
@@ -414,6 +441,12 @@ class TapUKEES(Tap):
 
     config_jsonschema = th.PropertiesList(
         th.Property("base_url", th.StringType, description="EES API base URL"),
+        th.Property(
+            "latest_only",
+            th.BooleanType,
+            description="Only fetch the latest release per publication (default: False — fetches all historical releases)",
+            default=False,
+        ),
     ).to_dict()
 
     def discover_streams(self):