Mark all pages from pages.jsonl as seeds (#2390)

Fixes #2389 

All pages from `pages/pages.jsonl` files now have `isSeed: True` in the
database, in addition to any pages that explicitly have `seed` set to
true in the actual JSONL.

Tests have been added to ensure that all pages from our fixture uploads
have `isSeed: True`.

---------

Co-authored-by: Ilya Kreymer <ikreymer@gmail.com>

backend/btrixcloud/pages.py

@@ -92,7 +92,7 @@ class PageOps:
                 if not page_dict.get("url"):
                     continue
 
-                if not page_dict.get("isSeed"):
+                if not page_dict.get("isSeed") and not page_dict.get("seed"):
                     page_dict["isSeed"] = False
 
                 if len(pages_buffer) > batch_size:

backend/btrixcloud/storages.py

@@ -607,7 +607,9 @@ class StorageOps:
 
         # pylint: disable=too-many-function-args
         def stream_page_lines(
-            pagefile_zipinfo: ZipInfo, wacz_url: str, wacz_filename: str
+            pagefile_zipinfo: ZipInfo,
+            wacz_url: str,
+            wacz_filename: str,
         ) -> Iterator[Dict[Any, Any]]:
             """Pass lines as json objects"""
             filename = pagefile_zipinfo.filename
@@ -621,6 +623,8 @@ class StorageOps:
             for line in line_iter:
                 page_json = _parse_json(line.decode("utf-8", errors="ignore"))
                 page_json["filename"] = os.path.basename(wacz_filename)
+                if filename == "pages/pages.jsonl":
+                    page_json["seed"] = True
                 yield page_json
 
         page_generators: List[Iterator[Dict[Any, Any]]] = []
@@ -637,7 +641,11 @@ class StorageOps:
                 ]
                 for pagefile_zipinfo in page_files:
                     page_generators.append(
-                        stream_page_lines(pagefile_zipinfo, wacz_url, wacz_file.name)
+                        stream_page_lines(
+                            pagefile_zipinfo,
+                            wacz_url,
+                            wacz_file.name,
+                        )
                     )
 
         return chain.from_iterable(page_generators)

backend/test/test_uploads.py

@@ -254,6 +254,7 @@ def test_get_upload_pages(admin_auth_headers, default_org_id, upload_id):
         assert page["ts"]
         assert page["filename"]
         assert page.get("title") or page.get("title") is None
+        assert page["isSeed"]
 
     page_id = pages[0]["id"]
     r = requests.get(
@@ -270,6 +271,7 @@ def test_get_upload_pages(admin_auth_headers, default_org_id, upload_id):
     assert page["ts"]
     assert page["filename"]
     assert page.get("title") or page.get("title") is None
+    assert page["isSeed"]
 
     assert page["notes"] == []
     assert page.get("userid") is None