Make archive.lookup_missing_hashes output bytes

All things equal, I think the bytes representation is better, and in
this case it works well for existing callers too.

swh/web/api/views/identifiers.py

@@ -110,12 +110,11 @@ def api_swhid_known(request):
     swhids_by_type = group_swhids(swhids)
     # search for hashes not present in the storage
     missing_hashes = {
-        k: set(map(hash_to_bytes, archive.lookup_missing_hashes({k: v})))
-        for k, v in swhids_by_type.items()
+        k: set(archive.lookup_missing_hashes({k: v})) for k, v in swhids_by_type.items()
     }
 
     for swhid in swhids:
-        if swhid.object_id not in missing_hashes[swhid.object_type]:
+        if hash_to_bytes(swhid.object_id) not in missing_hashes[swhid.object_type]:
             response[str(swhid)]["known"] = True
 
     return response

swh/web/common/archive.py

@@ -1406,7 +1406,7 @@ def lookup_object(object_type: ObjectType, object_id: str) -> Dict[str, Any]:
         raise ValueError(f"Unexpected object type variant: {object_type}")
 
 
-def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]:
+def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[bytes]:
     """Lookup missing Software Heritage persistent identifier hash, using
     batch processing.
 
@@ -1415,7 +1415,7 @@ def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]:
         keys: object types
         values: object hashes
     Returns:
-        A set(hexadecimal) of the hashes not found in the storage
+        A set(bytes) of the hashes not found in the storage
     """
     missing_hashes = []
 
@@ -1431,9 +1431,7 @@ def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]:
         elif obj_type == ObjectType.SNAPSHOT:
             missing_hashes.append(storage.snapshot_missing(obj_ids))
 
-    missing = set(
-        map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes))
-    )
+    missing = set(itertools.chain(*missing_hashes))
 
     return missing
 

swh/web/tests/common/test_archive.py

@@ -31,7 +31,7 @@ from swh.web.common import archive
 from swh.web.common.exc import BadInputExc, NotFoundExc
 from swh.web.common.typing import OriginInfo, PagedResult
 from swh.web.tests.conftest import ctags_json_missing, fossology_missing
-from swh.web.tests.data import random_content, random_sha1
+from swh.web.tests.data import random_content, random_sha1, random_sha1_bytes
 from swh.web.tests.strategies import new_origin, new_revision, visit_dates
 
 
@@ -941,18 +941,18 @@ def test_lookup_invalid_objects(invalid_sha1):
 
 
 def test_lookup_missing_hashes_non_present():
-    missing_cnt = random_sha1()
-    missing_dir = random_sha1()
-    missing_rev = random_sha1()
-    missing_rel = random_sha1()
-    missing_snp = random_sha1()
+    missing_cnt = random_sha1_bytes()
+    missing_dir = random_sha1_bytes()
+    missing_rev = random_sha1_bytes()
+    missing_rel = random_sha1_bytes()
+    missing_snp = random_sha1_bytes()
 
     grouped_swhids = {
-        ObjectType.CONTENT: [hash_to_bytes(missing_cnt)],
-        ObjectType.DIRECTORY: [hash_to_bytes(missing_dir)],
-        ObjectType.REVISION: [hash_to_bytes(missing_rev)],
-        ObjectType.RELEASE: [hash_to_bytes(missing_rel)],
-        ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)],
+        ObjectType.CONTENT: [missing_cnt],
+        ObjectType.DIRECTORY: [missing_dir],
+        ObjectType.REVISION: [missing_rev],
+        ObjectType.RELEASE: [missing_rel],
+        ObjectType.SNAPSHOT: [missing_snp],
     }
 
     actual_result = archive.lookup_missing_hashes(grouped_swhids)
@@ -967,16 +967,16 @@ def test_lookup_missing_hashes_non_present():
 
 
 def test_lookup_missing_hashes_some_present(content, directory):
-    missing_rev = random_sha1()
-    missing_rel = random_sha1()
-    missing_snp = random_sha1()
+    missing_rev = random_sha1_bytes()
+    missing_rel = random_sha1_bytes()
+    missing_snp = random_sha1_bytes()
 
     grouped_swhids = {
         ObjectType.CONTENT: [hash_to_bytes(content["sha1_git"])],
         ObjectType.DIRECTORY: [hash_to_bytes(directory)],
-        ObjectType.REVISION: [hash_to_bytes(missing_rev)],
-        ObjectType.RELEASE: [hash_to_bytes(missing_rel)],
-        ObjectType.SNAPSHOT: [hash_to_bytes(missing_snp)],
+        ObjectType.REVISION: [missing_rev],
+        ObjectType.RELEASE: [missing_rel],
+        ObjectType.SNAPSHOT: [missing_snp],
     }
 
     actual_result = archive.lookup_missing_hashes(grouped_swhids)