nucliadb 6.9.0.post5008__py3-none-any.whl → 6.9.0.post5009__py3-none-any.whl

nucliadb/search/search/hydrator/images.py

@@ -30,12 +30,16 @@ from nucliadb_protos import resources_pb2
 from nucliadb_utils.utilities import get_storage
 
 
-async def paragraph_source_image(kbid: str, paragraph: resources_pb2.Paragraph) -> Optional[Image]:
+async def paragraph_source_image(
+    kbid: str, paragraph_id: ParagraphId, paragraph: resources_pb2.Paragraph
+) -> Optional[Image]:
     """Certain paragraphs are extracted from images using techniques like OCR or
     inception. If that's the case, return the original image for this paragraph.
 
     """
     source_image = paragraph.representation.reference_file
+    if not source_image:
+        return None
 
     if paragraph.kind not in (
         resources_pb2.Paragraph.TypeParagraph.OCR,
@@ -43,7 +47,7 @@ async def paragraph_source_image(kbid: str, paragraph: resources_pb2.Paragraph)
     ):
         return None
 
-    field_id = ParagraphId.from_string(paragraph.key).field_id
+    field_id = paragraph_id.field_id
 
     # Paragraphs extracted from an image store its original image representation
     # in the reference file. The path is incomplete though, as it's stored in

nucliadb/search/search/hydrator/paragraphs.py

@@ -219,7 +219,9 @@ async def hydrate_paragraph(
                 hydrated.image = hydration_models.HydratedParagraphImage()
 
                 if config.image.source_image:
-                    hydrated.image.source_image = await paragraph_source_image(kbid, paragraph)
+                    hydrated.image.source_image = await paragraph_source_image(
+                        kbid, paragraph_id, paragraph
+                    )
 
             if config.page:
                 if hydrated.page is None:

nucliadb/train/generators/sentence_classifier.py

@@ -116,10 +116,7 @@ async def get_sentences(kbid: str, result: str) -> list[str]:
     if split is not None:
         text = extracted_text.split_text[split]
         for paragraph in field_metadata.split_metadata[split].paragraphs:
-            if paragraph.key == "":
-                key = f"{rid}/{field_type}/{field}/{paragraph.start}-{paragraph.end}"
-            else:
-                key = paragraph.key
+            key = f"{rid}/{field_type}/{field}/{paragraph.start}-{paragraph.end}"
             if key == result:
                 for sentence in paragraph.sentences:
                     splitted_text = text[sentence.start : sentence.end]
@@ -127,10 +124,7 @@ async def get_sentences(kbid: str, result: str) -> list[str]:
     else:
         text = extracted_text.text
         for paragraph in field_metadata.metadata.paragraphs:
-            if paragraph.key == "":
-                key = f"{rid}/{field_type}/{field}/{paragraph.start}-{paragraph.end}"
-            else:
-                key = paragraph.key
+            key = f"{rid}/{field_type}/{field}/{paragraph.start}-{paragraph.end}"
             if key == result:
                 for sentence in paragraph.sentences:
                     splitted_text = text[sentence.start : sentence.end]

{nucliadb-6.9.0.post5008.dist-info → nucliadb-6.9.0.post5009.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nucliadb
-Version: 6.9.0.post5008
+Version: 6.9.0.post5009
 Summary: NucliaDB
 Author-email: Nuclia <nucliadb@nuclia.com>
 License-Expression: AGPL-3.0-or-later
@@ -19,11 +19,11 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: <4,>=3.9
 Description-Content-Type: text/markdown
-Requires-Dist: nucliadb-telemetry[all]>=6.9.0.post5008
-Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.9.0.post5008
-Requires-Dist: nucliadb-protos>=6.9.0.post5008
-Requires-Dist: nucliadb-models>=6.9.0.post5008
-Requires-Dist: nidx-protos>=6.9.0.post5008
+Requires-Dist: nucliadb-telemetry[all]>=6.9.0.post5009
+Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.9.0.post5009
+Requires-Dist: nucliadb-protos>=6.9.0.post5009
+Requires-Dist: nucliadb-models>=6.9.0.post5009
+Requires-Dist: nidx-protos>=6.9.0.post5009
 Requires-Dist: nucliadb-admin-assets>=1.0.0.post1224
 Requires-Dist: nuclia-models>=0.50.0
 Requires-Dist: uvicorn[standard]

{nucliadb-6.9.0.post5008.dist-info → nucliadb-6.9.0.post5009.dist-info}/RECORD

@@ -274,8 +274,8 @@ nucliadb/search/search/chat/prompt.py,sha256=Bk69WVki5XIzXFYO6o1uQw1feHtuMwfEx_A
 nucliadb/search/search/chat/query.py,sha256=qWrwVEX_GrDV7LFRyC21BURtl-WsRt8BkIocY8njkKM,17147
 nucliadb/search/search/hydrator/__init__.py,sha256=3Pc-rcax4TI174qcrllnReE728DoJTaA8tpvBUFf98g,7005
 nucliadb/search/search/hydrator/fields.py,sha256=LhKw-aNU5eJqfZADtq3iB7AGXm0l_QabAAoSHJTk8Is,5962
-nucliadb/search/search/hydrator/images.py,sha256=qqjQbn82DgMUrNG21r8CNnpnOK4RnqHtnJ0jdQsvfyU,4550
-nucliadb/search/search/hydrator/paragraphs.py,sha256=yZXsTFAzH_c9Yf8sClLHsr2OwjUxoc2KQtUcV1RT7ug,12616
+nucliadb/search/search/hydrator/images.py,sha256=gS7-dr1e_DpRQ6XaGxwMW1AMEV9a-u73h_jTVy7XY38,4602
+nucliadb/search/search/hydrator/paragraphs.py,sha256=7JjifwhvY7V2TPSnXoIM6aBmh-i6WwTT29D5CvF2Nu0,12676
 nucliadb/search/search/hydrator/resources.py,sha256=1pNyUac8xWRnQVXU6FkDwsXHzmOZwqKRgKhGPCUdEhE,2004
 nucliadb/search/search/query_parser/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJn4bIXMa20ks,835
 nucliadb/search/search/query_parser/exceptions.py,sha256=sVl9gRNzhE-s480LBBVkiXzNRbKhYRQN5F3it5tNNp8,939
@@ -343,7 +343,7 @@ nucliadb/train/generators/image_classifier.py,sha256=46YShcl7nf1_iLXZklWTAFvUNII
 nucliadb/train/generators/paragraph_classifier.py,sha256=Jk3B8a2zkAf-2-59RECiNmNP1Nz7f0-hfnS4rIE69xA,2817
 nucliadb/train/generators/paragraph_streaming.py,sha256=axGNYjOTgxGsOcAAvCz_rTYzZCdZf0f1s-Hzn-VvIX0,3646
 nucliadb/train/generators/question_answer_streaming.py,sha256=YXjWQc_SJ_TyXXJRg8tGkC9IeIzbRDcn0spQw9AlRNY,5740
-nucliadb/train/generators/sentence_classifier.py,sha256=4JBGtInnWUQrM9wBd-P7Z2lTK1Dka6U3PPDqnLbGnFM,5177
+nucliadb/train/generators/sentence_classifier.py,sha256=izCAbjMgfH6isn48lj6z09gwCunouClqct8MuaYPwVc,4989
 nucliadb/train/generators/token_classifier.py,sha256=T8JOVR1vv5g7rn7HtcQcIZ3O5TFqh15uv5rOselLBVo,9617
 nucliadb/train/generators/utils.py,sha256=88cCuHUlOkn9UDP2NR4ru7A-5hbcLGd73sv9j0fMhmY,3590
 nucliadb/writer/__init__.py,sha256=S298mrZL3vr62OrBqi97mdLxgR5cReMlRJgnaQHZV7s,1304
@@ -384,8 +384,8 @@ nucliadb/writer/tus/local.py,sha256=7jYa_w9b-N90jWgN2sQKkNcomqn6JMVBOVeDOVYJHto,
 nucliadb/writer/tus/s3.py,sha256=vu1BGg4VqJ_x2P1u2BxqPKlSfw5orT_a3R-Ln5oPUpU,8483
 nucliadb/writer/tus/storage.py,sha256=ToqwjoYnjI4oIcwzkhha_MPxi-k4Jk3Lt55zRwaC1SM,2903
 nucliadb/writer/tus/utils.py,sha256=MSdVbRsRSZVdkaum69_0wku7X3p5wlZf4nr6E0GMKbw,2556
-nucliadb-6.9.0.post5008.dist-info/METADATA,sha256=lWKVwRZU7H-IiUnwX8V_02JAfoHhQ0Namxoqzqw0MDU,4158
-nucliadb-6.9.0.post5008.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-nucliadb-6.9.0.post5008.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
-nucliadb-6.9.0.post5008.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
-nucliadb-6.9.0.post5008.dist-info/RECORD,,
+nucliadb-6.9.0.post5009.dist-info/METADATA,sha256=hOXZOJfSkjAHZvN5oOM2KVltbSm_QmfnVzcB-jQIFRA,4158
+nucliadb-6.9.0.post5009.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+nucliadb-6.9.0.post5009.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
+nucliadb-6.9.0.post5009.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
+nucliadb-6.9.0.post5009.dist-info/RECORD,,