corvic-engine 0.3.0rc71__tar.gz → 0.3.0rc72__tar.gz

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/Cargo.lock

@@ -16,7 +16,7 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "corvic-engine"
-version = "0.3.0-rc.71"
+version = "0.3.0-rc.72"
 dependencies = [
  "numpy",
  "pyo3",
@@ -57,7 +57,7 @@ checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719"
 
 [[package]]
 name = "engine-walk"
-version = "0.3.0-rc.71"
+version = "0.3.0-rc.72"
 dependencies = [
  "version_check",
 ]

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/Cargo.toml

@@ -3,7 +3,7 @@ resolver = "2"
 members = ["python"]
 
 [workspace.package]
-version = "0.3.0-rc.71"
+version = "0.3.0-rc.72"
 edition = "2021"
 publish = false
 

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: corvic-engine
-Version: 0.3.0rc71
+Version: 0.3.0rc72
 Classifier: Environment :: Console
 Classifier: License :: Other/Proprietary License
 Classifier: Programming Language :: Python :: Implementation :: CPython

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/pyproject.toml

@@ -18,7 +18,7 @@ authors = [
 ]
 readme = "README.md"
 description = "Seamless embedding generation and retrieval."
-version = "0.3.0-rc.71" # sync this with tool.poetry.version below
+version = "0.3.0-rc.72" # sync this with tool.poetry.version below
 
 # TODO(ddn): Pull dependencies from tool.poetry.dependencies. We use maturin as
 # the build system but poetry for the development environment.
@@ -72,7 +72,7 @@ packages = [
 ]
 description = ""
 authors = []
-version = "0.3.0-rc.71"
+version = "0.3.0-rc.72"
 
 [tool.poetry.scripts]
 check = "corvic_check.cli:main"

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/python/corvic/model/_pipeline.py

@@ -293,6 +293,33 @@ class OcrPdfsPipeline(Pipeline):
                             op_graph.feature_type.categorical(),
                         ),
                         NewColumn("title", pl.String(), op_graph.feature_type.text()),
+                        NewColumn(
+                            "resource_id",
+                            pl.String(),
+                            op_graph.feature_type.identifier(),
+                        ),
+                        NewColumn(
+                            "page_number",
+                            pl.Int64(),
+                            op_graph.feature_type.numerical(),
+                        ),
+                        NewColumn(
+                            "bbox",
+                            pl.Struct(
+                                [
+                                    pl.Field("x1", pl.Float64()),
+                                    pl.Field("y1", pl.Float64()),
+                                    pl.Field("x2", pl.Float64()),
+                                    pl.Field("y2", pl.Float64()),
+                                ]
+                            ),
+                            op_graph.feature_type.embedding(),
+                        ),
+                        NewColumn(
+                            "created_at",
+                            pl.Datetime(),
+                            op_graph.feature_type.timestamp(),
+                        ),
                     ],
                 )
             )
@@ -360,6 +387,33 @@ class OcrPdfsPipeline(Pipeline):
                             pl.String(),
                             op_graph.feature_type.foreign_key(text_source.id),
                         ),
+                        NewColumn(
+                            "resource_id",
+                            pl.String(),
+                            op_graph.feature_type.identifier(),
+                        ),
+                        NewColumn(
+                            "page_number",
+                            pl.Int64(),
+                            op_graph.feature_type.numerical(),
+                        ),
+                        NewColumn(
+                            "bbox",
+                            pl.Struct(
+                                [
+                                    pl.Field("x1", pl.Float64()),
+                                    pl.Field("y1", pl.Float64()),
+                                    pl.Field("x2", pl.Float64()),
+                                    pl.Field("y2", pl.Float64()),
+                                ]
+                            ),
+                            op_graph.feature_type.embedding(),
+                        ),
+                        NewColumn(
+                            "created_at",
+                            pl.Datetime(),
+                            op_graph.feature_type.timestamp(),
+                        ),
                     ],
                 )
             )

{corvic_engine-0.3.0rc71 → corvic_engine-0.3.0rc72}/python/corvic/model/_proto_orm_convert.py

@@ -432,12 +432,14 @@ def feature_view_proto_to_orm(
             return err
     session.flush()
 
-    if not orm_obj.id:
+    if not orm_obj.id or not orm_obj.room_id:
         raise InternalError("internal assertion did not hold")
 
     new_fv_sources = list[orm.FeatureViewSource]()
     for fvs in proto_obj.feature_view_sources:
-        match _feature_view_source_proto_to_orm(fvs, orm_obj.id, session):
+        match _feature_view_source_proto_to_orm(
+            fvs, orm_obj.room_id, orm_obj.id, session
+        ):
             case orm.InvalidORMIdentifierError() | InvalidArgumentError() as err:
                 return err
             case Ok(fvs_orm):
@@ -481,6 +483,7 @@ def feature_view_proto_to_orm(
 
 def _feature_view_source_proto_to_orm(
     proto_obj: models_pb2.FeatureViewSource,
+    room_id: orm.RoomID,
     feature_view_id: orm.FeatureViewID,
     session: sa_orm.Session,
 ) -> Ok[orm.FeatureViewSource] | orm.InvalidORMIdentifierError | InvalidArgumentError:
@@ -491,6 +494,7 @@ def _feature_view_source_proto_to_orm(
             pass
 
     orm_obj = orm.FeatureViewSource(
+        room_id=room_id,
         table_op_graph=proto_obj.table_op_graph,
         drop_disconnected=proto_obj.drop_disconnected,
         source_id=source_id,