Fix/update cropus name embeddings model lang

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "welearn-database"
-version = "1.4.0"
+version = "1.4.2"
 description = "All stuff related to relationnal database from the WeLearn project"
 authors = [
     {name = "Théo",email = "theo.nardin@cri-paris.org"}

welearn_database/alembic/versions/b049924f7067_modify_corpus_name_embedding_model_lang_.py

@@ -0,0 +1,93 @@
+"""modify corpus_name_embedding_model_lang view
+
+Revision ID: b049924f7067
+Revises: f8602200fa99
+Create Date: 2026-03-31 16:09:12.085443
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "b049924f7067"
+down_revision: Union[str, None] = "f8602200fa99"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.execute(
+        """
+        DROP MATERIALIZED VIEW corpus_related.corpus_name_embedding_model_lang;
+        """
+    )
+    op.execute(
+        """
+
+CREATE MATERIALIZED VIEW corpus_related.corpus_name_embedding_model_lang
+TABLESPACE pg_default
+AS WITH ranked AS (
+SELECT
+    c.source_name,
+    cem.corpus_id,
+    cem.embedding_model_id,
+    em.title,
+    em.lang,
+    cem.used_since,
+    c.category_id,
+    ROW_NUMBER() OVER (
+      PARTITION BY cem.corpus_id,
+    em.lang
+ORDER BY
+    cem.used_since DESC
+    ) AS rn
+FROM
+    corpus_related.corpus_embedding_model cem
+JOIN corpus_related.corpus c ON
+    c.id = cem.corpus_id
+JOIN corpus_related.embedding_model em ON
+    em.id = cem.embedding_model_id
+WHERE
+    c.is_active
+)
+    SELECT
+        source_name,
+        corpus_id,
+        embedding_model_id,
+        title,
+        lang,
+        used_since,
+        category_id
+    FROM
+        ranked
+    WHERE
+        rn = 1
+
+WITH DATA;
+    """
+    )
+
+
+def downgrade() -> None:
+    op.execute(
+        """
+        DROP MATERIALIZED VIEW corpus_related.corpus_name_embedding_model_lang;
+        """
+    )
+    op.execute(
+        """
+        CREATE MATERIALIZED VIEW corpus_related.corpus_name_embedding_model_lang
+        TABLESPACE pg_default
+        AS SELECT corpus.source_name,
+            embedding_model.title,
+            embedding_model.lang
+           FROM corpus_related.corpus
+             JOIN corpus_related.corpus_embedding_model ON corpus_embedding_model.corpus_id = corpus.id
+             JOIN corpus_related.embedding_model ON embedding_model.id = corpus_embedding_model.embedding_model_id
+          WHERE corpus.is_active
+        WITH DATA;
+        """
+    )

welearn_database/data/models/corpus_related.py

@@ -107,8 +107,12 @@ class CorpusNameEmbeddingModelLang(Base):
     __table_args__ = {"schema": schema_name}
     __read_only__ = True
     source_name: Mapped[str] = mapped_column(primary_key=True)
+    corpus_id: Mapped[UUID]
+    embedding_model_id: Mapped[UUID]
     title: Mapped[str]
     lang: Mapped[str]
+    used_since: Mapped[datetime]
+    category_id: Mapped[UUID]
 
 
 class CorpusEmbeddingModel(Base):