🎨 Modularize and capitalize (#24)

* 🎨 Modularize and capitalize

* 🍱 Add a migration script

* 💚 Apply migration & fix

* 📝 Fix nb

docs/guide/models.ipynb → docs/guide/tables.ipynb

@@ -5,7 +5,7 @@
    "id": "2ee23aa0-22e1-45ae-8fe4-392702f95de5",
    "metadata": {},
    "source": [
-    "# Models"
+    "# Tables"
    ]
   },
   {
@@ -16,53 +16,20 @@
    "outputs": [],
    "source": [
     "from lnschema_bionty import (\n",
-    "    featureset,\n",
-    "    featureset_gene,\n",
-    "    featureset_protein,\n",
-    "    gene,\n",
-    "    protein,\n",
-    "    species,\n",
+    "    Gene,\n",
+    "    Protein,\n",
+    "    Species,\n",
     ")"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "f62cc02e-c0a9-4f55-b806-ec851659d56d",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "featureset()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "89141bf7-4d8c-4baa-a15d-a76f829097e0",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "featureset_gene()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "86ab44aa-0c5f-4b67-9196-fda8e0b6bb3b",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "featureset_protein()"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
    "id": "06582ada",
    "metadata": {},
    "outputs": [],
    "source": [
-    "gene()"
+    "Gene()"
    ]
   },
   {
@@ -72,7 +39,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "protein()"
+    "Protein()"
    ]
   },
   {
@@ -82,7 +49,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "species()"
+    "Species()"
    ]
   }
  ],

lnschema_bionty/__init__.py

@@ -1,4 +1,4 @@
-"""Schema module for universal biological entities (`zdno`).
+"""Biological entities (`zdno`).
 
 Import the package::
 
@@ -9,58 +9,50 @@
 .. autosummary::
    :toctree: .
 
-   gene
-   protein
-   cell_marker
-   cell_type
-   species
-   tissue
-   disease
-   featureset
+   Gene
+   Protein
+   CellMarker
+   CellType
+   Species
+   Tissue
+   Disease
 
-Link tables:
+Feature sets:
 
 .. autosummary::
    :toctree: .
 
-   featureset_gene
-   featureset_protein
-   featureset_cell_marker
+   Featureset
+   FeaturesetGene
+   FeaturesetProtein
+   FeaturesetCellMarker
 
-Tracking versions & migrations:
+Development tools:
 
 .. autosummary::
    :toctree: .
 
-   version_zdno
-   migration_zdno
-
-Auxiliary modules:
-
-.. autosummary::
-   :toctree: .
-
-   id
+   dev
 
 """
 # This is lnschema-module zdno.
 _schema_id = "zdno"
+_name = "bionty"
 _migration = "267d12e6f6f1"
 __version__ = "0.4.5"
 
-from . import id  # noqa
+from . import dev  # noqa
 from ._core import (  # noqa
-    cell_marker,
-    cell_type,
-    disease,
-    featureset,
-    featureset_cell_marker,
-    featureset_gene,
-    featureset_protein,
-    gene,
-    migration_zdno,
-    protein,
-    species,
-    tissue,
-    version_zdno,
+    CellMarker,
+    CellType,
+    Disease,
+    Featureset,
+    FeaturesetCellMarker,
+    FeaturesetGene,
+    FeaturesetProtein,
+    Gene,
+    Protein,
+    Species,
+    Tissue,
 )
+from .dev import id  # backward compat

lnschema_bionty/_core.py

@@ -1,13 +1,15 @@
-from datetime import datetime as datetime
 from typing import Optional  # noqa
 
-from lnschema_core._timestamps import CreatedAt
-from sqlmodel import Field, SQLModel
+from lnschema_core.dev.sqlmodel import schema_sqlmodel
+from sqlmodel import Field
 
-from . import id as idg
+from . import _name as schema_name
+from .dev import id as idg
 
+SQLModel, prefix, schema_arg = schema_sqlmodel(schema_name)
 
-class species(SQLModel, table=True):  # type: ignore
+
+class Species(SQLModel, table=True):  # type: ignore
     """Species."""
 
     id: str = Field(default_factory=idg.species, primary_key=True)
@@ -17,28 +19,34 @@ class species(SQLModel, table=True):  # type: ignore
     short_name: Optional[str] = None
 
 
-class featureset_gene(SQLModel, table=True):  # type: ignore
+class FeaturesetGene(SQLModel, table=True):  # type: ignore
     """Link table."""
 
-    featureset_id: str = Field(foreign_key="featureset.id", primary_key=True)
-    gene_id: str = Field(foreign_key="gene.id", primary_key=True)
+    __tablename__ = "{prefix}featureset_gene"
+
+    featureset_id: str = Field(foreign_key="bionty.featureset.id", primary_key=True)
+    gene_id: str = Field(foreign_key="bionty.gene.id", primary_key=True)
 
 
-class featureset_protein(SQLModel, table=True):  # type: ignore
+class FeaturesetProtein(SQLModel, table=True):  # type: ignore
     """Link table."""
 
-    featureset_id: str = Field(foreign_key="featureset.id", primary_key=True)
-    protein_id: str = Field(foreign_key="protein.id", primary_key=True)
+    __tablename__ = "{prefix}featureset_protein"
 
+    featureset_id: str = Field(foreign_key="bionty.featureset.id", primary_key=True)
+    protein_id: str = Field(foreign_key="bionty.protein.id", primary_key=True)
 
-class featureset_cell_marker(SQLModel, table=True):  # type: ignore
+
+class FeaturesetCellMarker(SQLModel, table=True):  # type: ignore
     """Link table."""
 
-    featureset_id: str = Field(foreign_key="featureset.id", primary_key=True)
-    cell_marker_id: str = Field(foreign_key="cell_marker.id", primary_key=True)
+    __tablename__ = "{prefix}featureset_cell_marker"
+
+    featureset_id: str = Field(foreign_key="bionty.featureset.id", primary_key=True)
+    cell_marker_id: str = Field(foreign_key="bionty.cell_marker.id", primary_key=True)
 
 
-class featureset(SQLModel, table=True):  # type: ignore
+class Featureset(SQLModel, table=True):  # type: ignore
     """Sets of biological features.
 
     See the corresponding link tables.
@@ -49,7 +57,7 @@ class featureset(SQLModel, table=True):  # type: ignore
     name: str = Field(default=None, unique=True)
 
 
-class gene(SQLModel, table=True):  # type: ignore
+class Gene(SQLModel, table=True):  # type: ignore
     """Genes."""
 
     id: str = Field(default_factory=idg.gene, primary_key=True)
@@ -63,12 +71,12 @@ class gene(SQLModel, table=True):  # type: ignore
     omim_id: Optional[int] = Field(default=None, index=True)
     synonyms: Optional[str] = Field(default=None, index=True)
     species_id: Optional[str] = Field(
-        default=None, foreign_key="species.id", index=True
+        default=None, foreign_key="bionty.species.id", index=True
     )
     version: Optional[str] = None
 
 
-class protein(SQLModel, table=True):  # type: ignore
+class Protein(SQLModel, table=True):  # type: ignore
     """Proteins."""
 
     id: str = Field(default_factory=idg.protein, primary_key=True)
@@ -77,62 +85,48 @@ class protein(SQLModel, table=True):  # type: ignore
     uniprotkb_name: str = Field(default=None, index=True)
     protein_names: Optional[str] = Field(default=None, index=True)
     length: Optional[int] = None
-    species_id: str = Field(default=None, foreign_key="species.id")
+    species_id: str = Field(default=None, foreign_key="bionty.species.id")
     gene_symbols: Optional[str] = None
     gene_synonyms: Optional[str] = None
     ensembl_transcript_ids: Optional[str] = Field(default=None, index=True)
     ncbi_gene_ids: Optional[str] = Field(default=None, index=True)
 
 
-class tissue(SQLModel, table=True):  # type: ignore
+class Tissue(SQLModel, table=True):  # type: ignore
     """Tissues."""
 
     id: str = Field(default_factory=idg.tissue, primary_key=True)
     ontology_id: Optional[str] = Field(default=None, index=True, unique=True)
     name: str = Field(default=None, index=True)
 
 
-class cell_type(SQLModel, table=True):  # type: ignore
+class CellType(SQLModel, table=True):  # type: ignore
     """Cell types."""
 
+    __tablename__ = "{prefix}cell_type"
+
     id: str = Field(default_factory=idg.cell_type, primary_key=True)
     ontology_id: str = Field(default=None, index=True, unique=True)
     name: str = Field(default=None, index=True)
 
 
-class disease(SQLModel, table=True):  # type: ignore
+class Disease(SQLModel, table=True):  # type: ignore
     """Diseases."""
 
     id: str = Field(default_factory=idg.tissue, primary_key=True)
     ontology_id: str = Field(default=None, index=True, unique=True)
     name: str = Field(default=None, index=True)
 
 
-class cell_marker(SQLModel, table=True):  # type: ignore
+class CellMarker(SQLModel, table=True):  # type: ignore
     """Cell markers: protein complexes."""
 
+    __tablename__ = "{prefix}cell_marker"
+
     id: str = Field(default_factory=idg.cell_marker, primary_key=True)
     name: str = Field(default=None, index=True, unique=True)
     gene_symbols: Optional[str] = None  # TODO: link table
     ncbi_gene_ids: Optional[str] = None  # TODO: link table
     protein_names: Optional[str] = None  # TODO: link table
     uniprotkb_ids: Optional[str] = None  # TODO: link table
-    species_id: str = Field(default=None, foreign_key="species.id")
-
-
-class version_zdno(SQLModel, table=True):  # type: ignore
-    """Schema versions."""
-
-    v: str = Field(primary_key=True)
-    migration: Optional[str] = None
-    user_id: str = Field(foreign_key="user.id")
-    created_at: datetime = CreatedAt
-
-
-class migration_zdno(SQLModel, table=True):  # type: ignore
-    """Latest migration.
-
-    This stores the reference to the latest migration script deployed.
-    """
-
-    version_num: Optional[str] = Field(primary_key=True)
+    species_id: str = Field(default=None, foreign_key="bionty.species.id")

lnschema_bionty/dev/__init__.py

@@ -0,0 +1,19 @@
+"""Development tools.
+
+Tracking versions & migrations:
+
+.. autosummary::
+   :toctree: .
+
+   version_zdno
+   migration_zdno
+
+Auxiliary modules:
+
+.. autosummary::
+   :toctree: .
+
+   id
+"""
+from . import id
+from ._versions import migration_zdno, version_zdno

lnschema_bionty/_id.py → lnschema_bionty/dev/_id.py

@@ -1,4 +1,4 @@
-from lnschema_core.id import base62
+from lnschema_core.dev.id import base62
 
 
 def featureset() -> str:

lnschema_bionty/dev/_versions.py

@@ -0,0 +1,24 @@
+from datetime import datetime as datetime
+from typing import Optional
+
+from lnschema_core._timestamps import CreatedAt
+from lnschema_core._users import CreatedBy
+from sqlmodel import Field, SQLModel
+
+
+class version_zdno(SQLModel, table=True):  # type: ignore
+    """Schema versions."""
+
+    v: str = Field(primary_key=True)
+    migration: Optional[str] = None
+    user_id: str = CreatedBy
+    created_at: datetime = CreatedAt
+
+
+class migration_zdno(SQLModel, table=True):  # type: ignore
+    """Latest migration.
+
+    This stores the reference to the latest migration script deployed.
+    """
+
+    version_num: Optional[str] = Field(primary_key=True)