feat: add license information to the api (#1482)

Author: jcpitre

.github/workflows/typescript-generator-check.yml

@@ -44,29 +44,14 @@ jobs:
         run: yarn generate:api-types:output
         env:
           OUTPUT_PATH_TYPES: src/app/services/feeds/generated/types.ts
-      
-      - name: Generate TypeScript gbfs validator types
-        working-directory: web-app
-        run: yarn generate:gbfs-validator-types:output
-        env:
-          OUTPUT_PATH_TYPES: src/app/services/feeds/generated/gbfs-validator-types.ts
 
       - name: Upload generated types
         uses: actions/upload-artifact@v4
         with:
           name: generated_types.ts
           path: web-app/src/app/services/feeds/generated/types.ts
-      
-      - name: Upload generated gbfs types
-        uses: actions/upload-artifact@v4
-        with:
-          name: generated_types.ts
-          path: web-app/src/app/services/feeds/generated/gbfs-validator-types.ts
 
       - name: Compare TypeScript types with existing types
         working-directory: web-app
         run: diff src/app/services/feeds/generated/types.ts src/app/services/feeds/types.ts || (echo "Types are different!" && exit 1)
 
-      - name: Compare gbfs validator TypeScript types with existing types
-        working-directory: web-app
-        run: diff src/app/services/feeds/generated/gbfs-validator-types.ts src/app/services/feeds/gbfs-validator-types.ts || (echo "Gbfs Validator Types are different!" && exit 1)

.github/workflows/typescript-generator-gbfs-validator-types-check.yml

@@ -0,0 +1,56 @@
+name: Verify TypeScript GBFS Validator Types Generation
+on:
+  pull_request:
+    branches: 
+      - main
+    paths:
+      - "docs/GbfsValidator.yaml"
+
+env:
+  NODE_VERSION: "18"
+
+jobs:
+  generate-and-compare:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: ${{ env.NODE_VERSION }}
+          cache: 'yarn'
+          cache-dependency-path: 'web-app/yarn.lock'
+
+      - name: Cache Yarn dependencies
+        uses: actions/cache@v4
+        id: yarn-cache
+        with:
+          path: |
+            **/node_modules
+            **/.eslintcache
+          key: ${{ runner.os }}-yarn-${{ hashFiles('web-app/yarn.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-yarn-
+
+      - name: Install dependencies
+        working-directory: web-app
+        run: yarn install --frozen-lockfile --prefer-offline
+      
+      - name: Generate TypeScript gbfs validator types
+        working-directory: web-app
+        run: yarn generate:gbfs-validator-types:output
+        env:
+          OUTPUT_PATH_TYPES: src/app/services/feeds/generated/gbfs-validator-types.ts
+      
+      - name: Upload generated gbfs types
+        uses: actions/upload-artifact@v4
+        with:
+          name: generated_types.ts
+          path: web-app/src/app/services/feeds/generated/gbfs-validator-types.ts
+      
+      - name: Compare gbfs validator TypeScript types with existing types
+        working-directory: web-app
+        run: diff src/app/services/feeds/generated/gbfs-validator-types.ts src/app/services/feeds/gbfs-validator-types.ts || (echo "Gbfs Validator Types are different!" && exit 1)

api/.openapi-generator/FILES

@@ -6,6 +6,8 @@ src/feeds_gen/apis/datasets_api.py
 src/feeds_gen/apis/datasets_api_base.py
 src/feeds_gen/apis/feeds_api.py
 src/feeds_gen/apis/feeds_api_base.py
+src/feeds_gen/apis/licenses_api.py
+src/feeds_gen/apis/licenses_api_base.py
 src/feeds_gen/apis/metadata_api.py
 src/feeds_gen/apis/metadata_api_base.py
 src/feeds_gen/apis/search_api.py
@@ -28,6 +30,9 @@ src/feeds_gen/models/gtfs_feed.py
 src/feeds_gen/models/gtfs_rt_feed.py
 src/feeds_gen/models/latest_dataset.py
 src/feeds_gen/models/latest_dataset_validation_report.py
+src/feeds_gen/models/license_base.py
+src/feeds_gen/models/license_rule.py
+src/feeds_gen/models/license_with_rules.py
 src/feeds_gen/models/location.py
 src/feeds_gen/models/metadata.py
 src/feeds_gen/models/redirect.py

api/src/feeds/impl/feeds_api_impl.py

@@ -3,7 +3,7 @@
 
 from sqlalchemy import or_
 from sqlalchemy import select
-from sqlalchemy.orm import joinedload, Session
+from sqlalchemy.orm import joinedload, contains_eager, selectinload, Session
 from sqlalchemy.orm.query import Query
 
 from feeds.impl.datasets_api_impl import DatasetsApiImpl
@@ -72,9 +72,13 @@ def get_feed(self, id: str, db_session: Session) -> Feed:
         is_email_restricted = is_user_email_restricted()
         self.logger.debug(f"User email is restricted: {is_email_restricted}")
 
+        # Use an explicit LEFT OUTER JOIN and contains_eager so the License relationship
+        # is populated from the same SQL result without causing N+1 queries.
         feed = (
             FeedFilter(stable_id=id, provider__ilike=None, producer_url__ilike=None, status=None)
             .filter(Database().get_query_model(db_session, FeedOrm))
+            .outerjoin(FeedOrm.license)
+            .options(contains_eager(FeedOrm.license))
             .filter(
                 or_(
                     FeedOrm.operational_status == "published",
@@ -115,7 +119,8 @@ def get_feeds(
         )
         # Results are sorted by provider
         feed_query = feed_query.order_by(FeedOrm.provider, FeedOrm.stable_id)
-        feed_query = feed_query.options(*get_joinedload_options())
+        # Ensure license relationship is available to the model conversion without extra queries
+        feed_query = feed_query.options(*get_joinedload_options(), selectinload(FeedOrm.license))
         if limit is not None:
             feed_query = feed_query.limit(limit)
         if offset is not None:

api/src/feeds/impl/licenses_api_impl.py

@@ -0,0 +1,44 @@
+from typing import List, Optional
+
+from feeds_gen.apis.licenses_api_base import BaseLicensesApi
+from feeds_gen.models.license_with_rules import LicenseWithRules
+from feeds_gen.models.license_base import LicenseBase
+from shared.database.database import with_db_session
+from shared.database_gen.sqlacodegen_models import License as LicenseOrm
+from feeds.impl.error_handling import raise_http_error
+from shared.db_models.license_with_rules_impl import LicenseWithRulesImpl
+from shared.db_models.license_base_impl import LicenseBaseImpl
+
+
+class LicensesApiImpl(BaseLicensesApi):
+    """
+    Implementation for the Licenses API.
+    """
+
+    @with_db_session
+    def get_license(self, id: str, db_session) -> LicenseWithRules:
+        """Return the license with the provided id."""
+        try:
+            lic: Optional[LicenseOrm] = db_session.query(LicenseOrm).filter(LicenseOrm.id == id).one_or_none()
+            if not lic:
+                raise_http_error(404, f"License '{id}' not found")
+
+            return LicenseWithRulesImpl.from_orm(lic)
+        except Exception as e:
+            # Use raise_http_error to convert into an HTTPException with proper logging
+            raise_http_error(500, f"Error retrieving license: {e}")
+
+    @with_db_session
+    def get_licenses(self, limit: int, offset: int, db_session) -> List[LicenseBase]:
+        """Return a list of licenses (paginated)."""
+        try:
+            query = db_session.query(LicenseOrm).order_by(LicenseOrm.id)
+            if limit is not None:
+                query = query.limit(limit)
+            if offset is not None:
+                query = query.offset(offset)
+            results = query.all()
+
+            return [LicenseBaseImpl.from_orm(lic) for lic in results]
+        except Exception as e:
+            raise_http_error(500, f"Error retrieving licenses: {e}")

api/src/main.py

@@ -24,6 +24,7 @@
 from feeds_gen.apis.feeds_api import router as FeedsApiRouter
 from feeds_gen.apis.metadata_api import router as MetadataApiRouter
 from feeds_gen.apis.search_api import router as SearchApiRouter
+from feeds_gen.apis.licenses_api import router as LicensesApiRouter
 
 # Using the starlettte implementaiton as fastapi implementation generates errors with CORS in certain situations and
 # returns 200 in the method response. More info, https://github.com/tiangolo/fastapi/issues/1663#issuecomment-730362611
@@ -54,6 +55,7 @@
 app.include_router(FeedsApiRouter)
 app.include_router(MetadataApiRouter)
 app.include_router(SearchApiRouter)
+app.include_router(LicensesApiRouter)
 
 
 @app.on_event("startup")

api/src/scripts/populate_db_test_data.py

@@ -12,12 +12,14 @@
     Gtfsfeed,
     Notice,
     Feature,
+    License,
     t_feedsearch,
     Location,
     Officialstatushistory,
     Gbfsversion,
     Gbfsendpoint,
     Gbfsfeed,
+    Rule,
 )
 from scripts.populate_db import set_up_configs, DatabasePopulateHelper
 from typing import TYPE_CHECKING
@@ -55,6 +57,51 @@ def populate_test_datasets(self, filepath, db_session: "Session"):
         with open(filepath) as f:
             data = json.load(f)
 
+        # Licenses (populate license table first so feeds can reference them)
+        if "licenses" in data:
+            for lic in data["licenses"]:
+                # skip if id missing
+                lic_id = lic.get("id")
+                if not lic_id:
+                    continue
+                existing = db_session.get(License, lic_id)
+                if existing:
+                    # optionally update existing fields if needed
+                    continue
+                license_obj = License(
+                    id=lic_id,
+                    type=lic.get("type", "standard"),
+                    is_spdx=lic.get("is_spdx", False),
+                    name=lic.get("name"),
+                    url=lic.get("url"),
+                    description=lic.get("description"),
+                    content_txt=lic.get("content_txt"),
+                    content_html=lic.get("content_html"),
+                    created_at=lic.get("created_at"),
+                    updated_at=lic.get("updated_at"),
+                )
+                db_session.add(license_obj)
+            db_session.commit()
+
+        # Rules (optional section to seed rule metadata used by license_rules)
+        if "rules" in data:
+            for rule in data["rules"]:
+                rule_name = rule.get("name")
+                if not rule_name:
+                    continue
+                existing_rule = db_session.get(Rule, rule_name)
+                if existing_rule:
+                    continue
+                db_session.add(
+                    Rule(
+                        name=rule_name,
+                        label=rule.get("label") or rule_name,
+                        type=rule.get("type") or "permission",
+                        description=rule.get("description"),
+                    )
+                )
+            db_session.commit()
+
         # GTFS Feeds
         if "feeds" in data:
             self.populate_test_feeds(data["feeds"], db_session)
@@ -130,6 +177,29 @@ def populate_test_datasets(self, filepath, db_session: "Session"):
                     db_session.query(Feature).filter(Feature.name == report_features["feature_name"]).first()
                 )
 
+        # License rules: populate association table by creating missing Rule rows and attaching them to License
+        if "license_rules" in data:
+            for lr in data["license_rules"]:
+                license_id = lr.get("license_id")
+                rule_id = lr.get("rule_id")
+                if not license_id or not rule_id:
+                    continue
+                license_obj = db_session.get(License, license_id)
+                if not license_obj:
+                    self.logger.error(f"No license found with id: {license_id}; skipping license_rule {rule_id}")
+                    continue
+                rule_obj = db_session.get(Rule, rule_id)
+                if not rule_obj:
+                    # Create a minimal Rule entry; label and type set conservatively
+                    rule_obj = Rule(name=rule_id, label=rule_id, type="permission", description=None)
+                    db_session.add(rule_obj)
+                    # flush so the relationship can reference it immediately
+                    db_session.flush()
+                # Attach if not already associated
+                if rule_obj not in license_obj.rules:
+                    license_obj.rules.append(rule_obj)
+            db_session.commit()
+
         # GBFS version
         if "gbfs_versions" in data:
             for version in data["gbfs_versions"]:
@@ -180,9 +250,13 @@ def populate_test_feeds(self, feeds_data, db_session: "Session"):
                 note=feed_data["note"],
                 authentication_info_url=None,
                 api_key_parameter_name=None,
-                license_url=None,
+                license_url=feed_data["source_info"]["license_url"],
                 feed_contact_email=feed_data["feed_contact_email"],
                 producer_url=feed_data["source_info"]["producer_url"],
+                # license_id may be missing or an empty string; coerce empty -> None to avoid FK violation
+                license_id=(feed_data["source_info"].get("license_id") or None),
+                # allow empty notes to stay as empty string; coerce if you prefer None
+                license_notes=(feed_data["source_info"].get("license_notes") or None),
                 operational_status="published",
             )
             locations = []

api/src/shared/db_models/basic_feed_impl.py

@@ -20,6 +20,11 @@ class Config:
     def from_orm(cls, feed: Feed | None) -> BasicFeed | None:
         if not feed:
             return None
+        # Determine license_is_spdx from the related License ORM if available
+        license_is_spdx = None
+        if getattr(feed, "license", None) is not None:
+            license_is_spdx = feed.license.is_spdx
+
         return cls(
             id=feed.stable_id,
             data_type=feed.data_type,
@@ -35,6 +40,9 @@ def from_orm(cls, feed: Feed | None) -> BasicFeed | None:
                 authentication_info_url=feed.authentication_info_url,
                 api_key_parameter_name=feed.api_key_parameter_name,
                 license_url=feed.license_url,
+                license_id=feed.license_id,
+                license_is_spdx=license_is_spdx,
+                license_notes=feed.license_notes,
             ),
             redirects=sorted([RedirectImpl.from_orm(item) for item in feed.redirectingids], key=lambda x: x.target_id),
         )

api/src/shared/db_models/license_base_impl.py

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+from typing import Optional
+
+from feeds_gen.models.license_base import LicenseBase
+from pydantic import ConfigDict
+
+from shared.database_gen.sqlacodegen_models import License as LicenseOrm
+
+
+class LicenseBaseImpl(LicenseBase):
+    """Pydantic model hydratable directly from a License ORM row."""
+
+    model_config = ConfigDict(from_attributes=True)
+
+    @classmethod
+    def from_orm(cls, license_orm: Optional[LicenseOrm]) -> Optional[LicenseBase]:
+        """Convert a SQLAlchemy License row into the base License model."""
+        if not license_orm:
+            return None
+
+        return cls(
+            id=license_orm.id,
+            type=license_orm.type,
+            is_spdx=license_orm.is_spdx,
+            name=license_orm.name,
+            url=license_orm.url,
+            description=license_orm.description,
+            created_at=license_orm.created_at,
+            updated_at=license_orm.updated_at,
+        )