db backed source creation

repub/model.py

@@ -84,6 +84,132 @@ def initialize_database(db_path: str | Path | None = None) -> Path:
     return resolved_path
 
 
+def source_slug_exists(slug: str) -> bool:
+    with database.connection_context():
+        return Source.select().where(Source.slug == slug).exists()
+
+
+def create_source(
+    *,
+    name: str,
+    slug: str,
+    source_type: str,
+    notes: str,
+    spider_arguments: str,
+    enabled: bool,
+    cron_minute: str,
+    cron_hour: str,
+    cron_day_of_month: str,
+    cron_day_of_week: str,
+    cron_month: str,
+    feed_url: str = "",
+    pangea_domain: str = "",
+    pangea_category: str = "",
+    content_type: str = "",
+    only_newest: bool = True,
+    max_articles: int | None = None,
+    oldest_article: int | None = None,
+    include_authors: bool = True,
+    exclude_media: bool = False,
+    include_content: bool = True,
+    content_format: str = "",
+) -> Source:
+    with database.connection_context():
+        with database.atomic():
+            source = Source.create(
+                name=name,
+                slug=slug,
+                source_type=source_type,
+                notes=notes,
+            )
+            if source_type == "feed":
+                SourceFeed.create(
+                    source=source,
+                    feed_url=feed_url,
+                )
+            else:
+                SourcePangea.create(
+                    source=source,
+                    domain=pangea_domain,
+                    category_name=pangea_category,
+                    content_type=content_type,
+                    only_newest=only_newest,
+                    max_articles=max_articles,
+                    oldest_article=oldest_article,
+                    include_authors=include_authors,
+                    exclude_media=exclude_media,
+                    include_content=include_content,
+                    content_format=content_format,
+                )
+            Job.create(
+                source=source,
+                enabled=enabled,
+                spider_arguments=spider_arguments,
+                cron_minute=cron_minute,
+                cron_hour=cron_hour,
+                cron_day_of_month=cron_day_of_month,
+                cron_day_of_week=cron_day_of_week,
+                cron_month=cron_month,
+            )
+            return source
+
+
+def load_sources() -> tuple[dict[str, object], ...]:
+    with database.connection_context():
+        sources = tuple(Source.select().order_by(Source.created_at.desc()))
+        source_ids = tuple(int(source.get_id()) for source in sources)
+        if not source_ids:
+            return ()
+        jobs = {
+            job.source_id: job for job in Job.select().where(Job.source.in_(source_ids))
+        }
+        feed_configs = {
+            config.source_id: config
+            for config in SourceFeed.select().where(SourceFeed.source.in_(source_ids))
+        }
+        pangea_configs = {
+            config.source_id: config
+            for config in SourcePangea.select().where(
+                SourcePangea.source.in_(source_ids)
+            )
+        }
+        return tuple(
+            _project_source(source, jobs, feed_configs, pangea_configs)
+            for source in sources
+        )
+
+
+def _project_source(
+    source: "Source",
+    jobs: dict[int, "Job"],
+    feed_configs: dict[int, "SourceFeed"],
+    pangea_configs: dict[int, "SourcePangea"],
+) -> dict[str, object]:
+    source_id = int(source.get_id())
+    job = jobs[source_id]
+    if source.source_type == "feed":
+        upstream = feed_configs[source_id].feed_url
+        source_type = "Feed"
+    else:
+        pangea = pangea_configs[source_id]
+        upstream = f"{pangea.domain} / {pangea.category_name}"
+        source_type = "Pangea"
+
+    return {
+        "name": source.name,
+        "slug": source.slug,
+        "source_type": source_type,
+        "upstream": upstream,
+        "schedule": (
+            f"cron: {job.cron_minute} {job.cron_hour} {job.cron_day_of_month} "
+            f"{job.cron_month} {job.cron_day_of_week}"
+        ),
+        "last_run": "Never run",
+        "state": "Enabled" if job.enabled else "Disabled",
+        "state_tone": "scheduled" if job.enabled else "idle",
+    }
+
+
 class BaseModel(Model):
     class Meta:
         database = database