pyproject.toml

@@ -9,7 +9,15 @@ authors = [{ name = "pdf", email = "git@nixon.mozmail.com" }]
 description = "Snowflake scraper"
 requires-python = ">=3.9"
 license = { text = "MIT" }
-dependencies = ["pydantic==1.10.10", "scrapy>=2.10.0", "feedparser>=6.0.10"]
+dependencies = [
+    "pydantic==1.10.10",
+    "scrapy>=2.10.0",
+    "feedparser>=6.0.10",
+    "uvloop>=0.17.0",
+    "rocketry>=2.5.1",
+    "fastapi>=0.101.0",
+    "uvicorn>=0.22.0",
+]
 
 [project.scripts]
 snowscraper = "snowscraper.cli:run"

snowscraper/api.py

@@ -0,0 +1,3 @@
+from fastapi import APIRouter
+
+app = APIRouter()

snowscraper/cli.py

@@ -1,16 +1,52 @@
 import argparse
+import asyncio
 import sys
 from datetime import datetime
 from datetime import timezone
+from types import FrameType
+
+import uvicorn
+import uvloop
 
 from .controller import import_scrapers
 from .controller import run_all
 
+app_fastapi = None
+app_rocketry = None
+
+
+class SnowScraper(uvicorn.Server):
+    def handle_exit(self, sig: int, frame: FrameType) -> None:
+        print("Shutting down SnowScraper")
+        return super().handle_exit(sig, frame)
+
+
+async def start_server() -> None:
+    print("Starting SnowScraper")
+    server = SnowScraper(config=uvicorn.Config(app_fastapi, workers=1, loop="uvloop"))
+    fastapi = asyncio.create_task(server.serve())
+    rocket = asyncio.create_task(app_rocketry.serve())
+    app_rocketry.task
+
+    await asyncio.wait([rocket, fastapi], return_when=asyncio.FIRST_COMPLETED)
+
 
 def main(args: argparse.Namespace) -> None:
+    global app_fastapi, app_rocketry
+
+    from .scheduler import app as app_rocketry
+    from .webserver import app as app_fastapi
+
     import_scrapers()
     run_all(args)
 
+    if sys.version_info >= (3, 11):
+        with asyncio.Runner(loop_factory=uvloop.new_event_loop) as runner:
+            runner.run(start_server())
+    else:
+        uvloop.install()
+        asyncio.run(start_server())
+
 
 def run():
     parser = argparse.ArgumentParser(description="Snowflake scraper")

snowscraper/controller.py

@@ -12,11 +12,9 @@ def register_scraper(cls):
 
 
 def run_all(args: argparse.Namespace):
-    results = {}
     for scraper_cls in SCRAPERS.values():
         scraper = scraper_cls(after=args.after)
-        results |= scraper.scrape()
+        scraper.scrape()
-    print(results)
 
 
 def import_scrapers():

snowscraper/helpers.py

@@ -2,8 +2,24 @@ from datetime import datetime
 
 
 def string_to_datetime(date_string):
+    """
+    Convert a date string to a datetime object.
+
+    This function supports the following formats:
+    - ISO 8601 with and without timezone offset: e.g., '2023-08-02T09:27:54-05:00' or '2023-08-07T18:32:05.537Z'
+    - RFC 1123: e.g., 'Mon, 07 Aug 2023 18:32:05 GMT'
+
+    Args:
+    - date_string (str): The date string to convert
+
+    Returns:
+    - datetime.datetime: The datetime representation of the provided date string.
+
+    Raises:
+    - ValueError: If the provided date string doesn't match any of the supported formats.
+    """
     try:
-        # try ISO 8601
+        # First, try ISO 8601 format
         if "Z" in date_string:
             return datetime.fromisoformat(date_string.replace("Z", "+00:00"))
         return datetime.fromisoformat(date_string)
@@ -11,9 +27,10 @@ def string_to_datetime(date_string):
         pass
 
     try:
-        # try RFC 1123
+        # Then, try RFC 1123 format
         return datetime.strptime(date_string, "%a, %d %b %Y %H:%M:%S %Z")
     except ValueError:
         pass
 
+    # If neither format matches, raise an exception
     raise ValueError(f"Unsupported date format: {date_string}")

snowscraper/scheduler.py

@@ -0,0 +1,6 @@
+from rocketry import Rocketry
+
+app = Rocketry(execution="async")
+
+if __name__ == "__main__":
+    app.run()

snowscraper/scrapers/medium.py

@@ -27,6 +27,7 @@ class MediumScraper(BaseScraper):
                     "published": string_to_datetime(entry["published"]),
                     "updated": updated,
                 }
+        # print(self.data)
         return self.data
 
     def transform(self):

snowscraper/scrapers/quickstarts.py

@@ -2,9 +2,7 @@ from datetime import datetime
 from datetime import timezone
 
 import scrapy
-from scrapy import signals
 from scrapy.crawler import CrawlerProcess
-from scrapy.signalmanager import dispatcher
 
 from ..controller import register_scraper
 from ..scraper import BaseScraper
@@ -25,29 +23,24 @@ class QuickstartScraper(BaseScraper, scrapy.Spider):
     def start_requests(self):
         yield scrapy.Request(url=QuickStartsURL, callback=self.parse)
 
-    def signal_handler(self, signal, sender, item, response, spider):
-        self.data[item["key"]] = item
-        self.data[item["key"]].pop("key")
-
     def scrape(self):
         print("Scraping Quickstarts")
-        dispatcher.connect(self.signal_handler, signal=signals.item_scraped)
         process = CrawlerProcess({"LOG_LEVEL": "ERROR"})
         process.crawl(QuickstartScraper, after=self.after)
         process.start()
-        return self.data
 
     def parse(self, response):
         for card in response.css("card-sorter#cards > a.codelab-card"):
             updated = string_to_datetime(card.attrib["data-updated"])
             if updated > self.after:
                 key = QuickStartsURL.rstrip("/") + card.attrib["href"]
-                yield {
+                self.data[key] = {
-                    "key": key,
                     "title": card.attrib["data-title"],
                     "updated": updated,
                     "tags": card.attrib["data-tags"],
                 }
+                # print(key, self.data[key])
+                yield self.data[key]
 
     def transform(self):
         return self.data

snowscraper/scrapers/youtube.py

@@ -0,0 +1,20 @@
+from datetime import datetime
+from datetime import timezone
+
+from ..controller import register_scraper
+from ..scraper import BaseScraper
+
+
+@register_scraper
+class YoutubeScraper(BaseScraper):
+    def __init__(self, after=datetime(1970, 1, 1, tzinfo=timezone.utc), *args, **kwargs):
+        super(YoutubeScraper, self).__init__(*args, **kwargs)
+        self.data = {}
+        self.after = after
+
+    def scrape(self):
+        print("Scraping YouTube... Unimplemented")
+        self.data = {}
+
+    def transform(self):
+        return self.data

snowscraper/webserver.py

@@ -0,0 +1,11 @@
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+
+from .api import app as api_v1
+
+app = FastAPI()
+app.add_middleware(CORSMiddleware, allow_origins=["*"], allow_methods=["GET"], allow_headers=["*"])
+app.include_router(api_v1, prefix="/api/v1")
+
+if __name__ == "__main__":
+    app.run()