Break coach listing download per-page.

.gitignore

@@ -3,3 +3,6 @@
 
 # A symlink produced by default when running `nix build`.
 result
+
+# The results of the `chess.com` coach scraping.
+data/

main.py

@@ -8,66 +8,70 @@ from bs4 import BeautifulSoup
 
 
 # References to paths we use to save any scraped content.
-DATA_COACH_LINKS = "data/coach_links.txt"
+DATA_COACH_LIST = "data/pages/{}.txt"
 DATA_COACH_DIR = "data/coach/{}/{}"
 
 
 async def scrape_coach_links(page_no):
     """Scrape a single coach page listing."""
     links = []
-    async with aiohttp.ClientSession() as session:
     href = f"https://www.chess.com/coaches?sortBy=alphabetical&page={page_no}"
+    async with aiohttp.ClientSession() as session:
         async with session.get(href) as response:
             if response.status != 200:
-                print(f"Encountered {response.status} when retrieving {href}")
+                print(f"Encountered {response.status} when retrieving {href}.")
                 return
             html = await response.text()
             soup = BeautifulSoup(html, "html.parser")
-            members = soup.find_all("a", class_="members-categories-username")
+            members = soup.find_all("a", class_="members-categories-blahblah")
             for member in members:
                 links.append(member.get("href"))
 
     return links
 
 
-async def scrape_all_coach_links(max_pages=62):
-    """Scans through chess.com/coaches for all member links."""
-    if os.path.isfile(DATA_COACH_LINKS):
-        with open(DATA_COACH_LINKS, "r") as f:
-            return f.readlines()
+async def scrape_all_coach_links(max_pages=64):
+    """Scan through https://www.chess.com/coaches for all member links."""
+    links = []
     for i in range(1, max_pages + 1):
-        # Nest the file context manager here so I can `tail -f` the file.
-        with open(DATA_COACH_LINKS, "a") as f:
+        filepath = DATA_COACH_LIST.format(i)
+        if os.path.isfile(filepath):
+            with open(filepath, "r") as f:
+                links = f.readlines()
+            print(f"{filepath} already exists.")
+        else:
             links = await scrape_coach_links(i)
+            with open(filepath, "w") as f:
                 for link in links:
                     f.write(f"{link}\n")
-            await asyncio.sleep(random.randint(2, 5))
+            print(f"Downloaded page {i} of coach list.")
+            await asyncio.sleep(random.randint(3, 7))
+
     return links
 
 
 async def download_member_info(member_name, filename, href):
     """Download member-specific content.
 
-    @return: True if we downloaded content. False if the results already
-    existed locally.
+    @return: True if we downloaded content. False if the download already
+    exists locally.
     """
     target = DATA_COACH_DIR.format(member_name, filename)
     if os.path.isfile(target):
         return False
-    with open(target, "w") as f:
     async with aiohttp.ClientSession() as session:
         async with session.get(href) as response:
             if response.status != 200:
                 print(f"Encountered {response.status} when retrieving {href}")
                 return
+            with open(target, "w") as f:
                 f.write(await response.text())
     return True
 
 
 async def main():
     links = await scrape_all_coach_links()
-    for link in links:
-        href = link.strip()
+    for href in [link.strip() for link in links]:
         member_name = href[len("https://www.chess.com/member/") :]
         downloaded = await asyncio.gather(
             download_member_info(
@@ -86,14 +90,14 @@ async def main():
                 f"https://www.chess.com/callback/member/stats/{member_name}",
             ),
         )
-        # Only want to sleep if the files didn't already exist.
         if any(downloaded):
-            await asyncio.sleep(random.randint(2, 5))
-            print(f"Downloaded {member_name}")
+            await asyncio.sleep(random.randint(3, 7))
+            print(f"Downloaded {member_name} info.")
         else:
-            print(f"Skipping {member_name}")
+            print(f"Skipping {member_name} download.")
 
 
 if __name__ == "__main__":
+    os.makedirs("data/pages", exist_ok=True)
     os.makedirs("data/coach", exist_ok=True)
     asyncio.run(main())