User cloudscraper

check.py

@@ -6,15 +6,13 @@ import html
 import os
 import re
 import smtplib
-import sys
 import warnings
-from dataclasses import dataclass
 from datetime import date
 from email.mime.text import MIMEText
 from email.utils import formatdate, make_msgid
 from urllib.parse import urlparse, urlunparse
 
-import cloudscraper  # type: ignore
+import cloudscraper
 import requests
 import yaml
 from requests.adapters import HTTPAdapter
@@ -23,8 +21,6 @@ from urllib3.util.url import parse_url
 
 from conference import LiveConference, config, load_yaml
 
-IS_TTY = sys.stdout.isatty()
-
 
 class AbsoluteDNSAdapter(HTTPAdapter):
     """A custom adapter for requests to ensure hostnames are treated as absolute."""
@@ -80,7 +76,6 @@ not_here_list = [
     "This page does not exist yet",
     "404 Not Found",
     "500 Internal Server Error",
-    "500: Internal Server Error",
     "Test Page for the Apache HTTP Server",
     "Site not found · GitHub Pages",
     "504: Gateway time-out",
@@ -108,17 +103,9 @@ not_here_list = [
     "This page doesn't currently exist",
     "ERROR 503 - Service Unavailable",
     "ERROR 503",
-    "503 Service Unavailable",
     "401 Authorization Required",
     "Authorization Required",
     "used Cloudflare to restrict access",
-    "Error 1014",
-    "CNAME Cross-User Banned",
-    "looks like there's no page here",
-    "404 page can’t be found either",
-    "503 self-signed certificate",
-    "504 Gateway Timeout",
-    "<h2>Pages</h2>",
 ]
 
 
@@ -148,133 +135,31 @@ def normalize_url(url: str) -> str:
     return urlunparse(parsed_url._replace(netloc=normalized_netloc))
 
 
-def url_to_filename(url: str) -> str:
-    """
-    Convert a URL to a valid filename by replacing invalid characters with underscores.
+def check_conference(
+    name: str, src_url: str, year: int
+) -> tuple[bool, str, str | None]:
+    """Check if conference is live."""
+    url = src_url.format(year=year)
+    past_url = src_url.format(year=year - 1)
+    no_dot = {"bsideskbh.dk", "pif.camp"}
+    session = s if all(hostname not in url for hostname in no_dot) else s_no_dot
+    try:
+        r = session.get(url)
+    except requests.exceptions.ConnectionError:
+        return (False, "connection refused", None)
 
-    Args:
-        url (str): The URL to be converted.
+    not_here = find_not_here_message(r.text)
+    if (
+        len(r.text) < 2048
+        and 'http-equiv="refresh"' in r.text
+        and str(year) not in r.text
+    ):
+        return (False, "redirect to URL without year", r.url)
 
-    Returns:
-        str: A valid filename.
-    """
-    # Replace invalid characters with underscores
-    return re.sub(r'[\/:*?"<>|]', "_", url)
+    if normalize_url(r.url) == normalize_url(past_url):
+        return (False, "redirect to previous year", r.url)
 
-
-def parse_opengraph_tags(html: str) -> dict[str, str]:
-    """
-    Locate Open Graph meta tags, and return them as a dictionary.
-
-    Keys will match the name following 'og:', e.g. 'title', 'image', 'url', 'site_name'.
-    """
-    pattern = re.compile(
-        r'<meta\s+property="og:([^"]+)"\s+content="([^"]+)"\s*/?>', flags=re.IGNORECASE
-    )
-    matches = pattern.findall(html)
-
-    og_tags = {}
-    for prop, content in matches:
-        og_tags[prop] = content
-
-    return og_tags
-
-
-@dataclass
-class Conference:
-    """Conference."""
-
-    name: str
-    src_url: str
-    year: int
-    response: requests.models.Response | None = None
-    redirect_to_url: str | None = None
-
-    @property
-    def url(self) -> str:
-        """Conference URL."""
-        return self.build_url(year=self.year)
-
-    def build_url(self, year: int) -> str:
-        """Build conference URL."""
-        return self.src_url.format(year=year, two_digit_year=year % 2000)
-
-    @property
-    def past_url(self) -> str:
-        """URL for previous year."""
-        return self.build_url(year=self.year - 1)
-
-    def check(self) -> tuple[bool, str, str | None]:
-        """Check if conference is live."""
-        no_dot = {"bsideskbh.dk", "pif.camp"}
-        url = self.url
-        session = s if all(hostname not in url for hostname in no_dot) else s_no_dot
-        try:
-            r = session.get(url)
-        except requests.exceptions.ConnectionError:
-            return (False, "connection refused", None)
-
-        self.response = r
-
-        if r.url.endswith("404.html") or r.url.endswith("404.htm"):
-            return (False, "URL ends with 404.html/404.htm", r.url)
-
-        if not r.text:
-            return (False, "empty response", r.url)
-
-        not_here = find_not_here_message(r.text)
-        if (
-            len(r.text) < 2048
-            and 'http-equiv="refresh"' in r.text
-            and str(self.year) not in r.text
-        ):
-            return (False, "redirect to URL without year", r.url)
-
-        if normalize_url(r.url) == normalize_url(self.past_url):
-            return (False, "redirect to previous year", r.url)
-
-        if not_here:
-            return (False, not_here, r.url)
-
-        return (True, get_title(r.text), r.url)
-
-    def og_tags(self) -> dict[str, str]:
-        """Open Graph tags."""
-        return parse_opengraph_tags(self.response.text) if self.response else {}
-
-    def check_web_site(self) -> bool:
-        """Check if an individual web site is live."""
-        assert "{year}" in self.src_url or "{two_digit_year}" in self.src_url
-        if IS_TTY:
-            print(f"Checking {self.name} {self.year}: {self.url}")
-        live, msg, redirect_to_url = self.check()
-        if not live:
-            if IS_TTY:
-                print(f"  Not live: {msg}")
-            return False
-
-        if IS_TTY:
-            print(f"  Live! Title: {msg}")
-
-        og = "".join(f"\n{key}: {value}" for key, value in self.og_tags().items())
-
-        if og:
-            og = "\n\nOpen Graph\n\n" + og
-
-        if redirect_to_url and normalize_url(redirect_to_url) == normalize_url(
-            self.url
-        ):
-            body = f"{self.name}\n{self.url}\n"
-        else:
-            self.redirect_to_url = redirect_to_url
-            body = f"{self.name}\n{self.url} redirects to {redirect_to_url}\n"
-
-        body += f"Web page title: {msg}{og}" ""
-        send_mail(f"Conference site live: {self.name} - {self.year}", body)
-        if IS_TTY:
-            print(f"  Email sent")
-
-        return True
+    return (False, not_here, r.url) if not_here else (True, get_title(r.text), r.url)
 
 
 def send_mail(subject: str, body: str) -> None:
@@ -300,6 +185,23 @@ def send_mail(subject: str, body: str) -> None:
     s.quit()
 
 
+def check_conference_web_site(name: str, src_url: str, year: int) -> bool:
+    """Check if an individual web site is live."""
+    assert "{year}" in src_url
+    live, msg, redirect_to_url = check_conference(name, src_url, year)
+    url = src_url.format(year=year)
+    if live:
+        if redirect_to_url and normalize_url(redirect_to_url) == normalize_url(url):
+            body = f"{name}\n{url}\nWeb page title: {msg}"
+        else:
+            body = f"""{name}
+{url} redirects to {redirect_to_url}
+Web page title: {msg}"""
+        send_mail(f"Conference site live: {name} - {year}", body)
+
+    return live
+
+
 def find_new_conference_web_sites(
     today: date, live: list[LiveConference]
 ) -> list[LiveConference]:
@@ -310,59 +212,24 @@ def find_new_conference_web_sites(
 
     live_set = {(c["conference"], c["year"]) for c in live}
     for name, src_url in load_yaml("conferences").items():
-        for year in (this_year, this_year + 1):
-            if (name, year) in live_set:
-                continue
-            conf = Conference(name, src_url, year)
-            if not conf.check_web_site():
-                continue
-            c: LiveConference = {
-                "conference": name,
-                "year": year,
-                "live": today,
-                "url": conf.url,
-            }
-            if conf.redirect_to_url:
-                c["redirect_to_url"] = conf.redirect_to_url
-            new.append(c)
+        new += [
+            {"conference": name, "year": year, "live": today}
+            for year in (this_year, this_year + 1)
+            if (name, year) not in live_set
+            and check_conference_web_site(name, src_url, year)
+        ]
     return new
 
 
-class NoAliasDumper(yaml.SafeDumper):
-    """Dumper that disables YAML anchors and aliases."""
-
-    def ignore_aliases(self, data):
-        """Skip alias generation."""
-        return True  # disables anchors and aliases
-
-
 def main(show_not_live: bool = False) -> None:
     """Check fow new conference web sites."""
-    if IS_TTY:
-        print("Loading existing live conferences...")
     live: list[LiveConference] = load_yaml("live")
-    if IS_TTY:
-        print(f"Found {len(live)} existing live conferences")
-        print("\nChecking for new conference websites...")
     if not (new := find_new_conference_web_sites(date.today(), live)):
-        if IS_TTY:
-            print("\nNo new conference websites found")
         return
 
-    if IS_TTY:
-        print(f"\n{len(new)} new conference(s) found! Updating live.yaml...")
     live_filename = os.path.expanduser(config["data"]["live"])
     with open(live_filename, "w") as out:
-        yaml.dump(
-            live + new,
-            stream=out,
-            sort_keys=False,
-            Dumper=NoAliasDumper,
-            allow_unicode=True,
-            default_flow_style=False,
-        )
-    if IS_TTY:
-        print(f"Updated {live_filename}")
+        yaml.dump(live + new, stream=out, sort_keys=False)
 
 
 if __name__ == "__main__":

conference/__init__.py

@@ -28,7 +28,6 @@ class LiveConference(typing.TypedDict, total=False):
     year: int
     live: date
     url: str | None
-    redirect_to_url: str | None
 
 
 def load_yaml(name: str) -> typing.Any:

requirements.txt

@@ -1,2 +1 @@
 requests
-cloudscraper