telemetry: Create a safe url generator.

tools/perf/profile_creators/profile_safe_url_generator.py

Index: tools/perf/profile_creators/profile_safe_url_generator.py
diff --git a/tools/perf/profile_creators/profile_safe_url_generator.py b/tools/perf/profile_creators/profile_safe_url_generator.py
new file mode 100644
index 0000000000000000000000000000000000000000..f375a7ad014a7980adfd4efc7f7f91d2d1586ec2
--- /dev/null
+++ b/tools/perf/profile_creators/profile_safe_url_generator.py
@@ -0,0 +1,89 @@
+# Copyright 2015 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+# This script requires the python package BeautifulSoup (v4). This package is
+# not included in the telemetry repository. You can install it with the
+# command: "pip install beautifulsoup4"
+
+import logging
+import urllib2
+import urlparse
+
+from bs4 import BeautifulSoup

[nednguyen, 2015/02/12 01:43:46]

Does bot machine has bs4?

[erikchen, 2015/02/12 01:46:57]

No.

+
+def _IsURLASCII(url):
+  """Whether |url| is ascii encoded."""
+  try:
+    url.decode("ascii")
+  except UnicodeEncodeError:
+    return False
+  return True
+
+def GenerateSafeUrls():
+  """
+  Generates a safe list of urls from a seed list. Each href in the HTML fetched
+  from the url from the seed list is placed into the safe list. The safe list
+  contains unsanitized urls.
+  """
+  # A list of websites whose hrefs are unlikely to link to sites that contain
+  # malware.
+  seed_urls = [
+    "https://www.cnn.com",
+    "https://www.youtube.com",
+    "https://www.facebook.com",
+    "https://www.twitter.com",
+    "https://www.yahoo.com",
+    "https://www.amazon.com",
+    "https://www.wikipedia.com",
+    "https://www.bing.com",
+    "https://www.dailymotion.com",
+    "https://www.stackoverflow.com",
+    "https://www.google.com/#q=dumpling",
+    "http://www.baidu.com/s?wd=rice",
+    "http://www.baidu.com/s?wd=cow",
+    "https://www.google.com/#q=fox",
+    "http://www.yahoo.co.jp/",
+    "http://www.yandex.ru/",
+    "https://www.imdb.com/",
+    "http://www.huffingtonpost.com/",
+    "https://www.deviantart.com/",
+    "http://www.wsj.com/",
+  ]
+
+  safe_urls = set()
+
+  for url in seed_urls:
+    try:
+      # Fetch and parse the HTML.
+      response = urllib2.urlopen(url)
+      html = response.read()
+      soup = BeautifulSoup(html)
+    except:
+      logging.exception("Error fetching or parsing url: %s", url)
+      raise
+
+    # Looks for all hrefs.
+    for link in soup.find_all('a'):
+      possibly_relative_url = link.get("href")
+      if not possibly_relative_url:
+        continue
+
+      # For simplicity, ignore urls that aren't ascii encoded.
+      if not _IsURLASCII(possibly_relative_url):
+        continue
+
+      absolute_url = urlparse.urljoin(url, possibly_relative_url)
+      safe_urls.add(absolute_url)
+
+  # Sort the urls, to make them easier to view in bulk.
+  safe_urls_list = list(safe_urls)
+  safe_urls_list.sort()
+
+  # Print out the safe urls in a format that is conducive to being copied into
+  # a python file.
+  for url in safe_urls_list:
+    print "\"%s\"," % (url)
+
+if __name__ == "__main__":
+  GenerateSafeUrls()