[utils,compat] Move struct_pack and struct_unpack to compat.py

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 7bcc85e2b530cb2eadb714e100b56d8f4637b87d..fa16a42ad4940bd28373713d27c0becd422ec29c 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -14,8 +14,8 @@ import email.utils
 import errno
 import functools
 import gzip
-import itertools
 import io
+import itertools
 import json
 import locale
 import math
@@ -24,9 +24,8 @@ import os
 import pipes
 import platform
 import re
-import ssl
 import socket
-import struct
+import ssl
 import subprocess
 import sys
 import tempfile
@@ -53,6 +52,7 @@ from .compat import (
     compat_urlparse,
     compat_xpath,
     shlex_quote,
+    struct_pack,
 )
 
 
@@ -89,6 +89,11 @@ KNOWN_EXTENSIONS = (
     'wav',
     'f4f', 'f4m', 'm3u8', 'smil')
 
+# needed for sanitizing filenames in restricted mode
+ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþÿ',
+                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOUUUUYP', ['ss'],
+                                        'aaaaaa', ['ae'], 'ceeeeiiiionoooooouuuuypy')))
+
 
 def preferredencoding():
     """Get preferred encoding.
@@ -251,9 +256,9 @@ def get_element_by_attribute(attribute, value, html):
 
     m = re.search(r'''(?xs)
         <([a-zA-Z0-9:._-]+)
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
          \s+%s=['"]?%s['"]?
-         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]+|="[^"]+"|='[^']+'))*?
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
         \s*>
         (?P<content>.*?)
         </\1>
@@ -365,6 +370,8 @@ def sanitize_filename(s, restricted=False, is_id=False):
     Set is_id if this is not an arbitrary string, but an ID that should be kept if possible
     """
     def replace_insane(char):
+        if restricted and char in ACCENT_CHARS:
+            return ACCENT_CHARS[char]
         if char == '?' or ord(char) < 32 or ord(char) == 127:
             return ''
         elif char == '"':
@@ -1754,24 +1761,6 @@ def escape_url(url):
         fragment=escape_rfc3986(url_parsed.fragment)
     ).geturl()
 
-try:
-    struct.pack('!I', 0)
-except TypeError:
-    # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument
-    # See https://bugs.python.org/issue19099
-    def struct_pack(spec, *args):
-        if isinstance(spec, compat_str):
-            spec = spec.encode('ascii')
-        return struct.pack(spec, *args)
-
-    def struct_unpack(spec, *args):
-        if isinstance(spec, compat_str):
-            spec = spec.encode('ascii')
-        return struct.unpack(spec, *args)
-else:
-    struct_pack = struct.pack
-    struct_unpack = struct.unpack
-
 
 def read_batch_urls(batch_fd):
     def fixup(url):