diff options
Diffstat (limited to 'synapse/rest/media/v1/_base.py')
| -rw-r--r-- | synapse/rest/media/v1/_base.py | 94 |
1 files changed, 53 insertions, 41 deletions
diff --git a/synapse/rest/media/v1/_base.py b/synapse/rest/media/v1/_base.py index 5fefee4dde..3689777266 100644 --- a/synapse/rest/media/v1/_base.py +++ b/synapse/rest/media/v1/_base.py @@ -17,7 +17,6 @@ import logging import os -from six import PY3 from six.moves import urllib from twisted.internet import defer @@ -30,6 +29,22 @@ from synapse.util.stringutils import is_ascii logger = logging.getLogger(__name__) +# list all text content types that will have the charset default to UTF-8 when +# none is given +TEXT_CONTENT_TYPES = [ + "text/css", + "text/csv", + "text/html", + "text/calendar", + "text/plain", + "text/javascript", + "application/json", + "application/ld+json", + "application/rtf", + "image/svg+xml", + "text/xml", +] + def parse_media_id(request): try: @@ -96,7 +111,14 @@ def add_file_headers(request, media_type, file_size, upload_name): def _quote(x): return urllib.parse.quote(x.encode("utf-8")) - request.setHeader(b"Content-Type", media_type.encode("UTF-8")) + # Default to a UTF-8 charset for text content types. + # ex, uses UTF-8 for 'text/css' but not 'text/css; charset=UTF-16' + if media_type.lower() in TEXT_CONTENT_TYPES: + content_type = media_type + "; charset=UTF-8" + else: + content_type = media_type + + request.setHeader(b"Content-Type", content_type.encode("UTF-8")) if upload_name: # RFC6266 section 4.1 [1] defines both `filename` and `filename*`. # @@ -135,27 +157,25 @@ def add_file_headers(request, media_type, file_size, upload_name): # separators as defined in RFC2616. SP and HT are handled separately. # see _can_encode_filename_as_token. -_FILENAME_SEPARATOR_CHARS = set( - ( - "(", - ")", - "<", - ">", - "@", - ",", - ";", - ":", - "\\", - '"', - "/", - "[", - "]", - "?", - "=", - "{", - "}", - ) -) +_FILENAME_SEPARATOR_CHARS = { + "(", + ")", + "<", + ">", + "@", + ",", + ";", + ":", + "\\", + '"', + "/", + "[", + "]", + "?", + "=", + "{", + "}", +} def _can_encode_filename_as_token(x): @@ -195,7 +215,7 @@ def respond_with_responder(request, responder, media_type, file_size, upload_nam respond_404(request) return - logger.debug("Responding to media request with responder %s") + logger.debug("Responding to media request with responder %s", responder) add_file_headers(request, media_type, file_size, upload_name) try: with responder: @@ -303,23 +323,15 @@ def get_filename_from_headers(headers): upload_name_utf8 = upload_name_utf8[7:] # We have a filename*= section. This MUST be ASCII, and any UTF-8 # bytes are %-quoted. - if PY3: - try: - # Once it is decoded, we can then unquote the %-encoded - # parts strictly into a unicode string. - upload_name = urllib.parse.unquote( - upload_name_utf8.decode("ascii"), errors="strict" - ) - except UnicodeDecodeError: - # Incorrect UTF-8. - pass - else: - # On Python 2, we first unquote the %-encoded parts and then - # decode it strictly using UTF-8. - try: - upload_name = urllib.parse.unquote(upload_name_utf8).decode("utf8") - except UnicodeDecodeError: - pass + try: + # Once it is decoded, we can then unquote the %-encoded + # parts strictly into a unicode string. + upload_name = urllib.parse.unquote( + upload_name_utf8.decode("ascii"), errors="strict" + ) + except UnicodeDecodeError: + # Incorrect UTF-8. + pass # If there isn't check for an ascii name. if not upload_name: |