Fixed a small issue with RFC 5987 headers arriving as unicode in parser

Also added API docs regarding header encoding, incl support for RFC 5987
for filename fields in Content-Disposition headers in multipart/form-data
parts, incl. an example of an upload request with a utf-8 encoded filename.

docs/api/fileops.rst

@@ -161,18 +161,10 @@ Upload file
       Content-Disposition: form-data; name="file"; filename="whistle_v2.gcode"
       Content-Type: application/octet-stream
 
-      ;Generated with Cura_SteamEngine 13.11.2
       M109 T0 S220.000000
       T0
-      ;Sliced at: Wed 11-12-2013 16:53:12
-      ;Basic settings: Layer height: 0.2 Walls: 0.8 Fill: 20
-      ;Print time: #P_TIME#
-      ;Filament used: #F_AMNT#m #F_WGHT#g
-      ;Filament cost: #F_COST#
-      ;M190 S70 ;Uncomment to add your own bed temperature line
-      ;M109 S220 ;Uncomment to add your own temperature line
-      G21        ;metric values
-      G90        ;absolute positioning
+      G21
+      G90
       ...
       ------WebKitFormBoundaryDeC2E3iWbTv1PwMC
       Content-Disposition: form-data; name="select"
@@ -188,7 +180,7 @@ Upload file
 
       HTTP/1.1 200 OK
       Content-Type: application/json
-      Location: http://example.com/api/files/sdcard/whistle_.gcode
+      Location: http://example.com/api/files/sdcard/whistle_v2.gcode
 
       {
         "files": {
@@ -208,6 +200,46 @@ Upload file
             }
           }
         },
+        "done": false
+      }
+
+   **Example with UTF-8 encoded filename following RFC 5987**
+
+   .. sourcecode:: http
+
+      POST /api/files/local HTTP/1.1
+      Host: example.com
+      X-Api-Key: abcdef...
+      Content-Type: multipart/form-data; boundary=----WebKitFormBoundaryDeC2E3iWbTv1PwMC
+
+      ------WebKitFormBoundaryDeC2E3iWbTv1PwMC
+      Content-Disposition: form-data; name="file"; filename*=utf-8''20mm-%C3%BCml%C3%A4ut-b%C3%B6x.gcode
+      Content-Type: application/octet-stream
+
+      M109 T0 S220.000000
+      T0
+      G21
+      G90
+      ...
+      ------WebKitFormBoundaryDeC2E3iWbTv1PwMC--
+
+   .. sourcecode:: http
+
+      HTTP/1.1 200 OK
+      Content-Type: application/json
+      Location: http://example.com/api/files/local/20mm-umlaut-box.gcode
+
+      {
+        "files": {
+          "local": {
+            "name": "20mm-umlaut-box",
+            "origin": "local",
+            "refs": {
+              "resource": "http://example.com/api/files/local/whistle_v2.gcode",
+              "download": "http://example.com/downloads/files/local/whistle_v2.gcode"
+            }
+          }
+        },
         "done": true
       }
 

docs/api/general.rst

@@ -74,7 +74,19 @@ Encoding
 
 OctoPrint uses UTF-8 as charset.
 
-.. _sec-api-cross-origin:
+That also includes headers in ``multipart/form-data`` requests, in order to allow the full UTF-8 range of characters
+for uploaded filenames. If a ``multipart/form-data`` sub header cannot be decoded as UTF-8, OctoPrint will also attempt
+to decode it as ISO-8859-1.
+
+Additionally, OctoPrint supports replacing the ``filename`` field in the ``Content-Disposition`` header of a
+multipart field with a ``filename*`` field following `RFC 5987, Section 3.2 <https://tools.ietf.org/html/rfc5987#section-3.2>`_,
+which allows defining the charset used for encoding the filename. If both ``filename`` and ``filename*`` fields are
+present, following the recommendation of the RFC ``filename*`` will be used.
+
+For an example on how to send a request utilizing RFC 5987 for the ``filename*`` attribute, see the second example
+in :ref:`Upload file <sec-api-fileops-uploadfile>`.
+
+.. _sec-api-general-crossorigin:
 
 Cross-origin requests
 =====================

src/octoprint/server/util/tornado.py

@@ -303,7 +303,7 @@ class UploadStorageFallbackHandler(tornado.web.RequestHandler):
 				filename = _extended_header_value(filename)
 			except:
 				# parse error, this is not RFC 5987 compliant after all
-				self._logger.warn("extended filename* value {!r} is not RFC 5987 compliant")
+				self._logger.warn("extended filename* value {!r} is not RFC 5987 compliant".format(filename))
 				self.send_error(400)
 				return
 		else:
@@ -494,11 +494,11 @@ def _extended_header_value(value):
 		# RFC 5987 section 3.2
 		from urllib import unquote
 		encoding, _, value = value.split("'", 2)
-		return unquote(value).decode(encoding)
+		return unquote(octoprint.util.to_str(value, encoding="iso-8859-1")).decode(encoding)
 
 	else:
 		# no encoding provided, strip potentially present quotes and call it a day
-		return _strip_value_quotes(value)
+		return octoprint.util.to_unicode(_strip_value_quotes(value), encoding="utf-8")
 
 
 class WsgiInputContainer(object):

tests/server/util/tornado.py

@@ -76,9 +76,14 @@ class StripValueQuotesTest(unittest.TestCase):
 class ExtendedHeaderValueTest(unittest.TestCase):
 
 	@data(
-		("", u""),
+		(u"", u""),
 		(None, None),
+		(u'"quoted-string"', u"quoted-string"),
+		(u'"qüöted-string"', u"qüöted-string"),
+		(u"iso-8859-1'en'%A3%20rates", u"£ rates"),
+		(u"UTF-8''%c2%a3%20and%20%e2%82%ac%20rates", u"£ and € rates"),
 		('"quoted-string"', u"quoted-string"),
+		('"qüöted-string"', u"qüöted-string"),
 		("iso-8859-1'en'%A3%20rates", u"£ rates"),
 		("UTF-8''%c2%a3%20and%20%e2%82%ac%20rates", u"£ and € rates")
 	)