From 274548f1b1e9efa0f776817efa1fb66de494d351 Mon Sep 17 00:00:00 2001
From: Henrik Levkowetz <henrik@levkowetz.com>
Date: Sun, 2 Jul 2017 21:32:44 +0000
Subject: [PATCH] Check the whole draft submission text when checking mime
 type, and catch decoding errors and raise ValidationError when converting
 submission form txt file to unicode for Draft().  - Legacy-Id: 13759

---
 ietf/submit/forms.py                | 11 +++++++++--
 ietf/submit/parsers/base.py         |  2 +-
 ietf/submit/parsers/plain_parser.py |  6 +++---
 3 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/ietf/submit/forms.py b/ietf/submit/forms.py
index 8c7d0eeae..6ea057454 100644
--- a/ietf/submit/forms.py
+++ b/ietf/submit/forms.py
@@ -203,11 +203,18 @@ class SubmissionUploadForm(forms.Form):
             # try to parse it
             txt_file = self.cleaned_data['txt']
             txt_file.seek(0)
-            self.parsed_draft = Draft(txt_file.read().decode('utf8'), txt_file.name)
+            bytes = txt_file.read()
+            txt_file.seek(0)
+            try:
+                text = bytes.decode('utf8')
+            except UnicodeDecodeError as e:
+                raise forms.ValidationError('Failed decoding the uploaded file: "%s"' % str(e))
+            #
+            self.parsed_draft = Draft(text, txt_file.name)
             self.filename = self.parsed_draft.filename
             self.revision = self.parsed_draft.revision
             self.title    = self.parsed_draft.get_title()
-            txt_file.seek(0)
+
 
         if not self.filename:
             raise forms.ValidationError("Could not extract a valid draft name from the upload"
diff --git a/ietf/submit/parsers/base.py b/ietf/submit/parsers/base.py
index 2c4707b3a..94c928a87 100644
--- a/ietf/submit/parsers/base.py
+++ b/ietf/submit/parsers/base.py
@@ -76,7 +76,7 @@ class FileParser(object):
 
     def parse_file_type(self):
         self.fd.file.seek(0)
-        content = self.fd.file.read(4096)
+        content = self.fd.file.read()
         mimetype = magic.from_buffer(content, mime=True)
         if not mimetype in self.mimetypes:
             self.parsed_info.add_error(u'Expected an %s file of type "%s", found one of type "%s"' % (self.ext.upper(), '" or "'.join(self.mimetypes), mimetype))
diff --git a/ietf/submit/parsers/plain_parser.py b/ietf/submit/parsers/plain_parser.py
index daf948840..4de4ff0b6 100644
--- a/ietf/submit/parsers/plain_parser.py
+++ b/ietf/submit/parsers/plain_parser.py
@@ -21,7 +21,7 @@ class PlainParser(FileParser):
     def parse_file_charset(self):
         import magic
         self.fd.file.seek(0)
-        content = self.fd.file.read(4096)
+        content = self.fd.file.read()
         if hasattr(magic, "open"):
             m = magic.open(magic.MAGIC_MIME)
             m.load()
@@ -31,8 +31,8 @@ class PlainParser(FileParser):
             m.cookie = magic.magic_open(magic.MAGIC_NONE | magic.MAGIC_MIME | magic.MAGIC_MIME_ENCODING)
             magic.magic_load(m.cookie, None)
             filetype = m.from_buffer(content)
-        if not 'ascii' in filetype:
-            self.parsed_info.add_error('A plain text ASCII document must be submitted.')
+        if not 'ascii' in filetype and not 'utf-8' in filetype:
+            self.parsed_info.add_error('A plain text ASCII document is required.  Found an unexpected encoding: "%s"' % filetype)
 
     def parse_name(self):
         self.fd.file.seek(0)