Tweaked comments related to mime boundary marker generation.

third_party/WebKit/Source/platform/network/FormDataEncoder.cpp

 /*
  * Copyright (C) 1999 Lars Knoll (knoll@kde.org)
  *           (C) 1999 Antti Koivisto (koivisto@kde.org)
  *           (C) 2001 Dirk Mueller (mueller@kde.org)
  * Copyright (C) 2004, 2005, 2006, 2007, 2008 Apple Inc. All rights reserved.
  *           (C) 2006 Alexey Proskuryakov (ap@nypop.com)
  * Copyright (C) 2008 Torch Mobile Inc. All rights reserved. (http://www.torchmobile.com/)
  *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Library General Public
@@ skipping 81 matching lines @@
     if (charset.isEmpty()) {
         if (defaultCharset.isEmpty())
             return WTF::UTF8Encoding();
 
         return defaultCharset;
     }
 
     return charset;
 }
 
-// TODO(lukasza): Reuse net::GenerateMimeMultipartBoundary instead.

[Łukasz Anforowicz, 2016/01/08 21:31:12]

Hmmm... ideally we would have only 1 piece of code for generating the mime
boundary.  I understand that we don't want Blink to be able to issue network
requests without going through the browser process + because of repo history
don't even allow blink -> base dependencies.  OTOH, it sounds that some blink ->
base dependencies will be eventually allowed and therefore maybe we just need to
move GenerateMimeMultipartBoundary into base?

WDYT?  Should I remove the "reuse" TODO or leave it here to see what happens
after Blink moves around and starts allowing some Chromium dependencies (like
base)?

[Ryan Sleevi, 2016/01/08 21:34:22]

No, we don't want base/ to be the dumping ground for what is unquestionably a
network layer dependency :)

The story of Blink and //net is even more complicated than the //base bit,
that's why I objected to the TODO - because it's not at all clear that this is a
good / right direction, but the TODO when read naturally suggests this is the
"decided decision"
Wholly remove.

 Vector<char> FormDataEncoder::generateUniqueBoundaryString()
 {
     Vector<char> boundary;
 
+    // TODO(rsleevi): crbug.com/575779: Follow the spec or fix the spec.
     // The RFC 2046 spec says the alphanumeric characters plus the
     // following characters are legal for boundaries:  '()+_,-./:=?
     // However the following characters, though legal, cause some sites
     // to fail: (),./:=+
+    //
     // Note that our algorithm makes it twice as much likely for 'A' or 'B'
     // to appear in the boundary string, because 0x41 and 0x42 are present in
     // the below array twice.
     static const char alphaNumericEncodingMap[64] = {
         0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48,
         0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50,
         0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58,
         0x59, 0x5A, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66,
         0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E,
         0x6F, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76,
@@ skipping 96 matching lines @@
         } else if (c == '\n' || (c == '\r' && (i + 1 >= length || string.data()[i + 1] != '\n'))) {
             append(buffer, "%0D%0A");
         } else if (c != '\r') {
             append(buffer, '%');
             appendByteAsHex(c, buffer);
         }
     }
 }
 
 }