Fix crash on incorrectly formatted MHTML documents

Fix crash on incorrectly formatted MHTML documents

On incorrectly formatted MHTML documents, MHTMLArchive::create() can return
null or a MHTMLArchive object whose main resource is null. We had a release
assertion to make sure MHTMLArchive::create() does not return null and no
null check for the main resource. This was causing us to crash in both cases.

This patch adds both null checks and handles them nicely.
This patch is based on initial work from Laszlo Gombos <l.gombos@samsung.com>;.

R=japhet@chromium.org

BUG=331898
TEST=mhtml/invalid-bad-boundary.mht
TEST=mhtml/invalid-bad-boundary2.mht

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=164584

[Nate Chapin, 2014-01-06 23:04:10]

LGTM. Feel free to ignore the comment if you disagree :)

https://codereview.chromium.org/125403003/diff/1/Source/core/loader/DocumentL...
File Source/core/loader/DocumentLoader.cpp (right):

https://codereview.chromium.org/125403003/diff/1/Source/core/loader/DocumentL...
Source/core/loader/DocumentLoader.cpp:289: if
(!isArchiveMIMEType(m_response.mimeType()) || !createArchive()) {
Should we consider changing this function's name back to maybeCreateArchive()
and move the isArchiveMIMEType() check into it?

That's how this code was a few months ago. I'm debating which one is more
readable.

[Inactive, 2014-01-07 00:38:16]

https://codereview.chromium.org/125403003/diff/1/Source/core/loader/DocumentL...
File Source/core/loader/DocumentLoader.cpp (right):

https://codereview.chromium.org/125403003/diff/1/Source/core/loader/DocumentL...
Source/core/loader/DocumentLoader.cpp:289: if
(!isArchiveMIMEType(m_response.mimeType()) || !createArchive()) {
On 2014/01/06 23:04:11, Nate Chapin wrote:
> Should we consider changing this function's name back to maybeCreateArchive()
> and move the isArchiveMIMEType() check into it?
> 
> That's how this code was a few months ago. I'm debating which one is more
> readable.

Done.

[commit-bot: I haz the power, 2014-01-07 09:59:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/125403003/270001

[commit-bot: I haz the power, 2014-01-07 09:59:05]

Failed to apply patch for LayoutTests/mhtml/invalid-bad-boundary2.mht:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  A         LayoutTests/mhtml/invalid-bad-boundary2.mht
  Copied LayoutTests/mhtml/content_transfer_encoding_none.mht ->
LayoutTests/mhtml/invalid-bad-boundary2.mht
  patching file LayoutTests/mhtml/invalid-bad-boundary2.mht
  Hunk #1 FAILED at 6.
  1 out of 1 hunk FAILED -- saving rejects to file
LayoutTests/mhtml/invalid-bad-boundary2.mht.rej

Patch:   NR 
LayoutTests/mhtml/content_transfer_encoding_none.mht->LayoutTests/mhtml/invalid-bad-boundary2.mht
Index: LayoutTests/mhtml/invalid-bad-boundary2.mht
diff --git a/LayoutTests/mhtml/content_transfer_encoding_none.mht
b/LayoutTests/mhtml/invalid-bad-boundary2.mht
similarity index 93%
copy from LayoutTests/mhtml/content_transfer_encoding_none.mht
copy to LayoutTests/mhtml/invalid-bad-boundary2.mht
index
c43f8399b9ec9ec71f4e706ef87739bf8569d5d8..fd9d657fa7deebb9878ff9249d0814dd47801098
100644
--- a/LayoutTests/mhtml/content_transfer_encoding_none.mht
+++ b/LayoutTests/mhtml/invalid-bad-boundary2.mht
@@ -6,7 +6,7 @@ Content-Type: multipart/related;
 	boundary="----=_Part_ADD9911_9D32DCC.1376472064834";
 	type="text/html"
 
-------=_Part_ADD9911_9D32DCC.1376472064834
+-----=_Part_ADD9911_9D32DCC.1376472064834
 Content-Location: http://localhost/content_transfer_encoding_none.html
 Content-Type: text/html;
 	charset="UTF-8"

[commit-bot: I haz the power, 2014-01-07 13:26:38]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/125403003/340001

[commit-bot: I haz the power, 2014-01-07 14:31:59]

Change committed as 164584