WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
Bug 116150
[Curl] Unable to download files.
https://bugs.webkit.org/show_bug.cgi?id=116150
Summary
[Curl] Unable to download files.
peavo
Reported
2013-05-15 06:38:01 PDT
For Curl, the WebDownload class (WebDownloadCurl.cpp) is not implemented.
Attachments
Patch
(20.20 KB, patch)
2013-05-15 07:34 PDT
,
peavo
no flags
Details
Formatted Diff
Diff
Patch
(22.34 KB, patch)
2013-05-29 07:16 PDT
,
peavo
no flags
Details
Formatted Diff
Diff
Patch
(26.48 KB, patch)
2013-05-31 04:44 PDT
,
peavo
no flags
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
peavo
Comment 1
2013-05-15 07:34:10 PDT
Created
attachment 201829
[details]
Patch
peavo
Comment 2
2013-05-29 07:16:51 PDT
Created
attachment 203200
[details]
Patch
peavo
Comment 3
2013-05-29 07:18:25 PDT
Modified the patch slightly by including a crash fix, and creating a new header file.
Brent Fulgham
Comment 4
2013-05-30 22:39:40 PDT
Comment on
attachment 203200
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=203200&action=review
Very nice! I had a few minor style/nit-picky comments, but this looks great. I do feel that we should place the "CurlDownload" class should be in its own set of files. It might make sense for the CurlDownload class to live in WebCore (platform/network/curl).
> Source/WebKit/win/ChangeLog:11 > + (CurlDownloadManager):
I would get rid of all of these method call-outs. The whole file is new, so no point listing them individually.
> Source/WebKit/win/ChangeLog:33 > + (CurlDownloadManager::CurlDownloadManager):
I think this should be in its own file. I know this means we touch the project files, but I think putting them in the same file is worse.
> Source/WebKit/win/WebDownloadCurl.cpp:208 > +// CurlDownloadManager -------------------------------------------------------------------
This should be in its own file.
> Source/WebKit/win/WebDownloadCurl.cpp:458 > + if (!m_destination.isEmpty())
Preference: I really prefer to handle the "no work" case as an early return. I know this method only has one line, but in my mind early return clarifies that if the destination file name is empty, we do nothing. As it's written, it's clear that we don't move the file when the name is empty, but it leaves open the possibility that other work might be done in this method if the file name was empty.
> Source/WebKit/win/WebDownloadCurl.cpp:459 > + MoveFile(m_tempPath.charactersWithNullTermination(), m_destination.charactersWithNullTermination());
Nit: We try to expose Win API calls with the global namespace, i.e., "::MoveFile"
peavo
Comment 5
2013-05-31 04:44:38 PDT
Created
attachment 203435
[details]
Patch
peavo
Comment 6
2013-05-31 05:02:12 PDT
(In reply to
comment #4
)
> (From update of
attachment 203200
[details]
) > View in context:
https://bugs.webkit.org/attachment.cgi?id=203200&action=review
>
Thanks for reviewing! I updated the patch according to your comments. The files CurlDownload.h/cpp were added to the WebCore/platform/network/curl folder.
Brent Fulgham
Comment 7
2013-05-31 22:19:57 PDT
Comment on
attachment 203435
[details]
Patch Looks great! Thanks for making the changes.
WebKit Commit Bot
Comment 8
2013-05-31 22:41:33 PDT
Comment on
attachment 203435
[details]
Patch Clearing flags on attachment: 203435 Committed
r151067
: <
http://trac.webkit.org/changeset/151067
>
WebKit Commit Bot
Comment 9
2013-05-31 22:41:34 PDT
All reviewed patches have been landed. Closing bug.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug