[webkit-reviews] review denied: [Bug 191645] [Curl] Implement Cookie Accept Policy. : [Attachment 362484] Patch

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Feb 20 02:22:19 PST 2019


Fujii Hironori <Hironori.Fujii at sony.com> has denied Takashi Komori
<Takashi.Komori at sony.com>'s request for review:
Bug 191645: [Curl] Implement Cookie Accept Policy.
https://bugs.webkit.org/show_bug.cgi?id=191645

Attachment 362484: Patch

https://bugs.webkit.org/attachment.cgi?id=362484&action=review




--- Comment #19 from Fujii Hironori <Hironori.Fujii at sony.com> ---
Comment on attachment 362484
  --> https://bugs.webkit.org/attachment.cgi?id=362484
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=362484&action=review

> Source/WebCore/platform/network/curl/CookieJarCurl.cpp:122
> +    if (auto cookies = cookieJarDB.searchCookies(firstParty, firstParty,
WTF::nullopt, WTF::nullopt, WTF::nullopt)) {

Why don't you pass the fourth argument (URL) to the second parameter of
searchCookies?
Is there a test case for this?

> Source/WebCore/platform/network/curl/CookieJarDB.cpp:333
> +    String urlHost = url.host().toString().convertToASCIILowercase();

This can be "url.host().convertToASCIILowercase()".

> Source/WebCore/platform/network/curl/CookieJarDB.cpp:346
> +	       statement.bindText(2, String("*.") + topPrivateDomain);

I think "*.example.com" doesn't match "example.com". Is this OK?

> Source/WebCore/platform/network/curl/CookieJarDB.cpp:362
> +    String requestHost =
requestUrl.host().toString().convertToASCIILowercase();

This can be requestUrl.host().convertToASCIILowercase().

> Source/WebCore/platform/network/curl/CookieJarDB.cpp:471
> +    if (!checkCookieAcceptPolicy(firstParty, cookieDomainUrl))

Do you need make a new URL? How about checkCookieAcceptPolicy(firstParty, url)?

> Source/WebCore/platform/network/curl/CookieUtil.cpp:93
> +    if (index != WTF::notFound && index + firstPartyDomain.length() ==
urlHost.length())

This doesn't seem right. "example.com" can match "xexample.com".


More information about the webkit-reviews mailing list