"--enable-file-cookies"?

[Mike West]

I'm poking at a change or two in //net/cookies/ that are somewhat complicated by the fact that we support `file:///` URLs if the `--enable-file-cookies` flag is set.

The command-line flag removal document[1] suggests that we're keeping the flag to support layout tests; AFAIK, the only layout test that requires this functionality is `fast/cookies/local-file-can-set-cookies.html`, which tests that the flag is working. `HistoryBrowserTest.VerifyHistoryLength3` also seems to fail consistently, but I'm sure we could rewrite it (haven't looked yet). The rest of the cookie tests seem to use the HTTP server.

Is there a non-layouttest reason to keep the flag? Will WebView folks be very, very sad? If not, would you folks be very, very sad if I removed it?

[1]: https://docs.google.com/a/chromium.org/spreadsheet/ccc?key=0ArH-w4dboLrgdGRqNU5FUm1Za2VieTFHOE91MW9hUVE&usp=sharing#gid=0 .

-mike

[Chris Bentzel]

+Yuzhu Shen

Looks like PPAPITestBase always sets this flag: https://code.google.com/p/chromium/codesearch#chromium/src/chrome/test/ppapi/ppapi_test.cc&q=kEnableFileCookies&sq=package:chromium&type=cs&l=136

I wouldn't be sad in general

--
You received this message because you are subscribed to the Google Groups "net-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to net-dev+u...@chromium.org.
To post to this group, send email to net...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/net-dev/CAKXHy%3DfTAKzNaEpgN3vQsJhT_w-nPbszYgFAtfB7dFLnzP2Tyg%40mail.gmail.com.

[Mike West]

In a fairly blunt patch (https://codereview.chromium.org/976553002), I removed that line (as well as a few other test-only enablings), and everything continued to pass other than that history test. It looks like WebView might have a few more things going on.

If there aren't fundamental objections, I'll dig into those to see if they're tractable. :)

-mike

-Mike

[Yuzhu Shen]

+CC David.

Sorry for late reply. I was on leave until today.

David and I took a look at the Pepper tests. We haven't used cookies in a while for test results. So it won't be an issue for Pepper.

Thanks for looking into it.

--

Best regards,
Yuzhu Shen.

[Mike West]

Sorry, I meant to follow up on this.

We can't drop the functionality, as it's exposed as part of the WebView API. Torne has suggested that we could deprecate it, but that's going to take some time*. I plan to throw a patch up that would drop the flag from Chrome, but keep the functionality in //net. I haven't done that yet because I'm not sure it's going to be substantial enough to actually commit. :)

-mike

* This is an understatement.

-Mike

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/net-dev/CABxY52Bsb8drwDH29bSx7hyE9aRRWzKzvyX6HTwRh2bw3zJ32w%40mail.gmail.com.

[Adam Rice]

I know from reading bug reports that some web developers use this command-line flag to avoid spinning up a local web server for testing. These people would be sad if the command-line flag went away.

However, from reading bug reports it appears that these people are mostly already sad because their pages behave differently when served by an HTTP server.

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/net-dev/CAAuiYA82x_2rYm%3Dj5xc27panM3nvPeSEZn3CyVDKOwRpHD7TWg%40mail.gmail.com.