[Webkit-unassigned] [Bug 168195] New: [GTK] Critical warnings in /webkit2/WebKitWebView/session-state-with-form-data

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Sun Feb 12 11:42:41 PST 2017


https://bugs.webkit.org/show_bug.cgi?id=168195

            Bug ID: 168195
           Summary: [GTK] Critical warnings in
                    /webkit2/WebKitWebView/session-state-with-form-data
    Classification: Unclassified
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: WebKit Gtk
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: mcatanzaro at igalia.com
                CC: bugs-noreply at webkitgtk.org, svillar at igalia.com

Created attachment 301312
  --> https://bugs.webkit.org/attachment.cgi?id=301312&action=review
Full backtrace

A couple criticals occur in the API test /webkit2/WebKitWebView/session-state-with-form-data:                 

(WebKitWebProcess:31443): GLib-CRITICAL **: g_string_append: assertion 'val != NULL' failed



(WebKitWebProcess:31443): GLib-CRITICAL **: g_string_append: assertion 'val != NULL' failed

OK

Anyway, I got a backtrace and it seems to be a bug with the form-controls-associated signal handler:

#3  0x00007f6be3796c77 in g_return_if_fail_warning (
    log_domain=0x7f6be37f44bf "GLib", 
    pretty_function=0x7f6be37f45d0 <__func__.10999> "g_string_insert_len", 
    expression=0x7f6be37f44d1 "len == 0 || val != NULL")
    at /home/mcatanzaro/Projects/GNOME/glib/glib/gmessages.c:2687
No locals.
#4  0x00007f6be37b3e1d in g_string_insert_len (string=0x2275c80, pos=-1, 
    val=0x0, len=-1) at /home/mcatanzaro/Projects/GNOME/glib/glib/gstring.c:430
        __func__ = "g_string_insert_len"
#5  0x00007f6be37b4341 in g_string_append (string=0x2275c80, val=0x0)
    at /home/mcatanzaro/Projects/GNOME/glib/glib/gstring.c:596
No locals.
#6  0x00007f6b947f5fcd in formControlsAssociatedCallback (webPage=0x226a360, 
    formElements=0x22b3460, extension=0x21a4ba0)
    at ../../Tools/TestWebKitAPI/Tests/WebKit2Gtk/WebExtensionTest.cpp:265
        domElement = 0x216d590
        i = 0
        formIdsBuilder = 0x2275c80
        __PRETTY_FUNCTION__ = "void formControlsAssociatedCallback(WebKitWebPage*, GPtrArray*, WebKitWebExtension*)"
        formIds = std::unique_ptr<char> containing 0x7ffd5f3162c0 ""
        data = 0x200216bc80

Note that the web process criticals do not cause the test to fail. That's bad! Even a non-critical warning in the UI process would be sufficient to cause our tests to fail, but this bug went undetected because it occurred in the web process. We should probably run all the tests with G_DEBUG=fatal-warnings to ensure this works for the web process in addition to the UI process.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170212/60baea23/attachment.html>


More information about the webkit-unassigned mailing list