Having one would be better, and we already have the ability to capture STDOUT from the server, so I don't think it will be too bad as long as it's easy to reproduce the issue in a test.
On Friday, May 4, 2012 at 2:50 PM, Mark Gibbins wrote:
> I was going to send one over, but I was in two minds about tests. Would you need one for this?
> On Wednesday, 2 May 2012 16:59:37 UTC+1, Joe Ferris wrote:
> > I haven't tried that yet; good find. Care to put together a pull request?
> > On Tuesday, May 1, 2012 at 7:25 PM, Mark Gibbins wrote:
> > > Have you tried defining 'QT_NO_DEBUG_OUTPUT' and 'QT_NO_WARNING_OUTPUT'?
Okay, I've done some more digging around and it seems the debug messages may come out even if you supply those flags when building capybara-webkit if you're using the pre-built binary from homebrew, or simply a debug version. The homebrew formula builds with the configure flag "-with-debug-and-release", making it pop out the warnings no matter what. The solution is to use a release build of Qt, or just build it yourself. With a release build, the flags work and capybara-webkit stays silent. I'm finishing the test(s) now and should have a pull request for you soon.
On Friday, 4 May 2012 21:37:05 UTC+1, Joe Ferris wrote:
> Having one would be better, and we already have the ability to capture > STDOUT from the server, so I don't think it will be too bad as long as it's > easy to reproduce the issue in a test.
> On Friday, May 4, 2012 at 2:50 PM, Mark Gibbins wrote:
> I was going to send one over, but I was in two minds about tests. Would > you need one for this?
> On Wednesday, 2 May 2012 16:59:37 UTC+1, Joe Ferris wrote:
> I haven't tried that yet; good find. Care to put together a pull request?
> On Tuesday, May 1, 2012 at 7:25 PM, Mark Gibbins wrote:
> Have you tried defining 'QT_NO_DEBUG_OUTPUT' and 'QT_NO_WARNING_OUTPUT'?
On Tuesday, May 8, 2012 at 7:27 PM, Mark Gibbins wrote:
> Okay, I've done some more digging around and it seems the debug messages may come out even if you supply those flags when building capybara-webkit if you're using the pre-built binary from homebrew, or simply a debug version. The homebrew formula builds with the configure flag "-with-debug-and-release", making it pop out the warnings no matter what. The solution is to use a release build of Qt, or just build it yourself. With a release build, the flags work and capybara-webkit stays silent. I'm finishing the test(s) now and should have a pull request for you soon.
> On Friday, 4 May 2012 21:37:05 UTC+1, Joe Ferris wrote:
> > Having one would be better, and we already have the ability to capture STDOUT from the server, so I don't think it will be too bad as long as it's easy to reproduce the issue in a test.
> > On Friday, May 4, 2012 at 2:50 PM, Mark Gibbins wrote:
> > > I was going to send one over, but I was in two minds about tests. Would you need one for this?
> > > On Wednesday, 2 May 2012 16:59:37 UTC+1, Joe Ferris wrote:
> > > > I haven't tried that yet; good find. Care to put together a pull request?
> > > > On Tuesday, May 1, 2012 at 7:25 PM, Mark Gibbins wrote:
> > > > > Have you tried defining 'QT_NO_DEBUG_OUTPUT' and 'QT_NO_WARNING_OUTPUT'?
Turns out this is a bit more complex than I laid out above. It seems that even with those flags and a fresh release build of Qt, it still prints the debug messages. It has something to do with the strange (at least to me!) way that the QDebug lib works. Apparently it doesn't really matter what flags you use, it just blurts out these messages. So in the end I had to actually root out the exact message being printed (I discovered it in th QtGui lib as part of text rendering), rewrite all the debug messages in that lib to check for debug flags and then recompile it.
I'm not sure it's much of a problem for Rails developers as it sends the messages to STDOUT, but I was testing a command line client and it was failing the tests. Besides, It's just plain annoying!
If you need more details on the exact edits let me know, as it might fall out of the scope of the project now.
I get the same annoying warnings. Since this bug seems to have been in Qt for years, perhaps there is a way to simply this particular message from the stderr stream? I haven't been successful in getting rid of them yet.
On Thursday, July 12, 2012 at 5:20 AM, Jan Maurits Faber wrote:
> I get the same annoying warnings. Since this bug seems to have been in Qt for years, perhaps there is a way to simply this particular message from the stderr stream? I haven't been successful in getting rid of them yet.
> On Tuesday, May 1, 2012 6:49:04 PM UTC+2, Brandon Hilkert wrote:
> > I'm getting these on the output of my spec request specs...
FWIW, the actual edits to completely silence the warnings are very minimal (a conditional define in one function). What about just overriding the method that prints this message?
On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> FWIW, the actual edits to completely silence the warnings are very minimal (a conditional define in one function). What about just overriding the method that prints this message?
There's an open issue on this on GitHub: https://github.com/thoughtbot/capybara-webkit/issues/485 ... sadly, it references this thread without any new information. This is the only warning I get in my Capbara Webkit integration tests.
On Wednesday, July 18, 2012 6:21:50 PM UTC-7, Joe Ferris wrote:
> Which function are you talking about? Are you suggesting that we override > something from Qt?
> On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> FWIW, the actual edits to completely silence the warnings are very minimal > (a conditional define in one function). What about just overriding the > method that prints this message?
You can suppress a number of warnings by installing a version of Qt that doesn't have DEBUG compiled in. Downloading the official versions (http://qt-project.org/downloads) is one way to do this.
We've also discussed a new feature that would allow people to filter messages by a pattern. I haven't had time to work on it yet, but if anybody's interested in contributing, please email me directly and I'll explain what would be involved in writing a pull request for it.
On Monday, March 18, 2013 at 8:02 PM, Amin Ariana wrote:
> There's an open issue on this on GitHub: https://github.com/thoughtbot/capybara-webkit/issues/485 ... sadly, it references this thread without any new information. This is the only warning I get in my Capbara Webkit integration tests.
> Wonder what others did to suppress this warning successfully in Rails / RSpec?
> On Wednesday, July 18, 2012 6:21:50 PM UTC-7, Joe Ferris wrote:
> > Which function are you talking about? Are you suggesting that we override something from Qt?
> > On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> > > FWIW, the actual edits to completely silence the warnings are very minimal (a conditional define in one function). What about just overriding the method that prints this message?
> -- > You received this message because you are subscribed to the Google Groups "capybara-webkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to capybara-webkit+unsubscribe@googlegroups.com (mailto:capybara-webkit+unsubscribe@googlegroups.com).
> For more options, visit https://groups.google.com/groups/opt_out.
FYI, fwiw I uninstalled my brew version of QT and installed the non-debug
official 4.8 version. A ran my Capybara Webkit tests and I still do get the
"QFont::setPixelSize: Pixel size <= 0 (0)" warning.
On Thu, Mar 21, 2013 at 7:21 AM, Joe Ferris <jfer...@thoughtbot.com> wrote:
> You can suppress a number of warnings by installing a version of Qt that
> doesn't have DEBUG compiled in. Downloading the official versions (
> http://qt-project.org/downloads) is one way to do this.
> We've also discussed a new feature that would allow people to filter
> messages by a pattern. I haven't had time to work on it yet, but if
> anybody's interested in contributing, please email me directly and I'll
> explain what would be involved in writing a pull request for it.
> -Joe
> On Monday, March 18, 2013 at 8:02 PM, Amin Ariana wrote:
> There's an open issue on this on GitHub:
> https://github.com/thoughtbot/capybara-webkit/issues/485 ... sadly, it
> references this thread without any new information. This is the only
> warning I get in my Capbara Webkit integration tests.
> Wonder what others did to suppress this warning successfully in Rails /
> RSpec?
> On Wednesday, July 18, 2012 6:21:50 PM UTC-7, Joe Ferris wrote:
> Which function are you talking about? Are you suggesting that we
> override something from Qt?
> On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> FWIW, the actual edits to completely silence the warnings are very minimal
> (a conditional define in one function). What about just overriding the
> method that prints this message?
> --
> You received this message because you are subscribed to the Google Groups
> "capybara-webkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to capybara-webkit+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
On Thursday, March 21, 2013 at 1:46 PM, Amin Ariana wrote:
> Thanks Joe -
> FYI, fwiw I uninstalled my brew version of QT and installed the non-debug official 4.8 version. A ran my Capybara Webkit tests and I still do get the "QFont::setPixelSize: Pixel size <= 0 (0)" warning.
> On Thu, Mar 21, 2013 at 7:21 AM, Joe Ferris <jfer...@thoughtbot.com (mailto:jfer...@thoughtbot.com)> wrote:
> > You can suppress a number of warnings by installing a version of Qt that doesn't have DEBUG compiled in. Downloading the official versions (http://qt-project.org/downloads) is one way to do this.
> > We've also discussed a new feature that would allow people to filter messages by a pattern. I haven't had time to work on it yet, but if anybody's interested in contributing, please email me directly and I'll explain what would be involved in writing a pull request for it.
> > -Joe
> > On Monday, March 18, 2013 at 8:02 PM, Amin Ariana wrote:
> > > There's an open issue on this on GitHub: https://github.com/thoughtbot/capybara-webkit/issues/485 ... sadly, it references this thread without any new information. This is the only warning I get in my Capbara Webkit integration tests.
> > > Wonder what others did to suppress this warning successfully in Rails / RSpec?
> > > On Wednesday, July 18, 2012 6:21:50 PM UTC-7, Joe Ferris wrote:
> > > > Which function are you talking about? Are you suggesting that we override something from Qt?
> > > > On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> > > > > FWIW, the actual edits to completely silence the warnings are very minimal (a conditional define in one function). What about just overriding the method that prints this message?
> > > -- > > > You received this message because you are subscribed to the Google Groups "capybara-webkit" group.
> > > To unsubscribe from this group and stop receiving emails from it, send an email to capybara-webkit+unsubscribe@googlegroups.com (mailto:capybara-webkit+unsubscribe@googlegroups.com).
> > > For more options, visit https://groups.google.com/groups/opt_out.
> > -- > > You received this message because you are subscribed to a topic in the Google Groups "capybara-webkit" group.
> > To unsubscribe from this topic, visit https://groups.google.com/d/topic/capybara-webkit/n1xMK0HnMWo/unsubsc....
> > To unsubscribe from this group and all its topics, send an email to capybara-webkit+unsubscribe@googlegroups.com (mailto:capybara-webkit%2Bunsubscribe@googlegroups.com).
> > For more options, visit https://groups.google.com/groups/opt_out.
> -- > You received this message because you are subscribed to the Google Groups "capybara-webkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to capybara-webkit+unsubscribe@googlegroups.com (mailto:capybara-webkit+unsubscribe@googlegroups.com).
> For more options, visit https://groups.google.com/groups/opt_out.
I uninstalled and reinstalled the gem, still get the same warning. Here is
the output of the command you suggested after the reinstall (and I double
checked that in between, the webkit_server directory was empty) :
QtWebKit.framework/Versions/4/QtWebKit (compatibility version 4.9.0,
current version 4.9.3)
QtGui.framework/Versions/4/QtGui (compatibility version 4.8.0, current
version 4.8.4)
QtCore.framework/Versions/4/QtCore (compatibility version 4.8.0, current
version 4.8.4)
QtNetwork.framework/Versions/4/QtNetwork (compatibility version 4.8.0,
current version 4.8.4)
/usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current version
52.0.0)
/usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version
1105.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version
159.1.0)
Here's the qmake / QT version:
qmake -v
QMake version 2.01a
Using Qt version 4.8.4 in /Library/Frameworks
This is on Mac OSX Lion if that matters. Also I do have QT 5 installed in
parallel to 4.8, but doesn't look like it's been used. I'm slightly
surprised by the version that QtWebKit is listing (4.9).
On Thu, Mar 21, 2013 at 10:50 AM, Joe Ferris <jfer...@thoughtbot.com> wrote:
> If you install a new version of Qt, you'll also need to recompile
> capybara-webkit by uninstalling and reinstalling.
> You can figure out which version of Qt capybara-webkit is using with this
> command on OS X:
> otool -L $(bundle show capybara-webkit)/bin/webkit_server
> I believe you can use `ldd` instead of `otool -L` on Linux.
> -Joe
> On Thursday, March 21, 2013 at 1:46 PM, Amin Ariana wrote:
> Thanks Joe -
> FYI, fwiw I uninstalled my brew version of QT and installed the non-debug
> official 4.8 version. A ran my Capybara Webkit tests and I still do get the
> "QFont::setPixelSize: Pixel size <= 0 (0)" warning.
> On Thu, Mar 21, 2013 at 7:21 AM, Joe Ferris <jfer...@thoughtbot.com>wrote:
> You can suppress a number of warnings by installing a version of Qt that
> doesn't have DEBUG compiled in. Downloading the official versions (
> http://qt-project.org/downloads) is one way to do this.
> We've also discussed a new feature that would allow people to filter
> messages by a pattern. I haven't had time to work on it yet, but if
> anybody's interested in contributing, please email me directly and I'll
> explain what would be involved in writing a pull request for it.
> -Joe
> On Monday, March 18, 2013 at 8:02 PM, Amin Ariana wrote:
> There's an open issue on this on GitHub:
> https://github.com/thoughtbot/capybara-webkit/issues/485 ... sadly, it
> references this thread without any new information. This is the only
> warning I get in my Capbara Webkit integration tests.
> Wonder what others did to suppress this warning successfully in Rails /
> RSpec?
> On Wednesday, July 18, 2012 6:21:50 PM UTC-7, Joe Ferris wrote:
> Which function are you talking about? Are you suggesting that we
> override something from Qt?
> On Saturday, July 14, 2012 at 6:11 AM, Mark Gibbins wrote:
> FWIW, the actual edits to completely silence the warnings are very minimal
> (a conditional define in one function). What about just overriding the
> method that prints this message?
> --
> You received this message because you are subscribed to the Google Groups
> "capybara-webkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to capybara-webkit+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
> --
> You received this message because you are subscribed to the Google Groups
> "capybara-webkit" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to capybara-webkit+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
