I posted a few threads the other day on the Watir General Google Group, one of which detailed my difficulty using set_no_wait on a validated text field (link below); and was asked by Jarmo Pertman whether it was in a frame. It is, and this concerns me a little; as sadly the site I'm testing is heavily frame-centric, with no immediate plans to modernise.
I recently migrated from an ageing Ruby1.8.6/Watir1.6.2 test environment to Ruby1.8.7-334/Watir2.0.1, and am now noticing more methods that are no longer working for me in frames. For instance, click_no_wait on a button that presents a new window works perfectly fine in my old environment, but not in my new one.
This is a question to the Watir developer group. Are no_wait methods known to be failing in frames in Watuir 2.0.1? If so, are they being addressed? If they're not, then I'll be forced to revert back to an archiac version of Ruby and Watir. For many reasons, I don't want to do this.
(Environment: Ruby 1.8.7-334. Watir 2.0.1. XP Pro. IE8)
http://groups.google.com/group/watir-general/browse_thread/thread/85a1a872d1e054dc
from the comments:
Something occured to me last night; I had installed DevKit, which was required to get the FastDebugger working in Netbeans 7.0.1. After completely reinstalling Ruby1.8.7/Watir2.0.1 without DevKit, click_no_wait and set_no_wait now seems to be working for me. It means I'll have to use something else other than Netbeans to run my scripts, but at least my scripts work again. (github.com/oneclick/rubyinstaller/wiki/Development-Kit) I hope this info helps someone else. GJHmf – GJHmf Sep 1 at 10:19
(inserting this answer because:
the OP was requested to but had not done so within a months time.
it removes this from the list of 'un-answered' watir related
questions
)
Related
ALL,
I'm trying to develop my app on Windows 8.1 with MSVC 2017.
I recently updated it (the original install was couple of years old) and now when I try to run it from the IDE I get an error:
verifier.pdb is not loaded.
What is this and why I didn't get it from the original version?
I presume that this verifier.dll is a good thing and will help me with the development in the long run, but right now it prevents me from debugging my application.
Could someone please help.
TIA!!
P.S.: I'm adding application-verifier tag, but I'm not sure that it is related. I apologize in advance if its not.
EDIT:
That question is related but the answer{s} there does NOT explain how to remove the error I'm seeing.
Moreover the linked question does not see the issue I'm having.
So those 2 - mine and the linked one - are NOT duplicates.
Also, I'd appreciate if the question I'm posting is not closed immediately - post the CONSIDERED dup and give me a time to respond.
On top of that the link referenced in one the comment still doesn't work and the updated one does NOT say how to cancel the specific error I'm seeing.
And finally - the lionked question was talking about debugging the application with the RT.
I am starting the application inside MSVC and so I don't have ANY means to continue running the application, while the presumed OP seems to imply the application was running fine after starting and therefore was able to perform the task semi-successfully.
TIA!!
It is application verifier.
I am guessing this is the original issue - Mysterious verifier.dll
I am working through the pluralsight videos on MonoTouch. At one point, the trainer right clicks on the name of a derived class, and in the 'refactor' menu there is a function to override/implement members of that class. When I click however (latest version), I see only 'rename.'
The person in this link had the same issue some time ago and has included screenshots - but noone replied to them in the MonoTouch discussion group:
http://monotouch.2284126.n4.nabble.com/Right-Click-Class-name-gt-Refactor-gt-Override-Implement-members-tt4655504.html#none
Has anyone experienced (and resolved!) this?
Some of the refactoring features were reorganized or removed (for now) in the rewrite of the code code completion engine that took place for MD 3.0.
You can still access this particular feature two ways:
1) After typing the "override" keyword, MD offers the members you can override/implement. Selecting one will cause it to be stubbed out.
2) You can override many members at once using the "Edit->Show Code Generation Window" command in the class body. This command doesn't have a keybinding on Mac by default, but you can assign one in Preferences.
MonoDevelop 3.0 (and later) removed some features (including a few about refactoring) since they were not as stable, fully functional (complete) or buggy.
The same features (or similar ones) are likely to come back in future releases.
I have a multi-form Windows mobile application that accesses the barcode scanner of the MC75 using the EMDK (version 2.6). I access the barcode scanner using a Singleton. When the application starts I call barcodereader.Actions.Enable() and when the application exits I call `barcodereader.Actions.Disable()'. Basically I've used the EMDK samples as a guide to get the scanner working.
While the application is running I add the event handler to the barcode reader `readNotify' on the form activation event and remove the event handler when the form activates. I have put code in to check that this is happening correctly and added a check to ensure no existing event handlers are present before a new one is added.
The application works fine but for some reason the barcode scanner suddenly cuts out and will not work when the trigger button is pressed. This always seems to happen in the same place (after around 20 odd scans spread across the three forms). No error is caught and the scanner light fails to come on. I have a set number of scans to do and the scanning stops on a form where I've done a successful scan seconds earlier. I can do more than 20 scans on a single form but I only seem to get the issue when doing scanning on different forms.
If I restart the application the scanner works again. The datawedge software is installed but is not used.
I have seen this issue during my research here, but my issue does not seem to be intermittent.
I saw a note on the Symbol support site stating it was not desirable to repeatedly call `.Actions.Enable' and '.Actions.Disable' when moving between forms and I don't do this.
I'm at a loss as to what to check in the application. Any ideas?
Details - using Windows Visual Studio 2008, device is running Windows Mobile 6.5 Professional, EMDK version 2.6, application written in C#.
I had this exact same issue on an MC70 device. Unfortunately it was at a previous employer so I no longer have the source code to look back on. I do remember having to create a somewhat ugly workaround to get it working.
I know that I ended up declaring/instantiating the barcodereader object inside of the parent form (so no need for the singleton class). Then every time I instantiated a child form, I passed either the parent form object or the barcodereader object BY REFERENCE into the child form. Then I was able to access it from the child form repeatedly without it crashing/locking up. I know that is a pretty vague description, but it's been a couple of years since I worked on that code. Hope this helps.
I knew this problem years ago!. The solution to that is: DON'T disable the barcode reader when doing form navigation (Keep it always Active ). It works perfectly for me (I did MC9062 6 years ago and now MC75A device). Just put some logic to tell your program to handle scanned data differently under different form.
Yes, it is important to use reader object as "Singleton".
I have some serious problems testing a sharepoint site with selenium/bromine. As I did't find an answer via various searches I hope someone here can point me in the right direction.
I am constantly getting timeouts opening the main page, but the server is definetly fast enough to answer the request and at 90% idle. Nevertheless I just get logs like these:
open http://username:passwd#10.13.110.54/default.aspx | Timed out after 90000ms
Test terminated The selenium server did not return OK
The auth popup is popping up at irregular intervals (every 5 to 10 clicks) although every open command uses the http://username:passwd#10.13.110.54/ as prefix
Clicking on elements is sometimes not registered, the logs show a successful
isElementPresent link=myLink
click link=myLink
but the browser doesn't react. These are mainly in-page links which open a new folder or an editing box.
I'm not sure whether I should have posted the in three separate questions, but I didn't want to spam.
Hope someone can help me, as I have these problems now for nearly 3 weeks.
Thanks in advance
Thomas
For your question number 2: Okay, this is a really late reply. I stumbled on this page looking for the answer myself. Given that I have solved it in the meantime, I figured I'd post my answer for other people stumbling onto this page.
General solution:
You need to create or use a profile that will let firefox automatically forward your credentials to the sharepoint website. You can create the profile manually and call it each time, see https://applicationtestingtips.wordpress.com/2009/12/21/seleniumrc-handle-windows-authentication-firefox/ for instructions.
Programmer solution: (works in python, should work similarly in Java)
Or you can create a new profile on the fly each time. I did that based on the information in the previously mentioned website. I use python for calling selenium, but this should be rather similar in whatever language you use to call selenium:
sharepointHosts = 'sharepoint1.mycompany.com,sharepoint2.mycompany.com' #have all your sharepoint hosts here in a comma-separated list
ffProfile.set_preference('network.automatic-ntlm-auth.trusted-uris', sharepointHosts)
ffProfile.set_preference('network.negotiate-auth.delegation-uris', sharepointHosts)
ffProfile.set_preference('network.negotiate-auth.trusted-uris', sharepointHosts)
driver = webdriver.Firefox(firefox_profile=ffProfile)
I see this referenced a lot: http://ajaxify.com/run/crossframe/
And I noticed now it's no longer working for the IFRAME (child) to change the parents hash.
I've been spending a few hours trying various things wondering why this isn't working anymore -- then I finally realized that the example I originally based it on was down too.
Can anybody confirm?
-
-
I need to use a cross domain iframe to take care of an order upload form that our shopping cart doesn't support, and I need the form to return an order ID to the parent, so that I can associate the data between the two servers.
Any recommendations or directions to head in would be appreciated.
I'm not looking for a shortcut or somebody to do my work. I've been reading all day... I just need a nudge in the right direction.
Thanks!
It is partly browser-specific. For instance, some browsers don't allow a nested child frame (iframe within an iframe) to change its parent's fragment. See this blog entry. But the basic functionality still works in recent browsers. This demo, which has an iframe change its parent, and vice versa (single level) works fine in Firefox 3.5.9 and Chrome 5.0.375.99.
The demo you gave also works both ways in that version of Firefox. It doesn't allow the child to change the parent in Chrome. The main difference seems to be that the working one uses parent.location, while the broken one uses parent.window.location.hash.
The best solution for recent browsers is postMessage. If necessary, you can also use a server proxy.