This is fixed now in git.
https://github.com/codenameone/CodenameOne/commit/8927c64972a0f1c838b4401b1bdf475b2560abc6

It will be included in next update on Friday.

Best regards

Steve

On Tue, Sep 4, 2018 at 4:00 PM, Steve Hannah <[email protected]>
wrote:

> Thanks.  I should have a fix for this shortly.
>
> On Tue, Sep 4, 2018 at 3:30 PM, Thomas <[email protected]> wrote:
>
>> OK. I was able to track down the issue. It appears that the 
>> android.webview.chromium
>> previously-failed class messages in logcat where actually unrelated to my
>> problem (the also appear in my working test cases). Actually, the issue is
>> that it seems that now, on android, a BrowserComponent can no longer be
>> used or controlled (to execute some js code) if not displayed. In the
>> simulator this is not the case (the BrowserComponent can still be used if
>> not displayed). I don't know if this is the garbage collector that acts
>> differently on android and on the simulator or if this is some other code
>> in the CN1 code that triggered this change but this is a real issue in my
>> test case. Actually, I am using a BrowserComponent to control a socketio
>> connection with my server (that is used during the whole life time of my
>> app). With the new BrowserComponent behaviour, this is no longer possible...
>> I updated my SocketIOTest test case (https://github.com/ramsestom/
>> CN1SocketIO_test) to test for this issue. If I have the line
>>
>> hi.add(BorderLayout.CENTER, browser);
>>
>> in my main MyAppplication class, everything is working. But If I comment
>> it (so the BrowserComponent handeling the socketio connection is no
>> longer displayed), the socketio connection do not work anymore (and if I
>> have a logger in my BrowserComponent it would fail with a null message like
>> in my first post).
>> Would it be possible to reverse the BrowserComponent behaviour back to
>> what it was a few weeks ago so that it can be used to execute js code in
>> background on any platform supporting it?
>> Thanks
>>
>>
>> On Tuesday, September 4, 2018 at 6:42:46 PM UTC+2, Steve Hannah wrote:
>>
>>> A self-contained test case always helps.  In any case, please file an
>>> issue in the issue tracker so this doesn't get lost.
>>>
>>> On Tue, Sep 4, 2018 at 9:32 AM, Thomas <[email protected]> wrote:
>>>
>>>> Actually, the issue seems to be related to webview that fails to be
>>>> loaded on my android device using CN1.
>>>> I commented the logger part in my code and my app still doesn't work
>>>> properly on my android device with this kind of message in the debugger
>>>> logcat:
>>>>
>>>> I/WebViewFactory: Loading com.google.android.webview version
>>>> 68.0.3440.91 (code 344009150)
>>>> I/art: Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.ServiceWorkerCo
>>>> ntrollerAdapter>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.ServiceWorkerCo
>>>> ntrollerAdapter>
>>>> I/art: Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.TokenBindingMan
>>>> agerAdapter>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.TokenBindingMan
>>>> agerAdapter>
>>>> I/cr_LibraryLoader: Time to load native libraries: 1 ms (timestamps
>>>> 8819-8820)
>>>> E/cr_VariationsUtils: Failed reading seed file
>>>> "/data/data/org.tbdlab.myapp/app_webview/variations_seed_new":
>>>> /data/data/org.tbdlab.myapp/app_webview/variations_seed_new: open
>>>> failed: ENOENT (No such file or directory)
>>>>                       Failed reading seed file "/data/data/org.tbdlab.
>>>> myapp/app_webview/variations_seed": 
>>>> /data/data/org.tbdlab.myapp/app_webview/variations_seed:
>>>> open failed: ENOENT (No such file or directory)
>>>> I/chromium: [INFO:library_loader_hooks.cc(36)] Chromium logging
>>>> enabled: level = 0, default verbosity = 0
>>>> I/cr_LibraryLoader: Expected native library version number
>>>> "68.0.3440.91", actual native library version number "68.0.3440.91"
>>>> I/cr_BrowserStartup: Initializing chromium process, singleProcess=true
>>>> I/cr_base: Android Locale: en_US requires .pak files: []
>>>> I/art: Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.WebResourceErrorAdapter>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.WebResourceErrorAdapter>
>>>> I/art: Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.WebViewContents
>>>> ClientAdapter$6>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.WebViewContents
>>>> ClientAdapter$6>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.SafeBrowsingRes
>>>> ponseAdapter>
>>>>        Rejecting re-init on previously-failed class
>>>> java.lang.Class<com.android.webview.chromium.SafeBrowsingRes
>>>> ponseAdapter>
>>>>
>>>>
>>>> my android test device is using android 5.1.1 and has the latest
>>>> chromium vebview version available for this device (I checked on the
>>>> playstore).
>>>> So I don't know if this issue was triggered by changes you made in CN1
>>>> that may have broke compatibility with some versions of android/webview or
>>>> if this is the latest vebview version that is bugged (my device may have
>>>> performed an autoupdate of webview chromium since last month)
>>>> I can probably make you a self-contained test case if necessary but I
>>>> am not sure it would help as the issue seems to be with webview that can
>>>> not load (so just trying to open a componentbrowser using webview might be
>>>> enaugh as a test)
>>>> What version of vebview runs on your Nexus 5. Is it the latest one?
>>>>
>>>>
>>>>
>>>>
>>>> On Tuesday, September 4, 2018 at 2:15:09 PM UTC+2, Steve Hannah wrote:
>>>>
>>>>> I just copied and pasted your sample into my own test case and it
>>>>> works fine in Android 6 on Nexus 5.  What version of Android are you
>>>>> using?  Can you provide a self-contained test case that will reproduce it?
>>>>>
>>>>> On Mon, Sep 3, 2018 at 10:50 PM, Shai Almog <[email protected]>
>>>>> wrote:
>>>>>
>>>>>> Steve made some performance improvements to JS interconnect which
>>>>>> might have triggered a race condition. We'll look into it.
>>>>>>
>>>>>> --
>>>>>> You received this message because you are subscribed to the Google
>>>>>> Groups "CodenameOne Discussions" group.
>>>>>> To unsubscribe from this group and stop receiving emails from it,
>>>>>> send an email to [email protected]
>>>>>> .
>>>>>> Visit this group at https://groups.google.com/grou
>>>>>> p/codenameone-discussions.
>>>>>> To view this discussion on the web visit
>>>>>> https://groups.google.com/d/msgid/codenameone-discussions/39
>>>>>> 9db7a2-33a8-451d-b788-504b0d86ab92%40googlegroups.com
>>>>>> <https://groups.google.com/d/msgid/codenameone-discussions/399db7a2-33a8-451d-b788-504b0d86ab92%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>>> .
>>>>>>
>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Steve Hannah
>>>>> Software Developer
>>>>> Codename One
>>>>> http://www.codenameone.com
>>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "CodenameOne Discussions" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>> Visit this group at https://groups.google.com/grou
>>>> p/codenameone-discussions.
>>>> To view this discussion on the web visit https://groups.google.com/d/ms
>>>> gid/codenameone-discussions/38d0a677-a0bc-42c4-9b0e-001fe5ad
>>>> 0f68%40googlegroups.com
>>>> <https://groups.google.com/d/msgid/codenameone-discussions/38d0a677-a0bc-42c4-9b0e-001fe5ad0f68%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>>
>>>
>>>
>>> --
>>> Steve Hannah
>>> Software Developer
>>> Codename One
>>> http://www.codenameone.com
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "CodenameOne Discussions" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> Visit this group at https://groups.google.com/grou
>> p/codenameone-discussions.
>> To view this discussion on the web visit https://groups.google.com/d/ms
>> gid/codenameone-discussions/3581ecb7-9a45-44ed-b90f-e8dba540
>> 6be0%40googlegroups.com
>> <https://groups.google.com/d/msgid/codenameone-discussions/3581ecb7-9a45-44ed-b90f-e8dba5406be0%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>
> --
> Steve Hannah
> Software Developer
> Codename One
> http://www.codenameone.com
>



-- 
Steve Hannah
Software Developer
Codename One
http://www.codenameone.com

-- 
You received this message because you are subscribed to the Google Groups 
"CodenameOne Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
Visit this group at https://groups.google.com/group/codenameone-discussions.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/codenameone-discussions/CAGOYrKWby4PHFb3tEHs7wtH_1buvWinjoYBe7yYL24vv3CM2FA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to