Workbench whining about not being able to instantiate a Browser

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Workbench whining about not being able to instantiate a Browser

spok
Dear all,

using carrot2 workbench 3.9.4 or older versions it´s not possible for me to open html files in the results list by clicking on them (Windows 7).

Reading the FAQ - last topic - I understood that I need a Webkit based browser to succeed.

I therefore installed Midori and configured it as external browser to use, but there is no difference.

What I´m doing wrong?

Many thanks for any hint.

Regards

spok
Reply | Threaded
Open this post in threaded view
|

Re: Workbench whining about not being able to instantiate a Browser

Dawid Weiss
> using carrot2 workbench 3.9.4 or older versions it´s not possible for me to
> open html files in the results list by clicking on them (Windows 7).

What's the symptom of "impossible"? Make sure that your Java version
is 32 bit or 64-bit (matching the system), then the browser component
(IE) should load just fine.

> Reading the FAQ - last topic - I understood that I need a Webkit based
> browser to succeed.

Not on Windows. On Windows you should stick with IE. You can also try
to diagnose what's wrong by opening the "browser" view in Workbench:
Window->Show View->Other, then General->Internal Web Browser.

If it opens, try to detect which IE component version your Workbench
is using by typing in one of the numerous browser-identification sites
(https://www.whatismybrowser.com/ or the like). It is quite
unfortunate but the version of the embedded browser control is global
on Windows, so if some other software required an IE7, for example,
then Workbench won't work. You can check out your current settings in
Windows registry at:

HKEY_CURRENT_USER\Software\Microsoft\Internet
Explorer\Main\FeatureControl\FEATURE_BROWSER_EMULATION

If there's a key "javaw" then remove it and try again.

Dawid

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers
Reply | Threaded
Open this post in threaded view
|

Re: Workbench whining about not being able to instantiate a Browser

spok
In reply to this post by spok
David,

thanks for the reply.

With "impossible" I meant that clicking on a result causes just nothing ...

But: I realized that there is a difference between html files on the web and on the filesystem ...

I set Preferences to "use internal browser" and its no problem to open websites. So far so good.

What I tried was: index html files on a local Windows file system with lucene and then using the lucene index to point to results. Works of course, but these results are listed, cannot be opened however.

Perhaps I have to set up a local webserver to point to them?

Regards

spok

BTW:

HKEY_CURRENT_USER\Software\Microsoft\Internet
Explorer\Main\FeatureControl\FEATURE_BROWSER_EMULATION

A key "javaw" has been there ...
Reply | Threaded
Open this post in threaded view
|

Re: Workbench whining about not being able to instantiate a Browser

Dawid Weiss
> What I tried was: index html files on a local Windows file system with
> lucene and then using the lucene index to point to results. Works of course,
> but these results are listed, cannot be opened however.

I don't think it's the browser component's problem then.

You need to index documents with an URL that can be opened with a
system browser. In Java the filesystem URIs have a different format
(file://c:/...) than what IE recognizes (file:///c:/foo). At least I
think, because I cannot verify this at the moment.

You can set up a web server and then index the documents with an URL
pointing to that web server, this would probably be more fool-proof
than trying to use local filesystem URIs.

Dawid

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Carrot2-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/carrot2-developers