Login
Username:

Password:

Remember me



Lost Password?

Register now!

Sections

Who's Online
15 user(s) are online (10 user(s) are browsing Forums)

Members: 0
Guests: 15

more...

Support us!

Headlines

 
  Register To PostTopic is Locked  

« 1 (2) 3 »
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

Quote:

joerg wrote:
@LiveForIt

Quote:
1)
As I found out, many CmakeFiles where not correct, also files where missing in the sand-labs svn,
There never were any CMake files for the AmigaOS 4.x port from me because I don't use it. The broken CMake files currently on the OWB SVN are from afxgroup.
There is nothing missing in the AmigaOS 4.x parts, they are just several months old and of course it can't work if you use newer versions of the rest of OWB.


So what do you use to build it then? LiveForIt has just spent weeks getting the SVN version to build using the CMakefiles. If there's a faster way then I'm sure that he would like to hear about it.

Quote:
Quote:
3)
Sand-labs also only accepted diff files.
Enforced code review is a good thing, even if it failed for the broken CMake files from afxgroup.


If you want to keep your independant AmigaOS 4.x port of OWB outside of the official OWB SVN like the AROS and MorphOS porters do you wont get any help from me at all, and I wont include changes from you in my versions of OWB.


That's rather harsh.

I have no idea what LiveForIt's intentions are but, given the slow review process for commits to the main SVN repository, I'd also keep my own private repository for local changes (if I were working on the OWB port). Naturally, I'd make sure that I regularly submit diffs to the master repository, but there is great value in having a local one. It's the difference between losing just a day's work, and losing over a month's worth of work. Plus, I have found rolling back or reviewing the diffs over the last week has been very useful with other projects; that's something that isn't possible unless you're able to commit small changes to the repository.

An alternative would be to have a private branch within the master SVN repository, but I doubt that the OWB team would be okay with that.

Hans

Join Kea Campus' Amiga Corner and support Amiga content creation
https://keasigmadelta.com/ - see more of my work
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

I'm open for ideas, we now how at least 3 or 4 people interested in work on this for AmigaOS4, we need to organize thing so that source code is available to every one.

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@Hans

Quote:
So what do you use to build it then?
I'm building OWB on AmigaOS 4.1, since there is no CMake port (and the AmigaOS ports of Perl and Python can't be used for the scripts used by OWB either) I update the makefiles manually.

Quote:
LiveForIt has just spent weeks getting the SVN version to build using the CMakefiles.
I guess he tried to use the broken AmigaOS 4.x parts from afxgroup in the CMakefiles as base instead of starting from scatch, he mixed old AmigaOS 4.x parts with much newer versions of other parts of OWB, which can't work, and IIRC he even put some parts of the MorphOS port into it. No wonder it took that long with such a mess. Additionally he tries to cross-build it using an unsupported system as host, even a native Cygwin build of OWB isn't possible without changes, which might have caused additional problems.
Using AmigaOS 4.x parts with the matching revision of the rest of the OWB sources, and using a supported system as host for the cross-build (for example Linux), it should have taken only a few minutes to add AmigaOS 4.x support to the CMakefiles.

Quote:
given the slow review process for commits to the main SVN repository
IIRC it took about 3 days on average to get my AmigaOS 4.x changes reviewed by a Pleyo employee and submitted to the SVN. For faster commits at least 2 AmigaOS 4.x developers would have been required.
I have reviewer and committer permissions on sand-labs.org, but of course I'm not allowed to review and commit my own changes

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@LiveForIt

Quote:
I'm open for ideas, we now how at least 3 or 4 people interested in work on this for AmigaOS4, we need to organize thing so that source code is available to every one.
As long as the sand-labs OWB SVN is working I wont switch to something else.

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@LiveForIt

Quote:
4)
Keep the source code only on your hard-drive is not smart.


I never say that, at that point it could be better to send your diffs directly to Joerg, then eventually he can upload them on Sand-Labs SVN as he have access to it

BTW, what is the scope to fork it to make another Reaction release ?

if you have choosed Reaction it could be logic to work side-by-side with Joerg, conversely there is a mutch better version, (the MUI one) that wait only to be compiled and adapted on OS4

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@samo79

And Jurge is going to send his diff files to me etc?

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

You have not uploaded anything to main SVN for years what are you taking about.

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

Quote:
I update the makefiles manually


Well your makefiles are unobtainable (on your broken Amiga),
and I don't know hot edit the makesfiles.

Quote:
I guess he tried to use the broken AmigaOS 4.x parts from afxgroup in the CMakefiles as base instead of starting from scatch


No I don't wont start from scratch, I wont continue the work.

Quote:
he mixed old AmigaOS 4.x parts with much newer versions of other parts of OWB


I have updated the class manually, all AmigaOS4 dependent parts.
Again the newer ones are on your broken Amiga so they are unobtainable.

Quote:
No wonder it took that long with such a mess.


I'm not done yet, just found out used the wrong compile option.

Quote:
Additionally he tries to cross-build it using an unsupported system as host, even a native Cygwin build of OWB isn't possible without changes, which might have caused additional problems.


I spent long time getting this to work, I have no doubt it works.

Quote:
it should have taken only a few minutes to add AmigaOS 4.x support to the Cmakefiles.


Well maybe, if I know how the CMAKE system worked before starting on this project even so, it takes time find out whats broken and not.

It takes 30min to check if every things work, that time it takes to compile the libwebkit-owb.so file

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@LiveForIt

Are you sure that Joerg's machine is again broken ?
If you fellow the bugtracker you can see that he restart to work on it, generally a good tradition is to send a mail or a message to the involved dev, then only when talked with him, setup somethings and start ...

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@LiveForIt

Quote:
You have not uploaded anything to main SVN for years what are you taking about.
How about searching for my changes before writing such a nonsense? The last one is Changeset 1176 (sand-labs account required) which is only 8 months old.
IIRC I updated the AmigaOS 4.x parts on the OWB SVN more often before that, but since there never was anyone else working on the AmigaOS 4.x port of OWB it wasn't important to update the AmigaOS 4.x parts more often.

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

You do a lot of bugfixes on OWB core, so uploading changes often into the SVN can help automatically all other versions

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Home away from home
Home away from home


See User information
@joerg

Quote:
How about searching for my changes before writing such a nonsense?


I'm sorry, I only downloaded the files in the main trunk as explained in download section on sand-labs there I could not find any newer files then 2007 in the main trunk

Quote:
The last one is Changeset 1176 (sand-labs account required) which is only 8 months old.


I have created a account, but the url you sent me does not work.

Quote:
403 Forbidden (Insufficient permissions to access /branches/NewTrunk/BAL/Events/WebCore/AmigaOS4/BCEventHandlerAmigaOS4.cpp)

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@LiveForIt

Quote:
I'm sorry, I only downloaded the files in the main trunk as explained in download section on sand-labs there I could not find any newer files then 2007 in the main trunk
There are no OWB source archives, at least nothing still usable, you have to check out the sources with SVN. It's best to use the revision with the last AmigaOS 4.x changes, i.e. "svn co -r 1176 http://www.sand-labs.org/svn/trunk OWB". If you'd use the current trunk revision instead you'd get problems and would have to adapt some of the AmigaOS 4.x parts to changes in other parts.

Quote:
I have created a account, but the url you sent me does not work.
Strange, but being logged in is only required to view change sets, for browsing some branches of the sources, reporting bugs and submitting diffs, etc. For checking out the sources with SVN an account isn't required.

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@joerg

Quote:
You can't use MarkStackPosix.cpp on AmigaOS 4.x, you have to use MarkStackAmigaOS4.cpp instead (if it's not in the SVN yet: IIRC it's the same as MarkStackNone.cpp).


I have downloaded pukapuka, and I can find neither MarkStackAmigaos4.cpp nor MarkStackNone.cpp in the archive. How is it supposed to look?

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Quite a regular
Quite a regular


See User information
@joerg

I must have missed something along the way here... I noticed the creation of this thread but I didn't know what was going on behind the scenes...

What exactly is going on? you handing over the development of OWB or is someone else wanting to start there own branch or is something taking shape that allows more than just yourself to work on OWB sources etc?

I just want to make it clear that I will not support anything that places what we had in the beginning (e.g. Joerg working by himself on OWB and we getting nice wee updates) in the firing line and risking termination!



Everyone play fair, this is a great time for us!

~Yes I am a Kiwi, No, I did not appear as an extra in 'Lord of the Rings'~
1x AmigaOne X5000 2.0GHz 2gM RadeonR9280X AOS4.x
3x AmigaOne X1000 1.8GHz 2gM RadeonHD7970 AOS4.x
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@Slayer

Actually I'm not sure that I know what is going on with OWB. My own project is just to get the Qt WebKit module to work, and for that purpose I'm "stealing" small snips of code from OWB.

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@alfkil

Quote:
I have downloaded pukapuka, and I can find neither MarkStackAmigaos4.cpp nor MarkStackNone.cpp in the archive. How is it supposed to look?
http://www.sand-labs.org/svn/trunk/Ja ... runtime/MarkStackNone.cpp

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Quite a regular
Quite a regular


See User information
@alfkil

hmmm, well at least you're willing to elaborate in your direction... I'm hoping for the best in any case

~Yes I am a Kiwi, No, I did not appear as an extra in 'Lord of the Rings'~
1x AmigaOne X5000 2.0GHz 2gM RadeonR9280X AOS4.x
3x AmigaOne X1000 1.8GHz 2gM RadeonHD7970 AOS4.x
Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@joerg

Thanks Joerg . I must be needing glasses...

Go to top
Re: I have some good news, 98% of OWB compiles in cygwin :-)
Just can't stay away
Just can't stay away


See User information
@Slayer

Quote:
hmmm, well at least you're willing to elaborate in your direction... I'm hoping for the best in any case
If Alfkil gets QtWebKit working without problems and can build Arora with it you'll get a (simple) web browser and using something like OWB instead isn't required any more.

Go to top

  Register To PostTopic is Locked
« 1 (2) 3 »

 




Currently Active Users Viewing This Thread: 1 ( 0 members and 1 Anonymous Users )




Powered by XOOPS 2.0 © 2001-2024 The XOOPS Project