www.tombraiderforums.com  

Go Back   www.tombraiderforums.com > Tomb Raider Modding > Tomb Raider Level Editor > Software Development

Reply
 
Thread Tools
Old 21-06-16, 10:18   #1941
vvsgh
Student
 
Join Date: Jun 2015
Posts: 134
Default

Quote:
Originally Posted by Lwmte View Post
There is one problem, however. I did significant scripting upgrade with already C++-ficated master, so certain script functions need to be back-ported to TeslaRus branch from current (deprecated) master.
Shouldn't we think about implementing something like SCUMMVM modular framework? That would make integration of different branches easier. Current engine architecture is monolithic.

Quote:
Originally Posted by Kikiloco View Post
i suggest it have to be regrouped, checks how much developers still current interested, which skill they have to contribute, under which condition they agree to collaborate, what goals they want to archive, and if they can work based in a plan and assigned task.
The OpenTomb is a small project. There is not enough skilled developers to separate work like that. There is no permanent developers either.

Quote:
Originally Posted by stohrendorf View Post
Yes, protect it! (And do the same with the backup branch.)
That's what only admins can do. The collaborator's rights are insufficient in this case.
vvsgh is offline   Reply With Quote
Old 21-06-16, 10:36   #1942
TeslaRus
Student
 
TeslaRus's Avatar
 
Join Date: Jan 2013
Posts: 195
Default

maybe it will be better if merging of my repo was:
1) Backup master (done);
2) Make git patch (all my branch commits);
3) In master: git reset hard c9404a0; (maybe was not works easy because pull request commits had been earlier than c9404a0)...
4) Apply patch;
5) Upload - history was saved...

About scripts: for correct work use LUA v5.2... + LUA calls every frame (so use engine frame time in time depended scripts); some features was moved from master;
About debug: added OutTextXYZ(...) - see examples in DebugInfo in engine.cpp;

Issues must be updated...
TeslaRus is offline   Reply With Quote
Old 21-06-16, 11:14   #1943
vvsgh
Student
 
Join Date: Jun 2015
Posts: 134
Default

Quote:
Originally Posted by TeslaRus View Post
maybe it will be better if merging of my repo was:
1) Backup master (done);
2) Make git patch (all my branch commits);
3) In master: git reset hard c9404a0; (maybe was not works easy because pull request commits had been earlier than c9404a0)...
4) Apply patch;
5) Upload - history was saved...
That history would be synthetic rather then real. The only benefit I can see is bisection and it can be done in a local repository if needed, because we kept real history in backup branch. FWIW I even have copies of old mercurial repositories, but never needed it.

Pulling history from your branch allows to automatically merge between them. You can do git pull and git push from your repository. If we do the patch there will be only one common ancestor, i.e c9404a0.

EDIT: Sorry, I have to correct myself. There will be no common ancestors at all - git doesn't know anything about your repository root.

Quote:
Originally Posted by TeslaRus View Post
Issues must be updated...
Yes, we'll do it eventually.

Last edited by vvsgh; 21-06-16 at 11:28.
vvsgh is offline   Reply With Quote
Old 21-06-16, 11:34   #1944
TeslaRus
Student
 
TeslaRus's Avatar
 
Join Date: Jan 2013
Posts: 195
Default

Quote:
Originally Posted by vvsgh View Post
Pulling history from your branch allows to automatically merge between them. You can do git pull and git push from your repository. If we do the patch there will be only one common ancestor, i.e c9404a0.
Ok, thanks for information;
About repository: do not commit to src/core - that is rarely changed, low level modules folder, if someone need some new functionality in core modules - write request to me (may be functionality already exists or locates in other place);
For first time: do changes in feature branches and make merge requests - I can understood what functionality are needed (and how it planned to implements) and make advises to use some or other functions + it will be more stable;
TeslaRus is offline   Reply With Quote
Old 21-06-16, 15:56   #1945
TeslaRus
Student
 
TeslaRus's Avatar
 
Join Date: Jan 2013
Posts: 195
Default Upd:

At home... I begin to write simple documentation of engine functions;
UPD: Triggers are implemented in trigger.cpp module (instead LUA scripts) + original triggers may be overwritten by script commands (it may be debugged); use "y" key to switch debug mode, it has triggers information output;
Al library was updated: it uses SDL backend and did not crash debug!
TeslaRus is offline   Reply With Quote
Old 21-06-16, 16:58   #1946
vvsgh
Student
 
Join Date: Jun 2015
Posts: 134
Default

BTW, before we forget about it. TRosettaStone is now buried under backup branch.
vvsgh is offline   Reply With Quote
Old 21-06-16, 17:07   #1947
TeslaRus
Student
 
TeslaRus's Avatar
 
Join Date: Jan 2013
Posts: 195
Default

it is accessible in stohrendorf branch
TeslaRus is offline   Reply With Quote
Old 21-06-16, 17:13   #1948
stohrendorf
Hobbyist
 
stohrendorf's Avatar
 
Join Date: Jul 2015
Location: Aachen, Germany
Posts: 77
Default

Yeah well... once someone visits opentomb.earvillage.net, and the server-saved timestamp is older than 15 minutes, the master branch will automatically be pulled, and if that succeeds, opentomb.earvillage.net/OpenTomb/doc/trosettastone.html will vanish. How to cope with that?
stohrendorf is offline   Reply With Quote
Old 21-06-16, 17:21   #1949
vvsgh
Student
 
Join Date: Jun 2015
Posts: 134
Default

Quote:
Originally Posted by Nickotte View Post
Ah, I see. But what about the previous commits, why wasn't it rolled back once it was found to be not ideal?
Because nobody was ready to admit that it was a mistake.

Quote:
Originally Posted by stohrendorf View Post
opentomb.earvillage.net/OpenTomb/doc/trosettastone.html will vanish. How to cope with that?
Somebody should back port TRosettaStone from backup to master.

EDIT: Did a back port. Everything seems ok. Review it in my GitHub PR.

Last edited by vvsgh; 21-06-16 at 18:40.
vvsgh is offline   Reply With Quote
Old 21-06-16, 19:01   #1950
TeslaRus
Student
 
TeslaRus's Avatar
 
Join Date: Jan 2013
Posts: 195
Default

https://github.com/opentomb/OpenTomb...ster/ENGINE.md - first engine documentation version;
doc was restored in master;
I will upload and write plan / decomposition little bit later (may be tomorrow);
and: if was question about engine, please ask about, I will try to explain

Last edited by TeslaRus; 21-06-16 at 19:04. Reason: add info / update
TeslaRus is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT. The time now is 15:23.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.