Support for multiple engine versions
Moderator: Moderators
Re: Support for multiple engine versions
Purpose of the lobby information is to notify client that newer "stable" version exists at all..
Re: Support for multiple engine versions
..means finding last directory (when sorting on name) without + in it's name.
Re: Support for multiple engine versions
Yeah but if you dont have that version locally(yet), something could tell the lobby that new one was released..
Re: Support for multiple engine versions
for the springcontent and such, what about hashing non-synced files like images and widgets separately from synced files, or just not hashing them?
Re: Support for multiple engine versions
We could add springignore.txt to the archive with patterns for such files.
However, I think the only true long term solution is working with 2 hashes for each archive; one depending on only the archive itself (for torrent system etc.), and one which includes entire dependency chain (for testing sync etc.)
However, I think the only true long term solution is working with 2 hashes for each archive; one depending on only the archive itself (for torrent system etc.), and one which includes entire dependency chain (for testing sync etc.)
Re: Support for multiple engine versions
well, you don't need to sync: anything in luaui, images/textures, etc
you do need to sync: models, fbi/luafbi, gadgets...
you do need to sync: models, fbi/luafbi, gadgets...
Re: Support for multiple engine versions
Im ready to implement downloader related changes .. so if plan is agreed up someone could make unitsync related changes..
I think that recent release shows such feature would be really usefull.
I think that recent release shows such feature would be really usefull.
Re: Support for multiple engine versions
Bumping this in relation to other idea.
Also I realized that a wrapper unitsync which calls the correct unitsync for a particular Spring version would be a pain to make & maintain IMO because of all the language bindings present in unitsync. (All methods would have to be wrapped for all bindings...)
Also I realized that a wrapper unitsync which calls the correct unitsync for a particular Spring version would be a pain to make & maintain IMO because of all the language bindings present in unitsync. (All methods would have to be wrapped for all bindings...)
Re: Support for multiple engine versions
I have been exploring using the JNA java library instead of Sun JNI in order to call unitsync. While I havent put enough time into it to build a viable replacement, this would allow spring to remove the JNI headers and thus for a lot of people remove the jdk dependency.
However I have been toying with this for a while and have little time to push forward so if anybody wishes to help then feel free to work on it. A basic class with a handful of functions and a test is in the latest battlehub SDK in svn IIRC
However I have been toying with this for a while and have little time to push forward so if anybody wishes to help then feel free to work on it. A basic class with a handful of functions and a test is in the latest battlehub SDK in svn IIRC
Re: Support for multiple engine versions
i have an AWK script that takes a C file and outputs a Java file with JNA wrapper functions for all function pointers in the C file.
should be easy to rewrite it to create JNA wrapper functions for all C functions in a file.
that should work well for unitsync then. thats what you would need, right?
i may do it, or will give you the AWK code, if you want.
should be easy to rewrite it to create JNA wrapper functions for all C functions in a file.
that should work well for unitsync then. thats what you would need, right?
i may do it, or will give you the AWK code, if you want.
Re: Support for multiple engine versions
Hmm maybe this awk script would be usable to actually generate wrapper unitsync too.
Re: Support for multiple engine versions
AWK is a tool that is installed on all linuxes and unixes (and i guess the BSDs to) and there is also a single exe for windows.
it is a text processing tool. you feed it text (eg a file) and it uses regex magic and a few basic functions to output text again.
it works well for translating code.
http://en.wikipedia.org/wiki/Awk
edit:
Tobi:
tell me more about it...
what is wrapper unitsync?
should i post the script? or should i try to get it to do the wraper unitsync you mention (after you told me what it is)?
it is a text processing tool. you feed it text (eg a file) and it uses regex magic and a few basic functions to output text again.
it works well for translating code.
http://en.wikipedia.org/wiki/Awk
edit:
Tobi:
tell me more about it...
what is wrapper unitsync?
should i post the script? or should i try to get it to do the wraper unitsync you mention (after you told me what it is)?
Re: Support for multiple engine versions
So any progress on this?