Hello,
The following decisions & plans were made:
Disisions:
o CSH/TCSH development postponed untill we actually need it. o Porting of dependencies that have uses outside of the OOo project to gain momentum.
Tasks:
o AC will investigate the feasability of writing a wrapper for libusb o SieGel will investigate/port autoconf & automake o Bean will investigate/port aspell o tonyw will investigate/port BerkleyDB o Archprogrammer will investigate the currently available ports for usuability o Archprogramer will investigate the work required to port NAS.
I've set up the needed tasks & projects on http://dev.amigaopenoffice.org, and assigned them to the proper people, however, that doesn't mean anyone else can't help out ;)
Also, everyone with access to dev.amigaopenoffice.org has to fill in the proper details in their user account.
The IRC log is available here: http://amigaopenoffice.org/index.php?option=com_wikidoc&task=viewcontent...
Kind regards
Well, it's getting quiet here (somehow), or it's just my impression ?
o AC will investigate the feasability of writing a wrapper for libusb o SieGel will investigate/port autoconf & automake o Bean will investigate/port aspell o tonyw will investigate/port BerkleyDB o Archprogrammer will investigate the currently available ports for usuability o Archprogramer will investigate the work required to port NAS.
So, those are your immediate tasks ? Anything I can help out?
I've set up the needed tasks & projects on http://dev.amigaopenoffice.org, and assigned them to the proper people, however, that doesn't mean anyone else can't help out ;)
Also, everyone with access to dev.amigaopenoffice.org has to fill in the proper details in their user account.
I have no access to that site. If someone would like to add me, go ahead.
BTW: I don't know if anyone did that already...I just compiled the single csh files with the gcc3.4 but didn't link them yet. Suddenly I have only private libraries (no dedicated 'struct DOSBase' ?? Have to check out what's wrong with my SDK). I just did it for some personal reasons and I am not sure if this can help you at all (I guess you discussed that already). It runs internal commands only and passes everthing else streight into the Amiga shell (I guess thru a System call, didn't look into the code yet), I don't know if you have access to it's local envs after that point. The command abbreviation is a pain (e.g. type Make instead of make, because otherwise finds the first (AmigaOS) command starting with 'make', which is makelink instead). The parser has to be extended to understand more sh keywords and sub shells are not possible (without a major rewrite). Command abbreviation has to be switched off, unix path names should be default (-lunix ?). So, csh requires a major rewrite. Didn't look into (the mentioned) tcsh, yet.
And no, I have no clue how a sh really works. I am no unix guru...
cheers Jürgen
On 2005-01-26, Jürgen Schober wrote:
Well, it's getting quiet here (somehow), or it's just my impression ?
No, it has become a tad more quiet - which I think is both because the most urgent questions seem to be resolved and because people are discovering that they have a life also. :-)
The sub-projects are also starting to take form so people might be working on them. :-)
<SNIP>
So, those are your immediate tasks ? Anything I can help out?
Certainly. One suggestion off the top of my head is to help out with untangling the dependancy tree of OOo. SieGeL posted a list of the subdirectories of the OOo source tree recently and improving it would help.
Otherwise you can always come up with your own suggestions.
Perhaps our benevolent non-dictator has a suggestion?
BTW: I don't know if anyone did that already...I just compiled the single csh files with the gcc3.4 but didn't link them yet. Suddenly I have only
Already examined it. As you already discovered, it seems far too incomplete to be useful for running anything but the simplest csh scripts.
<SNIP>
(-lunix ?). So, csh requires a major rewrite. Didn't look into (the mentioned) tcsh, yet.
I also examined tcsh a bit and even if it appears to be fairly portable there is still the problem of fork() in addition to some smaller things like termios, a few signals etc.
The general consensus seems to be that it is easier to write a ne c-shell intepreter. (i.e. a c-shell without any command-line features)
And no, I have no clue how a sh really works. I am no unix guru...
Don't worry. It seems we have at least a couple of people with several years of Unix experience on the list. Feel free to ask if there is anything in particular you wonder.
-Peter aka. Archprogrammer
Reality is for people who cannot face ScienceFiction. Only lefthanded people are in their right minds.
Hello Jürgen
On 26/01/2005, you wrote:
Well, it's getting quiet here (somehow), or it's just my impression ?
Aslong as they still work i'm not complaining :D
o AC will investigate the feasability of writing a wrapper for libusb o SieGel will investigate/port autoconf & automake o Bean will investigate/port aspell o tonyw will investigate/port BerkleyDB o Archprogrammer will investigate the currently available ports for usuability o Archprogramer will investigate the work required to port NAS.
So, those are your immediate tasks ? Anything I can help out?
Depends, what can you do? Are you good with USB stacks? Network related code?
I've set up the needed tasks & projects on http://dev.amigaopenoffice.org, and assigned them to the proper people, however, that doesn't mean anyone else can't help out ;)
Also, everyone with access to dev.amigaopenoffice.org has to fill in the proper details in their user account.
I have no access to that site. If someone would like to add me, go ahead.
Just checked my list of names, i don't have your nickname to add you, so maybe you should tell that first ;)
BTW: I don't know if anyone did that already...I just compiled the single csh files with the gcc3.4 but didn't link them yet. Suddenly I have only private libraries (no dedicated 'struct DOSBase' ?? Have to check out what's wrong with my SDK). I just did it for some personal reasons and I am not sure if this can help you at all (I guess you discussed that already). It runs internal commands only and passes everthing else streight into the Amiga shell (I guess thru a System call, didn't look into the code yet), I don't know if you have access to it's local envs after that point. The command abbreviation is a pain (e.g. type Make instead of make, because otherwise finds the first (AmigaOS) command starting with 'make', which is makelink instead). The parser has to be extended to understand more sh keywords and sub shells are not possible (without a major rewrite). Command abbreviation has to be switched off, unix path names should be default (-lunix ?). So, csh requires a major rewrite. Didn't look into (the mentioned) tcsh, yet.
And no, I have no clue how a sh really works. I am no unix guru...
The csh & tsch ports we currently have require alot of work to make them, well, work, so those are a dead end, when the time comes we'll need to revisit this issue, but for now it's best we focus on tasks we can use on the short term.
Regards
So, those are your immediate tasks ? Anything I can help out?
Depends, what can you do? Are you good with USB stacks? Network related code?
Well, I I don't know...I did some socket stuff recently but I am not really into networks. I have no clue how USBworks, to be honest. OTHO, doing CCCC/C++ since 15 years, I guess Ihave no problem challanging new frontiers ;) The only thing is, I don't know if I have much time.
I have no access to that site. If someone would like to add me, go ahead.
Just checked my list of names, i don't have your nickname to add you, so maybe you should tell that first ;)
Use juergen ;) That's easy.
The csh & tsch ports we currently have require alot of work to make them, well, work, so those are a dead end, when the time comes we'll need to revisit this issue, but for now it's best we focus on tasks we can use on the short term.
I still want to look into it... I have started it, so at least I'd like to get it going. Helps me to understand DOS internals. And I dont know .. Following the tcsh/fork discussion ... using csh, collecting all globals into a (what I call) context and use the parser from csh + adding pthreads and make the code reentrant and threadsave would result in a new shell (this is in fact a rewrite, but I don't have to re-invent the wheel). I'll do that...I cannot guarantee anything, but maybe this will be of any use for you... even if you go another route.
BTW:I looked into the other mail (what was his name, Andre Siegel?), IMO it seems this is a complete packages with all the tools involved -> much more than we really need. I'll look into this to get a better idea what is really needed. And if I have a better picture about the module structure I can tell you what I can do... Or you tell me, what you need, and I tell you if I can do that ;)
----- Original Message ----- From: "Jürgen Schober" juergen.schober@pointdesign.at To: openoffice-os4@samfundet.no Sent: Wednesday, January 26, 2005 7:00 PM Subject: Re: [OO.org-OS4] Re: Results of IRC meeting 24/01/05
Well, I I don't know...I did some socket stuff recently but I am not really into networks. I have no clue how USBworks, to be honest. OTHO, doing CCCC/C++ since 15 years, I guess Ihave no problem challanging new frontiers ;)
What? and I thought we were doing that in AmigaBasic ?? ;-) Now THERE'S a challenge ;-)
And now I await the flames ;-).....hehehe
Jürgen Schober wrote:
So, those are your immediate tasks ? Anything I can help out?
Depends, what can you do? Are you good with USB stacks? Network related code?
Well, I I don't know...I did some socket stuff recently but I am not really into networks.
I would just like to add here that I'm perfectly capable of writing networking stuff for _Linux_, but how one would even BEGIN writing networking stuff for AmigaOS is way beyond me. I miss ixnet.library :-P
Ole-Egil
On 2005-01-27, Ole-Egil Hvitmyren wrote:
<SNIP>
I would just like to add here that I'm perfectly capable of writing networking stuff for _Linux_, but how one would even BEGIN writing networking stuff for AmigaOS is way beyond me. I miss ixnet.library :-P
<SNIP>
How difficult can it be? I mean, an ethernet frame is an ethernet frame - right? :-P
I think you will find that sockets are about the same so I think you will do just fine.
But perhaps we should put you on porting the UNO _bridges_? :-P
-Peter aka. Archprogrammer
PS: I received the A1 board yesterday so work is going slow (both this and my regular work :-) as I am installing and configuring the new machine. Another thanks to Stefan!
Reality is for people who cannot face ScienceFiction. Only lefthanded people are in their right minds.
Peter Bengtsson wrote:
On 2005-01-27, Ole-Egil Hvitmyren wrote:
<SNIP>
I would just like to add here that I'm perfectly capable of writing networking stuff for _Linux_, but how one would even BEGIN writing networking stuff for AmigaOS is way beyond me. I miss ixnet.library :-P
<SNIP>
How difficult can it be? I mean, an ethernet frame is an ethernet frame - right? :-P
Not if you don't find any relevant includes :-P
I think you will find that sockets are about the same so I think you will do just fine.
Well, I've finally found some documentation about how to do it, so maybe I'll manage :-)
Hello Ole-Egil
On 27/01/2005, you wrote:
Peter Bengtsson wrote:
On 2005-01-27, Ole-Egil Hvitmyren wrote:
<SNIP>
I would just like to add here that I'm perfectly capable of writing networking stuff for _Linux_, but how one would even BEGIN writing networking stuff for AmigaOS is way beyond me. I miss ixnet.library :-P
<SNIP>
How difficult can it be? I mean, an ethernet frame is an ethernet frame - right? :-P
Not if you don't find any relevant includes :-P
You need the amitcp dev kit IIRC
I think you will find that sockets are about the same so I think you will do just fine.
Well, I've finally found some documentation about how to do it, so maybe I'll manage :-)
IF you get stuck, just tell us ;)
Regards
Hello Jürgen
On 27/01/2005, you wrote:
So, those are your immediate tasks ? Anything I can help out?
Depends, what can you do? Are you good with USB stacks? Network related code?
Well, I I don't know...I did some socket stuff recently but I am not really into networks. I have no clue how USBworks, to be honest. OTHO, doing CCCC/C++ since 15 years, I guess Ihave no problem challanging new frontiers ;) The only thing is, I don't know if I have much time.
Good to know ;)
I have no access to that site. If someone would like to add me, go ahead.
Just checked my list of names, i don't have your nickname to add you, so maybe you should tell that first ;)
Use juergen ;) That's easy.
Done, see private mail
The csh & tsch ports we currently have require alot of work to make them, well, work, so those are a dead end, when the time comes we'll need to revisit this issue, but for now it's best we focus on tasks we can use on the short term.
I still want to look into it... I have started it, so at least I'd like to get it going. Helps me to understand DOS internals. And I dont know .. Following the tcsh/fork discussion ... using csh, collecting all globals into a (what I call) context and use the parser from csh + adding pthreads and make the code reentrant and threadsave would result in a new shell (this is in fact a rewrite, but I don't have to re-invent the wheel). I'll do that...I cannot guarantee anything, but maybe this will be of any use for you... even if you go another route.
We do need csh, but not in the near future, you now have access to dev.amigaopenoffice org, i'll let you in charge of the csh/tcsh part, can you make an estimate of the work involved by 4 Februari? (Not the time needed, but what steps need to be taken, what you will need to make it work, those things ;)
BTW:I looked into the other mail (what was his name, Andre Siegel?), IMO it seems this is a complete packages with all the tools involved -> much more than we really need. I'll look into this to get a better idea what is really needed. And if I have a better picture about the module structure I can tell you what I can do... Or you tell me, what you need, and I tell you if I can do that ;)
You should contact SieGel talk this over with him, to avoid double work.
Regards
Use juergen ;) That's easy.
Done, see private mail
OK.
We do need csh, but not in the near future, you now have access to dev.amigaopenoffice org, i'll let you in charge of the csh/tcsh part, can you make an estimate of the work involved by 4 Februari? (Not the time needed, but what steps need to be taken, what you will need to make it work, those things ;)
Done.
You should contact SieGel talk this over with him, to avoid double work.
Yes, but I need an overview anyhow. If I don't know how it works, I cannot really tell what I could do...(I would see this as prject analysis or such)
<Big Snip>
The csh & tsch ports we currently have require alot of work to make them, well, work, so those are a dead end, when the time comes we'll need to revisit this issue, but for now it's best we focus on tasks we can use on the short term.
Sorry for missing the meeting on Monday, I broke down in the fast lane on the M1. A rather harrowing experience I can tell you. Onto the message
Whilst I hate to contradict our benevolent dictator, this could be a mistake why?
well, whilst its very good to get the other utilities(dependancies) ported and it will provide some output for the project (always a good thing), if we dont start thinking about csh/tcsh now, and getting something at least started we will come to a point (fairly rapidly from what i can tell - at least as far as projects go) where we will come to a complete stop because of their lack of availability. What we should be doing is using a schedules to plan when we will be doing something about the issue, now we obviously cant plan based on time, as all the programmers are doing this in their spare time, we can do it based on acheivements (provided the project plan is set up correctly), so ie when 50% of dependancies are complete we need to start on build environment etc. This is one of the reasons it makes sense to have a top level porject (say dependencies) and then sub projects for all of the apps that need porting, you can then see at a glance the percentage complete of the dependencies top level project.
This also applies to the bash shell (abc-shell??) which will be needed to help port the dependancies I believe (along with automake/conf etc)
Mark
Hello Mark
On 27/01/2005, you wrote:
<Big Snip>
The csh & tsch ports we currently have require alot of work to make them, well, work, so those are a dead end, when the time comes we'll need to revisit this issue, but for now it's best we focus on tasks we can use on the short term.
Sorry for missing the meeting on Monday, I broke down in the fast lane on the M1. A rather harrowing experience I can tell you. Onto the message
Whilst I hate to contradict our benevolent dictator, this could be a mistake why?
You dare question my authority?! j/k ;)
well, whilst its very good to get the other utilities(dependancies) ported and it will provide some output for the project (always a good thing), if we dont start thinking about csh/tcsh now, and getting something at least started we will come to a point (fairly rapidly from what i can tell - at least as far as projects go) where we will come to a complete stop because of their lack of availability. What we should be doing is using a schedules to plan when we will be doing something about the issue, now we obviously cant plan based on time, as all the programmers are doing this in their spare time, we can do it based on acheivements (provided the project plan is set up correctly), so ie when 50% of dependancies are complete we need to start on build environment etc. This is one of the reasons it makes sense to have a top level porject (say dependencies) and then sub projects for all of the apps that need porting, you can then see at a glance the percentage complete of the dependencies top level project.
That was actually what i had in mind, port a fair deal of the dependencies, focussing on the more directyl usuable ones, and when a good amount of them are done, then reinitiate the csh development.
However, Jurgen has offered to continue on the csh/tcsh part, and i've agreed to let him work it out, the rest of us can continue with their currently assigned tasks.
As for the project layout, you do raise a good point (the better overview of the achieved work), i'll get back to you about that after i had some food
This also applies to the bash shell (abc-shell??) which will be needed to help port the dependancies I believe (along with automake/conf etc)
SieGel is investigating automake and related tools.
Regards
However, Jurgen has offered to continue on the csh/tcsh part, and i've agreed to let him work it out, the rest of us can continue with their currently assigned tasks.
Oh man...that's sounds so serious. Means I have to deliver something any time (soon), now ;) And that at a time I get my own department organized here...becoming a slave of myself again. Though those times were over ;-> Damn...
cheers Jurgen
-
Christophe Ochal -
Jürgen Schober -
Mark bond -
Ole-Egil Hvitmyren -
Peter Bengtsson -
stephane richard