iamb-dev-internal Mailing List for IAMB -Iamb A Message Board (Page 3)
Status: Beta
Brought to you by:
rbowen
Menu
▾
▴
iamb-dev-internal — Discussion of the gory details involved in a rewrite for the next major version
You can subscribe to this list here.
| 2001 |
Jan
(48) |
Feb
(60) |
Mar
(16) |
Apr
(11) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Ken R. <kr...@qx...> - 2001-02-07 03:36:33
|
> I wasn't thinking so much of converting databases now. What I'm more > interested in is a method that I can in some sense feed an entire thread > to, and get a stree. I'm not quite sure if it's even an important > question, but it will certainly help in the testing phase to be able to > get a set of article objects and create a stree out of them. Since I seem to be the local stree surgeon, I'll tackle it, if you can tell me the structure of the thread you want to feed it. It won't happen until tomorrow, though, since I still have to write a test with 3 versions tonight. -- Ken |
|
From: Rich B. <rb...@rc...> - 2001-02-07 03:35:09
|
Matt Cashner wrote: > > On Tue, 6 Feb 2001, Rich Bowen wrote: > > > I wasn't thinking so much of converting databases now. What I'm more > > interested in is a method that I can in some sense feed an entire thread > > to, and get a stree. I'm not quite sure if it's even an important > > question, but it will certainly help in the testing phase to be able to > > get a set of article objects and create a stree out of them. > > then write it. it'll be in cvs in a few hours. if not, send me a patch. A question, then, to the API-meister, if he is here and not writing tests. What horticultural term would amuse you on this one? I might get to this in the little bit, and I might not, depending on how two other little things cooperate. (a coldsync conduit, and a TOC, not my family, in case you were wondering.) -- Rich Bowen -- Director of Web Application Development http://www.cre8tivegroup.com/ -- ri...@cr... Come see me at Apachecon! -- http://www.apachecon.com/ |
|
From: Ken R. <kr...@qx...> - 2001-02-07 03:33:00
|
> i was just going to commit source. i will wait for your patch. Viola, or whatever that French thingy is. -- Ken |
|
From: Matt C. <su...@qx...> - 2001-02-07 03:30:03
|
On Tue, 6 Feb 2001, Rich Bowen wrote: > I wasn't thinking so much of converting databases now. What I'm more > interested in is a method that I can in some sense feed an entire thread > to, and get a stree. I'm not quite sure if it's even an important > question, but it will certainly help in the testing phase to be able to > get a set of article objects and create a stree out of them. then write it. it'll be in cvs in a few hours. if not, send me a patch. m. |
|
From: Rich B. <rb...@rc...> - 2001-02-07 03:23:56
|
Matt Cashner wrote: > > On Tue, 6 Feb 2001, Rich Bowen wrote: > > > Well, it makes, passes 'make test', and installs OK, but it's a little > > hard to test it without a database handy. Will it be possible to > > construct a stree Ab Nihil, if all I have is a set of articles? That is, > > will it be possible to take an existing pre-2.0 discussion forum and > > convert it to strees? > > um, its a little premature to be converting the database yet. but yes, > there will be a converter. I wasn't thinking so much of converting databases now. What I'm more interested in is a method that I can in some sense feed an entire thread to, and get a stree. I'm not quite sure if it's even an important question, but it will certainly help in the testing phase to be able to get a set of article objects and create a stree out of them. -- Rich Bowen -- Director of Web Application Development http://www.cre8tivegroup.com/ -- ri...@cr... Come see me at Apachecon! -- http://www.apachecon.com/ |
|
From: Matt C. <su...@qx...> - 2001-02-07 03:14:31
|
On Tue, 6 Feb 2001, Rich Bowen wrote: > Well, it makes, passes 'make test', and installs OK, but it's a little > hard to test it without a database handy. Will it be possible to > construct a stree Ab Nihil, if all I have is a set of articles? That is, > will it be possible to take an existing pre-2.0 discussion forum and > convert it to strees? um, its a little premature to be converting the database yet. but yes, there will be a converter. m. |
|
From: Matt C. <su...@qx...> - 2001-02-07 03:06:14
|
On Tue, 6 Feb 2001, Ken Rietz wrote: > I am currently taking a break in between writing tests. If you want, I > can create the diff in less than 10 minutes. Is that OK, or have you > already gone ahead? i was just going to commit source. i will wait for your patch. m. |
|
From: Matt C. <su...@qx...> - 2001-02-07 03:01:43
|
i would like to suggest that we move the source tree off of sourceforge. i'd like it someplace that a) is not so damn slow b) is not so damn broken c) is in our control. frankly, i'm just tired of dealing with the issues of sourceforge when have the ability to host these services jsut as well if not better than sourceforge. m. |
|
From: Ken R. <kr...@qx...> - 2001-02-07 02:59:23
|
> ken, do you have a diff yet for the changes you want made? i'm probably > going to import the code into the tree here in a few minutes without your > revisions. you can make changes then through cvs. I am currently taking a break in between writing tests. If you want, I can create the diff in less than 10 minutes. Is that OK, or have you already gone ahead? -- Ken |
|
From: Matt C. <su...@qx...> - 2001-02-07 02:40:57
|
ken, do you have a diff yet for the changes you want made? i'm probably going to import the code into the tree here in a few minutes without your revisions. you can make changes then through cvs. m. |
|
From: Rich B. <rb...@rc...> - 2001-02-07 02:10:13
|
Matt Cashner wrote: > > On Mon, 5 Feb 2001, Matt Cashner wrote: > > > that's fine. i just want this as a module pretty quickly so we can begin > > to develop code against it as quickly as possible. > > and here it is. if this passes muster, i'll dump the current cvs tree, and > reinitialize it with this module in a pm/ tree. > > criticize, shred, but for gods sake give feedback. > > and yes, it passes its own 'make test' :) Well, it makes, passes 'make test', and installs OK, but it's a little hard to test it without a database handy. Will it be possible to construct a stree Ab Nihil, if all I have is a set of articles? That is, will it be possible to take an existing pre-2.0 discussion forum and convert it to strees? -- Rich Bowen -- Director of Web Application Development http://www.cre8tivegroup.com/ -- ri...@cr... Come see me at Apachecon! -- http://www.apachecon.com/ |
|
From: Matt C. <su...@qx...> - 2001-02-06 04:26:50
|
On Mon, 5 Feb 2001, Ken Rietz wrote: > The documentation for $stree -> prune(); can be pulled from the API. ok folks. lesson 1: how to create a patch when you dont have cvs. copy file. edit copy. diff -u file.original file.edit > patch email patch to list :) ken, give me a wondrous patch with your corrections and i will cvs this bad mama. m. |
|
From: Ken R. <kr...@qx...> - 2001-02-06 03:46:14
|
> criticize, shred, but for gods sake give feedback.
+++++
The documentation for $stree -> prune(); can be pulled from the API.
+++++
I was intending to use "plant" for "new" -- you guys were the ones
that said that "new" was overused, and this was one of the reasons
I went with the gardening names.
However, the functionality of plant needs to be slightly different.
Check the API. The intention I had was that new not only created
the stree, but initialized it with its $ArticleID as well:
sub plant {
my ($class, $ArticleID) = @_;
my $stree = '[' . $ArticleID . ']';
bless $stree, $class;
return $stree;
}
The call to plant would then look like this:
~~~~ Database creation of the root article, giving $ArticleID ~~~~
$nextstree = Stree -> plant ( $ArticleID );
That should work, unless you want to use a call like
$nextstree = Stree -> plant ( ARTICLE => $ArticleID );
to be consistent. In that case, the method would be:
sub plant {
my ($class, %args) = @_;
my $stree = '[' . $args{ARTICLE} . ']';
bless $stree, $class;
return $stree;
}
I'd actually prefer the second of those, mostly on the grounds
of consistency, since we will be using named parameters
everywhere else (I hope).
-- Ken
|
|
From: Matt C. <su...@qx...> - 2001-02-06 03:14:39
|
On Mon, 5 Feb 2001, Matt Cashner wrote: > that's fine. i just want this as a module pretty quickly so we can begin > to develop code against it as quickly as possible. and here it is. if this passes muster, i'll dump the current cvs tree, and reinitialize it with this module in a pm/ tree. criticize, shred, but for gods sake give feedback. and yes, it passes its own 'make test' :) m. |
|
From: Matt C. <su...@qx...> - 2001-02-06 01:10:21
|
On Mon, 5 Feb 2001, Ken Rietz wrote: > Well, there are remaining portions to code, too, but it sounds like that's fine. i just want this as a module pretty quickly so we can begin to develop code against it as quickly as possible. /me wonders if he has enough cat5 to reach the living room. grrr... i need wireless. figure an hour or two before i get this module out there. m. |
|
From: Ken R. <kr...@qx...> - 2001-02-06 00:55:39
|
Matt Cashner wrote: > > i'm going to convert this to a module with a test suite right off. once > ken approves my module as acceptable, we'll patch against that. Well, there are remaining portions to code, too, but it sounds like a little bit of forward movement after a long time. Be still my pounding heart .... :-) -- Ken |
|
From: Matt C. <su...@qx...> - 2001-02-06 00:38:06
|
i'm going to convert this to a module with a test suite right off. once ken approves my module as acceptable, we'll patch against that. m. |
|
From: Ken R. <kr...@qx...> - 2001-02-06 00:22:42
|
> ken, is the end of the week a reasonable deadline for code to back yoru > api? Probably, but hard to say. On the other hand, I do have some code splashed together now. It includes all the harder stuff, specifically dissect, prune, and heritage. (Bud, level, and siblings are all quite easy. Specifically, there was code at one time for siblings, if I remember correctly.) At this point, they are all in a single file that is not yet a package, i.e., not yet an object. I'm debugging the algorithms first. The OO stuff is easy to tack on. Feel free, if you want to do that, but I am more interested in polishing (speeding) up the algorithms. Some of these will be called repeatedly, and if we are going to see code speed increases, it is best to find them here. So, please, hack at the code mercilessly, from an efficiency standpoint. The overall code is commented, with extra notes thrown in, and there is a test stree (stress test stree?) and commented out code, to be sure that it all works. Some design decisions did not get commented, I expect, so ask all the questions you want. -- Ken |
|
From: Matt C. <su...@qx...> - 2001-02-05 23:51:13
|
On Sun, 4 Feb 2001, Rich Bowen wrote: > OK, I'm tired of talking, and I want to write some code. We had our amen. we're now 5 months after that meeting and we still dont have an stree module. yes i know you're busy ken. i'm not angry with you. its just frustrating. > since then. Everything that I want to do involves substantial breakage > in the stable tree, but if I need to do a major tag and release on the > stable tree, and then start breaking things over there, then so be it. the problem is that we're starting over from scratch over here. i really dont like the fact that there's even code in our tree right now. i just havent dumped the tree yet.and until we get ken's module, we are dead in the water. feel free to write code which depends on ken's api. there's just no way to test code for the meanwhile. ken, is the end of the week a reasonable deadline for code to back yoru api? frankly, i join rich in his frustration at the delay and the yammering. if something doesnt start happening soo, it will either be time to move on or time to start writing most of it myself. of course then it will be the poe thingy i mentioned. and i'll probably start that soon anyway. *shrug* like i said, we're sort of in holding until ken gets this module done and then the code can flourish. but until then... m. |
|
From: Rich B. <rb...@rc...> - 2001-02-04 20:40:32
|
OK, I'm tired of talking, and I want to write some code. We had our
little meeting in September, and no substantial code has been written
since then. Everything that I want to do involves substantial breakage
in the stable tree, but if I need to do a major tag and release on the
stable tree, and then start breaking things over there, then so be it.
But if we are to a point where I can start writing things in the dev
branch, I'd rather do that. Either way, the distinction between the two
branches is becoming less relevant, because either way the stable branch
is at a stopping point until I can be at liberty to start the breakage.
The way that I understand this strees stuff, it is a performance
enhancement. The methods that look up that sort of thing can look in the
strees, or it can look directly into the forest, as it were, for this
information. It makes very little difference to the CGI code, which is
just making a method call, and cares not a whit where that data comes
from.
So, with all of the above in mind, I'd like to write some code. If I
can't do it in the dev tree, I want to start doing it in the stable
tree, making the stable tree not so stable.
Thoughts? Comments? Strange remarks about anything other than whether a
certain person is wearing pants?
--
Rich Bowen -- Director of Web Application Development
http://www.cre8tivegroup.com/ -- ri...@cr...
Have trouble remembering things? http://www.mymissinghead.com/
|
|
From: Matt C. <su...@qx...> - 2001-02-04 20:11:58
|
On Sat, 3 Feb 2001, Rich Bowen wrote: > I think that it would be really interesting to write a real non-CGI > thingy in Perl. ok. thought about side project that could greatly benefit iamb. what if we all build a real honest to god perl web server, using poe? maybe just hork code from modules so it doesnt depend on any modules. or not. but i think a real perl web server with modperl-like capabilities would a) be fun b) be challenging c) be of great benefit to the community d) might be talk worthy for yapc. :) thoughts? m. |
|
From: Rich B. <rb...@rc...> - 2001-02-03 23:57:53
|
Matt Cashner wrote:
>
> On Sat, 3 Feb 2001, Rich Bowen wrote:
>
> > The neat thing about this is that if it actually works, we end up
> > working on a "real" product, rather than just CGI and CGI-ish stuff. I'm
> > about ready to move onto something different
>
> you serious? about moving on that is.
Well, we put it in a different CVs repository, and if it sucks, we
forget it as a bad idea and move on with life.
I think that it would be really interesting to write a real non-CGI
thingy in Perl.
So, yeah, I guess I'm serious.
I am marginally concerned about the 4-way spit of the development
effort, but since we're not working towards any fixed deadline, I don't
see any real harm in this. And if we do the internals right, we won't
actually be losing anything.
--
Rich Bowen -- Director of Web Application Development
http://www.cre8tivegroup.com/ -- ri...@cr...
Have trouble remembering things? http://www.mymissinghead.com/
|
|
From: Matt C. <su...@qx...> - 2001-02-03 23:49:24
|
On Sat, 3 Feb 2001, Rich Bowen wrote: > The neat thing about this is that if it actually works, we end up > working on a "real" product, rather than just CGI and CGI-ish stuff. I'm > about ready to move onto something different you serious? about moving on that is. m. |
|
From: Rich B. <rb...@rc...> - 2001-02-03 23:45:19
|
Matt Cashner wrote:
...
> thoughts anyone? i'm going to work on the webserver end of this just for
> fun but it would be nice if iamb could benefit from it as well.
The neat thing about this is that if it actually works, we end up
working on a "real" product, rather than just CGI and CGI-ish stuff. I'm
about ready to move onto something different
But I still think that it would be pretty cool if we could have a
something that had multiple interfaces, one of which was perhaps CGI,
and perhaps mod_perl, as well as other things. I'm not quite sure what
that meant, but I think it meant something interesting.
--
Rich Bowen -- Director of Web Application Development
http://www.cre8tivegroup.com/ -- ri...@cr...
Have trouble remembering things? http://www.mymissinghead.com/
|
|
From: Matt C. <su...@qx...> - 2001-02-03 22:45:49
|
On Sat, 3 Feb 2001, Ken Rietz wrote: > Who is dngor? Presumably someone reliable, but if we base IAMB on > his/her code, I'd really like to know more. um the author of POE and the god of all things POE. > It sounds like you are going to re-write the relevant chunk of > Apache. We'd have to do everything Apache does for us "by hand", > such as authentication, error stuff, etc. Since I am not sure of > all that Apache does, I'd defer to Rich on that one. apache does a lot of silly crap. we'd need auth and logs but all that's easy and the code dngor gave me does all that :) > Would there only be one Perl interpreter running at a time, ala > modperl, or would one get fired up for each connection? one interp per server. we can eval the code and run it in Safe.pm (see perldoc Safe) > Does POE > take care of that? (I still don't know what POE is, but clearly > if we intend to communicate, I'd better find out. What are some > refs to it?) cpan> install POE linux$ perldoc POE :) i can send out some sample code if you like > If we could get IAMB independent of server, that would *rock*. > It is certainly worth some exploration. groovy m. |
2 messages has been excluded from this view by a project administrator.
