According to [EMAIL PROTECTED]: > Thank you for the heads up on the state of htMerge... We may be able to > help (here at ViWire) correct this if we could get more information about > the exact nature of the problem holding up the fix of htMerge.
Well, I think Geoff should comment on this, because I'm not even sure if the mifluz code merge is tied to the memory problems in htmerge or not. I think Geoff has a better handle on this specific problem in htmerge than I do. What do you think, Geoff? Are they related? > It seems to me (based the little information I have) that mifluz is the > root of the problem. Could you or anyone who is involved with this part of > the code send me details about what mifluz does and/or what it is intended > to do? Is is a functional part of the code, or can it be safely extracted > from the mainline program? Is it easier to remove or to finish > integrating? > > Thanks, and I'm looking forward to contributing to the quality of this > project! > > -Rav I know that there are a few problems that are supposed to be fixed (or become fixable) once we have the newest version of mifluz in place. Mifluz is essentially the core functionality of htdig's 3.2 series, managing the word database. It uses Berkeley DB at the center, but the DB code was patched to allow compression of the word database. There were two programmers from France (Loic Dachary and Marcel Bosc) who develped this code back in the winter of 1999, starting with the htdig code, but developing this as a separate set of C++ classes that could be dropped into other search tools, and then reintegrating these new mifluz libraries into what is the current 3.2 development code for htdig. The idea was that mifluz and htdig would be maintained in parallel, so that any updates to one project would benefit the other. That worked for about 6 months until these programmers got busy with other things. At that point, they made a lot of changes and bug fixes in mifluz, but never reintegrated these changes into htdig. Geoff tried to integrate the new code back in the summer/fall of last year, but there were all sorts of things that stopped working after that, so he had to back out those changes until we could figure out what went wrong and what needs to be changed to htdig to accomodate the changes in mifluz. This was supposed to be a temporary situation, and Loic was eventually supposed to help us figure out the changes, but it's been over a year now with no progress. So, now we have a difficult situation. htdig is tied to an old version of a package that is for all intents and purposes unsupported at this time, but this package is crucial to the functioning of the engine. The question is how is the best way to dissociate htdig from this package. We can work with the code we have now, and hunt down and fix the bugs, but we may end up duplicating a lot of effort that Loic and Marcel have already put into the latest version of mifluz. The other option is to make another attempt at merging in the new mifluz code, and then spending the time squashing the bugs that result from the introduced incompatibilities. I think the second approach is preferable, but I doubt we can count on any help from the mifluz developers for this, so it may or may not be easier. Either way, it's a big project, and all the htdig developers are too busy right now to put in the time required. If you guys are willing to help out, it would be much appreciated. I'm sure Geoff and I could help get you up to speed on things. -- Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]> Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/~grdetil Dept. Physiology, U. of Manitoba Phone: (204)789-3766 Winnipeg, MB R3E 3J7 (Canada) Fax: (204)789-3930 _______________________________________________ htdig-general mailing list <[EMAIL PROTECTED]> To unsubscribe, send a message to <[EMAIL PROTECTED]> with a subject of unsubscribe FAQ: http://htdig.sourceforge.net/FAQ.html
