At 10:29 PM +0200 7/12/00, Gaillard Pierre-Olivier wrote:
> - handling occurrences as mifluz is said to do it (I have not checked
>yet, so please don't flame if I did not understand the change list
>comments) would be a big improvement other htdig. It is the best
>solution for rare words
Loic took my original (crummy) htword code for 3.2 and used that as a
base for mifluz. In some sense, mifluz is ht://Dig's htword and the
two CVS trees are synced at a reasonable rate. Obviously both Loic
and I would love to see them keep more in sync, but it often takes
some times. But saying "ht://Dig should use mifluz" is a bit
redundant. :-)
You'd probably be interested to know that there is work in mifluz to
incorporate something closer to what you describe. However, it's not
based on files for individual words, but rather on documents. The
indexing speed would be much faster if it just dumped an index of a
document into one given file and later merged it.
In any case, the way things are shaping up, it seems that the more
technical discussions on the format of the indexes and so on is going
on through the mifluz list and ht://Dig is centering more on
user-level issues. Of course Loic and I (among others) subscribe to
both lists.
Cheers,
-Geoff
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED]
You will receive a message to confirm this.
