FreeLists / haiku-development / [haiku-development] Re: TrackerGrep [was Re: missing -lm?]

["François Revol" <revol@xxxxxxx>]

> > > I think that idea is fine and support it.
> > >
> > > What would also be good is to add some indexing of files, a la
> > > Beagle,
> > > Spotlight and all that. Though that is certainly a lot more work
> > > than
> > > just adding TrackerGrep.
> >
> > This is definitely a must have, though not for R1. We will
> >  probably implement
> > the indexing in a separate server. The most interesting question is
> >  how to do
> > that in a fast way. We might need to add some special FS support
> > for it.
>
> That's one way of doing it.  And fairly safe.  Another way is to have
> attributes which contain multiple elements.  I did some experiments
> with

No that's the same way, you'd still need an indexing server to set the
attributes whatever they are layed out.

> that, using space separated strings interpreted as a list of
> keywords.
> It fitted in with the current system fairly well.

That's useless, queries work perfectly well with pattern matching, you
can query for META:keyw=="*foo*".
Of course it matches bigger words too, but you could just make sure you
have a space at start and end so you can query for "* foo *".

> The applications (or a background thread running a Translator) would
> have
> to be adapted to write a standard "META:keywords" attribute with all
> the
> keywords for a modified document.

Bookmarks uses META:keyw, not META:keywords.

> The file system would need only a few changes.  When adding an
> attribute
> to the index, each word in the attribute is added separately to the
> index,
> with a back pointer to the file as usual.  Thus the file is in the
> index
> multiple times.  Deleting the keywords is similar to adding.
> Searching
> (and live queries) is a bit different, since you may get the same
> file
> multiple times.  The simple solution is to have user level code
> filter
> out the duplicates.

Again that's useless complexity.

François.