Re: Advanced search with wildcard using notmuch for mutt

[io <io@ooeeeoo.com>]

what xapian 'indexing system' did was to index the entire sentence 'xxx_yyy' and you will not be able to find any sentence which contain the word 'yyy'?
xapian should have this simple wildcard feature which 'grep'(search) offer. ($grep '*word*' file). It is strange that xapian restrict the search to 'trailing wildcard' only.
Novice user who get introduce to notmuch just want to run the search and get the result straight away. 

-- 
Best regards,
io
On Thu, Nov 30, 2023 at 09:31:19AM -0400, David Bremner wrote:
> Michael J Gruber <michaeljgruber+grubix+git@gmail.com> writes:
> 
> 
> > Using xapian commands, one could extract all stems and grep those for a
> > term which one "remembers partially" (often happened to me), and then feed
> > that into notmuch. Might be worthwhile scripting or even integrating into
> > notmuch (sexp?).
> 
> The words are stored unstemmed as well, so in principle we could extract
> those and do some kind of fuzzy search on them to construct
> queries. This is what notmuch already does for regex searches on fields
> other than from, subject, and mid. The reason this is not too attractive
> for the message body is that it works on a per word basis, and most uses
> of regex (although not the one under discussion) involve matching
> multiple words. Currently Xapian only supports trailing wildcards (which
> would not help here), but we could do that for body words. It just
> doesn't (or hasn't) seemed like such a common use case. 
> 
> Alas, none of the forward looking discussion really helps the original
> poster.