acedb query language proposals

Dave Matthews matthews at GREENGENES.CIT.CORNELL.EDU
Sat Oct 5 17:30:56 EST 1996


> From BIOSCI-REQUEST at net.bio.net Fri Sep 20 14:07:01 1996
> To: acedb at net.bio.net
> From: <rd at sanger.ac.uk>
> Subject: acedb query language proposals

The proposed new query language sounds great.  I have an important user-
submitted query I can't do with the existing language, "find all Loci
within x units of Locus A on any Map", for lack of the num_op (arithmetic)
functionality included in the proposal.  So I'm really looking forward to
it.  A few suggestions:

1.
The "right_of" relation Jean mentioned, e.g. the relation between c and d in

FOO f
tag a b
tag c d

is crucial.  Maybe it could be given its own operation name?


2.  Should there be an escape prefix like \ for the wildcard characters *
and ?, as well as for ( and ) and any other syntax-reserved characters?
Currently (well, ace4_3c), using "" to protect such characters in my
queries is not working.


3.  
> 3) possible member specific operations on tree classes?  e.g.  things to
> test interval overlap/containment for maps and sequences.  Perhaps we might
> even want interval objects?

Any furtherance of this would be good.  Actually we need a graphical 
interface to select and compare (multimap) subsections of Maps.


4. 
Should there be a flag to request case-dependent searches?  Both in text and
in classes that are defined as -CaseSensitive.


5. 
Will the old syntax still be available?  Or some query client with similar
syntax?  I can learn to write the new syntax but I'd hate to have to tell my 
users they have to learn it.


6. 
I hope the new version will sort lists correctly even if they're very long,
instead of piecewise like the current Tablemaker.

- Dave




More information about the Acedb mailing list