comp.lang.idl-pvwave archive: archive » Avoiding multiple or complex WHEREs?

Home » Public Forums » archive » Avoiding multiple or complex WHEREs?

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Avoiding multiple or complex WHEREs? [message #75224]

Mon, 21 February 2011 13:31

cgguido
Messages: 195
Registered: August 2005

Senior Member

;Say I have an array of point IDs and times where p[0,*] are IDs and
;p[1,*] are times, something like:

p=[[1,1],[1,2],[2,3]]

;Typically, the array has many points, many times and many gaps
;I want to be able to get a subset of 'p' that contains a range
;[id0:idn] or a list [id0,id2,id7] of IDs and Ts the most efficient
;way possible.

;Right now, I build an array 'i' with one FOR loop where i[id,t] is
the row index to 'p'
;with the correct id and t. If that combinations does not exist, then
;i[id,t]=NaN. So:

i[0,0]=NaN
i[2,0]=NaN
i[1,*]=[NaN,0,1,NaN]

;So if I want all instances of particle 1 in array p I do

i1=i[1,where(i[1,*] ge 0)]
p1=p[*,i1]

;This gets handy when I want particles 1,3,4 and times [23:100],
;because I don't have to do where(id eq 1 OR id eq 3 OR id EQ 4....
etc.)

i_cool=i[1,where(i[[1,3,4],[23:100]] ge 0)]
p_cool=p[*,i_cool]

;Now, I am betting that a histogram/histobin trick might be even
;better... Any ideas?

;PS: This does not deal nicely with IDs or Ts that are bigger than the
;biggest contained in 'p'

Report message to a moderator

[Message index]

		Avoiding multiple or complex WHEREs? By: cgguido on Mon, 21 February 2011 13:31
		Re: Avoiding multiple or complex WHEREs? By: cgguido on Tue, 22 February 2011 12:02

Previous Topic:	Re: Extract numbers from text string
Next Topic:	Re: howto select and deselect plotting windows?

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sat Oct 11 12:39:30 PDT 2025

Total time taken to generate the page: 1.52087 seconds