It would be nice if you could address columns by number, though.
For example, if you have a stored procedure which you cannot change the
definition of, and it returns columns with invalid or duplicate names (i.e. CF
cannot refer to them at all).
Jon Hall wrote:
>
> Thank you Jim! You realize though that you have said Ben Forta is wrong...I
> wouldn't be standing in the middle of any fields during a thunderstorm if I
> were you. :-)
>
> jon
> ----- Original Message -----
> From: "Jim McAtee" <[EMAIL PROTECTED]>
> To: "CF-Talk" <[EMAIL PROTECTED]>
> Sent: Friday, October 06, 2000 3:21 AM
> Subject: Re: accessing a query as an array
>
> > > I am trying to populate an array from a query and I am running into =
> > > problems...
> > >
> > > From what I have infered from the CF docs, CFQuery's are actually =
> > > arrays.=20
> > > If we assume that then, #QueryName[1][4]# should return the data in the
> =
> > > first row and the fourth column......right??
> > >
> > > My confusion comes from what seems to me to be conflicting info in Ben =
> > > Forta's book and the CF online docs.=20
> > > Ben states on page 529 (4.0) that "Database queries returned by the =
> > > <CFQUERY> tag are actually 2D arrays..."
> > > The online docs say that "Query column data can be referenced using =
> > > array-like syntax. For example, myquery.col_name[1] references data in =
> > > the first row in the column col_name."
> > >
> > > If the latter is true then it seems to me that queries are not acutally
> =
> > > true 2D arrays but "arraylike" as it says...
> >
> >
> > Internally, queries are probably stored very much like two-dimensional
> > arrays. I don't believe CF lets you address them like this, however. One
> > of the problems is that there's no set order for the columns returned by a
> > SQL query. That should be a bit more obvious when you do a SELECT *.
> Which
> > database column would you expect the fourth column of the array to be?
> >
> > CF _will_ let you address each column as a one-dimensional array, as
> you've
> > shown. Doing this lets you address the Nth row of the query randomly,
> > rather than having to step through the query using <cfloop>.
> >
> > One simple way to populate a two-d array from a query is shown below.
> This
> > way, you _know_ that 'name' is the first column, 'address' is the seconde,
> > etc.
> >
> > <cfset a = ArrayNew(2)>
> > <cfloop query="myquery">
> > <cfset a[myquery.currentrow, 1] = myquery.name>
> > <cfset a[myquery.currentrow, 2] = myquery.address>
> > <cfset a[myquery.currentrow, 3] = myquery.city>
> > <cfset a[myquery.currentrow, 4] = myquery.state>
> > <cfset a[myquery.currentrow, 5] = myquery.zip>
> > </cfloop>
> >
> > Depending on what you need to do with your array, you might also use an
> > array of structs to keep the query results. An array of structs lets you
> > address the columns by name, rather than by column number. Then again,
> > there's no reason why you can't leave the data in the original CF query.
> > There are CF function to let you manipulate the query data by adding rows,
> > adding columns, and changing values.
> >
> > Jim
> >
> > --------------------------------------------------------------------------
> ----
> > Archives: http://www.mail-archive.com/[email protected]/
> > To Unsubscribe visit
> http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or
> send a message to [EMAIL PROTECTED] with 'unsubscribe' in
> the body.
>
> ------------------------------------------------------------------------------
> Archives: http://www.mail-archive.com/[email protected]/
> To Unsubscribe visit
>http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
>message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
--
David Cummins
Systems Developer
Ubiquity Software Ltd.
Ph: (09) 309-1921
------------------------------------------------------------------------------
Archives: http://www.mail-archive.com/[email protected]/
To Unsubscribe visit
http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk or send a
message to [EMAIL PROTECTED] with 'unsubscribe' in the body.
