[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [patch #5466] Encapsulated procedure.c
From: |
Ben Pfaff |
Subject: |
Re: [patch #5466] Encapsulated procedure.c |
Date: |
Thu, 19 Oct 2006 21:40:35 -0700 |
User-agent: |
Gnus/5.110006 (No Gnus v0.6) Emacs/21.4 (gnu/linux) |
John Darrington <address@hidden> writes:
> On Thu, Oct 19, 2006 at 02:03:25PM +0000, Ben Pfaff wrote:
>
> Follow-up Comment #1, patch #5466 (project pspp):
>
> This is clearly an improvement. It looks good. Thank you for writing
> it.
> If it passes "make check", please check it in.
>
> Done.
>
> I'm not actually sure it makes the DATASET stuff easier, though. I
> thought I
> had concluded that those commands were essentially equivalent to PSPP
> scratch
> files (which I think we should drop once the DATASET commands are
> working).
>
> It seems that we have different ideas of what the DATASET commands
> actually do. As I understand it, a DATASET is a completely new
> operating environment. Each dataset has its own dictionary, active file,
> transformation set etc, etc. The commands allow you to create new
> ones, and swtich between existing ones.
I was under the impression that, when you switch active datasets,
pending transformations get executed, as if the EXECUTE command
had been run. Effectively, this is like saving the dictionary
and active file.
I haven't studied the DATASET commands in detail, though, so this
might be a wrong first impression.
--
"If a person keeps faithfully busy each hour of the working day, he
can count on waking up some morning to find himself one of the
competent ones of his generation."
--William James