[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Octave-bug-tracker] [bug #55065] Implement uiopen and uisave
From: |
Nicholas Jankowski |
Subject: |
[Octave-bug-tracker] [bug #55065] Implement uiopen and uisave |
Date: |
Thu, 23 Mar 2023 14:59:04 -0400 (EDT) |
Follow-up Comment #7, bug #55065 (project octave):
@ristephens, these are actually in pretty good shape. a couple initial
thoughts:
uisave:
while i like the fact that applying an empty array as a placeholder for the
variable names ( uisave([], "filename") allows you to save all variables
without specifying them but still pre-name the file, i noticed that matlab
doesn't behave this way. if you pass [] as the variable names, it gives you a
"no variables to save" error.
whenever you don't specify a variable (either by uisave() or by the syntax
above with []), an "ans" variable is created in the workspace that appears to
be a cell array containing the variables saved to the file. This probably
should be suppressed somehow. (Matlab's function doesn't change anything in
the workspace.)
Last, while both programs give you an error if you specify a variable that
doesn't exist, your uisave still creates an empty mat file. (well, it is empty
of variables. it contains the "# Created by Octave ..." header line)
uiopen - no issues i could see with a quick check and a few tests. for this
file, since there's a user-interaction free syntax, maybe it would be possible
to write a few BISTs that create a few workspace variables then writes them to
a mat file and uiopens them with uiopen('filenam',true) ? would need to get
someone else's opinion on how temp file writes are best handled within the
test environment though.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?55065>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/