genius r752 - trunk
- From: jirka svn gnome org
- To: svn-commits-list gnome org
- Subject: genius r752 - trunk
- Date: Fri, 20 Mar 2009 15:12:57 +0000 (UTC)
Author: jirka
Date: Fri Mar 20 15:12:57 2009
New Revision: 752
URL: http://svn.gnome.org/viewvc/genius?rev=752&view=rev
Log:
update TODO
-Jiri
Modified:
trunk/TODO
Modified: trunk/TODO
==============================================================================
--- trunk/TODO (original)
+++ trunk/TODO Fri Mar 20 15:12:57 2009
@@ -1,59 +1,52 @@
* Logfiles and/or save session
* Whack the readline nonsense in the GUI
-* Check this strange thing of locals living on the global context somehow,
- try the testprogfamily.gel and look at vars and notice there are things
- with context > 0 on the toplevel context. It's strange. And I can't
- repro it now
-
+* Native polynomial operation
+* Add continued fraction type of a loop
+* Add user argument checking functions for parameters (to make calling GEL
+ functions that check their arguments faster)
+* fix matrix exp function and add matrix ln function
+* matrix powers for rational numbers, and scalar^matrix (as matlab does it)
+* Graph of 1/x or atan must not have vertical lines. Futhermore maybe we
+ should perhaps think about removing the vertical line if two consecutive
+ points are much further from each other than the previous two points on the
+ graph. This would graph the step function correctly.
* Implement max_nodes in the command line version
* Implement MaxNodes parameter
+* Kill plugin interface maybe? It is just a pain to maintain and no one uses
+ AFAIK. It was originally done to make Dr.Genius possible, but that's long
+ over.
+Possibly old
+------------
-This is somewhat out of date, though some of these may be fine ideas still.
-
----------------
-
-* Make mod arithmetic propagation into user functions possible
+* Check this strange thing of locals living on the global context somehow,
+ try the testprogfamily.gel and look at vars and notice there are things
+ with context > 0 on the toplevel context. It's strange. And I can't
+ repro it now
-* Gauss needs to be mod aware
+This is out of date
+-------------------
-* Add flags to arguments to function to not-mod them where appropriate
- (for example I(n))
+* Gauss needs to be mod aware (isn't it now?)
* Check variable substitution of references, that could refer
to freed data I think
-* Add continued fraction type of a loop
-
* Check rational powers in mod mode to work correctly
* Check modular arithmetic to be correct (such as: "expr mod n"),
that is, when do we mod is the question.
-* Add extended gcd
-
* Perhaps latex and troff mode should not be used on error output,
only on toplevel. Alternantively everything should be latex/troffized
I suppose. I like the first alternative more though.
-* Native polynomial operation
-
-* Add user argument checking functions for parameters
-
-* fix matrix exp function and add matrix ln function
-
-* matrix powers for rational numbers, and scalar^matrix (as matlab
- does it)
-
* add try/catch operator for exceptions (need special exception
variable like return_val)
* interrupt internal mpwrap calculation (need new error type)
-* result display (optional) ... for matrix display and normal
- numbers ...
-
* cleanup engine
* make one line expression output more sane and make it work right
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]