File: BUGS

package info (click to toggle)
geda-gnetlist 20050313-1
  • links: PTS
  • area: main
  • in suites: sarge
  • size: 2,092 kB
  • ctags: 702
  • sloc: lisp: 5,519; ansic: 3,809; sh: 3,647; makefile: 291
file content (35 lines) | stat: -rw-r--r-- 1,218 bytes parent folder | download | duplicates (5)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

Bugs (no particular order): 
----------------------

	- If a component cannot be read, then this situation is not dealt 
	  with.  Technically at this point, gnetlist can't do anything so 
	  it really should exit.  Add code to do this later...

	- memory leaks (lots right now) A LOT MORE now!

	- Ordering of pin outputs in spice netlist ouput for components like 
	  BJT transistors B C E.  This is wrong for now. (might be fixed now)

	- (SPICE backend) It is possible to have named a net something (say 1)
	  and then have an unnamed net called 1 as well.  Obviously my
	  unnamed net naming is lame.

	- gnetlist:get-toplevel-attribute only looks for the first occurrance
	  of the toplevel attribute.  Make this somehow better?

	- Figure out why the post process screwed up net names when it was
	  done in the wrong place (after every sheet), low priority
	
	- gnetlist doesn't handle order very well on some machines?
	  non-linux only? 

Finished stuff:

FIXED	embedded schematics don't netlist at all.

OKAY	infinite net connection loops

FIXED  	attach an label to a net which is going to no pin and it won't
        show up. (this is because mid point connections are not being
	dealt with properly...)