RE to remove ,

Some parts of this are a hack, but it works. You might want to explore regular expressions for this as well.

Text version of the code:

axlCmdRegister("parsenetstring" 'ParseNetString)

procedure(ParseNetString( )
let(((b_validIndex t)
(s_inPutString "")
(l_output list())
(l_temp list())
(s_curChar "")
(b_indexFound nil)
)

s_inPutString = "signal_A{2,4,6,8:10},signal_B,signal_C[10,12:14,16]_E"

for(i 1 strlen(s_inPutString)
b_indexFound = nil
s_curChar = buildString(list(getchar(s_inPutString i)))

;Ignore any commas inside brackets
if(s_curChar == "{" || s_curChar == "[" then
b_validIndex = nil
)

;Character is outside of brackets, so it could be a valid comma
if(s_curChar == "}" || s_curChar == "]" then
b_validIndex = t
)

;The current character could be a valid comma
if(b_validIndex then
;Valid comma found. Store the
if(s_curChar == "," then
b_indexFound = t
l_output = cons(buildString(reverse(l_temp) "") l_output)
l_temp = list()
)
)

;Couldn't find an easy way to remove a portion of a string by a known index so we are
; just creating a list of chars which we will build a new string from when a valid
; comma is found
if(!b_indexFound then
l_temp = cons(s_curChar l_temp)
)

;We are at the end of the string, so add the final value to the list
if(i==strlen(s_inPutString) then
l_output = cons(buildString(reverse(l_temp) "") l_output)
)
i++
)

l_output = reverse(l_output)

println(l_output)

)
)

Some parts of this are a hack, but it works. You might want to explore regular expressions for this as well.

Text version of the code:

axlCmdRegister("parsenetstring" 'ParseNetString)

procedure(ParseNetString( )
let(((b_validIndex t)
(s_inPutString "")
(l_output list())
(l_temp list())
(s_curChar "")
(b_indexFound nil)
)

s_inPutString = "signal_A{2,4,6,8:10},signal_B,signal_C[10,12:14,16]_E"

for(i 1 strlen(s_inPutString)
b_indexFound = nil
s_curChar = buildString(list(getchar(s_inPutString i)))

;Ignore any commas inside brackets
if(s_curChar == "{" || s_curChar == "[" then
b_validIndex = nil
)

;Character is outside of brackets, so it could be a valid comma
if(s_curChar == "}" || s_curChar == "]" then
b_validIndex = t
)

;The current character could be a valid comma
if(b_validIndex then
;Valid comma found. Store the
if(s_curChar == "," then
b_indexFound = t
l_output = cons(buildString(reverse(l_temp) "") l_output)
l_temp = list()
)
)

;Couldn't find an easy way to remove a portion of a string by a known index so we are
; just creating a list of chars which we will build a new string from when a valid
; comma is found
if(!b_indexFound then
l_temp = cons(s_curChar l_temp)
)

;We are at the end of the string, so add the final value to the list
if(i==strlen(s_inPutString) then
l_output = cons(buildString(reverse(l_temp) "") l_output)
)
i++
)

l_output = reverse(l_output)

println(l_output)

)
)

Hi mstaub, it works and looks good, thx.

yes, using rex functions can be an alternate option,

and just to share a workaround  i can think so far:

1. Use rexCompile and rexExecute to take out (first one Only) the substring of { } and [ ]

2. Use rexReplace all the " , " from 1 with a string of (non meta) characters that can never be found in a signal name. (say,  "___") 

3. Use rexReplace (first One Only) to put 2 back into the initial string.

4. Use while loop for 1-3, and the rexExecute in 1 as a loop checker. 

5. after 4, parseString with " , " and replace each "___" with " ,  "

Hi !

Another way to do that. Try below code.

(defun test ()
inPutString = "signal_A{2,4,6,8:10},signal_B,signal_C[10,12:14,16]_E"
out_str = nil
foreach(item reverse(parseString(inPutString "signal"))
item_len = strlen(item)
if(nindex(item ",") == item_len then
tmp_str = strcat("signal" subs_tring(item 1 item_len-1))
else
tmp_str = strcat("signal" item)
);end if
out_str = cons(tmp_str out_str)
);end foreach
printf("OUT: %L\n", out_str)
);end defun

Note: please remove underline on command sub_string to run code. I can't post without underline on that command.

Luan.

I had looked into using 'parseString(inPutString "signal")' as well but you making the assumption that 'signal' will always be the prefix. Which could be the case, and if so that would work.