ParseString2List

Function: list ParseString2List(string src, list separators, list spacers, integer ParseStringKeepNulls);

Returns a list that is src broken into a list, discarding separators, keeping spacers.

if ParseStringKeepNulls == FALSE

Same as llParseString2List, but not limited to 8 spacers or separators.

Thus substitute a call to the llParseString2List function by a call to ParseString2List whenever you have more than 8 separators or more than 8 spacers.

if ParseStringKeepNulls != FALSE

Same as llParseStringKeepNulls, but not limited to 8 spacers or separators.

Thus substitute a call to the llParseStringKeepNulls function by a call to ParseString2List whenever you have more than 8 separators or more than 8 spacers.

Faster

This version is slightly faster but uses a bit more bytecode. <lsl>list ParseString2List(string src, list separators, list spacers, integer ParseStringKeepNulls) {//works just like llParseString2List and llParseStringKeepNulls

//Instead of each list being limited to 8 items, it is now 1024.
//The max length of src is 2,097,151 bytes.
   integer i = ~(separators != []);
   integer r = (spacers != []);
   spacers += separators;
   list out = "" + (separators = []);
   string p;
   integer offset;
   while((i = -~i) < r)
       if(!~llListFindList(out, (list)(p = llList2String(spacers, i))))
           if(~(offset = llSubStringIndex(src, p)))
           {
               separators += ((offset + 0xFFF00000) << 11) | (i + 0x400);
               out += p;
           }
   out = [];
   offset = 0xFFF00000;
   while(separators != [])//Can't use just "while(separators)" because of JIRA:SVC-689
   {
       if(offset ^ (i = ((r = llList2Integer(separators = llListSort(separators, 1, TRUE), 0)) >> 11)) || ParseStringKeepNulls)
           out += llDeleteSubString(src, i - offset, -1);
       src = llDeleteSubString(src, 0, ~(offset - (i += llStringLength(p = llList2String(spacers, r - 0x400)))));
       if(r & 0x400)
           out += p;
       offset = i;
       do
       {
           if(~(i = llSubStringIndex(src, p)))
               separators = llDeleteSubList(separators, 0, 0) + ((i + offset) << 11) | (0x7FF & r);
           else if((separators = llDeleteSubList(separators, 0, 0)) == [])//UGLY
               jump out;
       } while(offset > ((r = llList2Integer(separators, 0)) >> 11));
   }
   @out;
   if(src != "" || ParseStringKeepNulls)
       out += src;
   return out;

}//Strife Onizuka</lsl>

Slower

This version is slightly slower but uses less bytecode than the other. <lsl>list ParseString2List(string src, list separators, list spacers, integer ParseStringKeepNulls) {//works just like llParseString2List and llParseStringKeepNulls

//Instead of each list being limited to 8 items, it is now 1024.
//The max length of src is 2,097,151 bytes.
   integer i = ~(separators != []);
   integer r = (spacers != []);
   spacers += separators;
   list out = "" + (separators = []);
   string p;
   integer offset;
   while((i = -~i) < r)
       if(!~llListFindList(out, (list)(p = llList2String(spacers, i))))
           if(~(offset = llSubStringIndex(src, p)))
           {
               separators += ((offset + 0xFFF00000) << 11) | (i + 0x400);
               out += p;
           }
   out = [];
   offset = 0xFFF00000;
   while(separators != [])//Can't use just "while(separators)" because of JIRA:SVC-689
   {
       p = llList2String(spacers, (r = llList2Integer(separators = llListSort(separators, 1, TRUE), 0)) - 0x400);
       if(offset <= (i = (r >> 11)))
       {
           if(offset ^ i || ParseStringKeepNulls)
               out += llDeleteSubString(src, i - offset, -1);
           if(r & 0x400)
               out += p;
           src = llDeleteSubString(src, 0, ~(offset - (i += llStringLength(p))));
           offset = i;
       }
       separators = llDeleteSubList(separators, 0, 0);
       if(~(i = llSubStringIndex(src, p)))
           separators += ((i + offset) << 11) | (r & 0x7FF);
   }
   if(src != "" || ParseStringKeepNulls)
       out += src;
   return out;

}//Strife Onizuka</lsl>

<lsl>//Use for testing the function. string test(string src, list separators, list spacers, integer nulls) {

   list t = [];
   if(nulls)
       t = llParseStringKeepNulls(src, separators, spacers);
   else
       t = llParseString2List(src, separators, spacers);
   string a = llList2CSV(t);
   string b = llList2CSV(ParseString2List(src, separators, spacers, nulls));
   return (string)(a==b) + " : " + a + "            " + b;

}

default {

   state_entry()
   {
       llOwnerSay("----------------  " + (string)llGetFreeMemory());
       llOwnerSay(test("abcdefg", ["b"], ["b"], FALSE));
       llOwnerSay(test("abcdefg", ["b"], ["bc"], FALSE));
       llOwnerSay(test("abcdefg", ["bc"], ["b"], FALSE));
       llOwnerSay(test("abcdefg", ["b"], ["ab"], FALSE));
       llOwnerSay(test("abcdefg", ["b", "g"], ["ab"], FALSE));
       llOwnerSay(test("abcdefg", ["b"], ["ab", "g"], FALSE));
       llOwnerSay(test("abcdefg", ["b", "a"], ["a", "b"], FALSE));
       llOwnerSay(test("abcdefg", ["a", "b"], ["b", "a"], FALSE));
       llOwnerSay(test("abcdefg", ["b", "c"], ["a", "b"], FALSE));
       llOwnerSay(test("abcdefg", ["c", "b"], ["b", "a"], FALSE));
       llOwnerSay(test("abcdefg", ["b", "a"], ["c", "b"], FALSE));
       llOwnerSay(test("abcdefg", ["a", "b"], ["b", "c"], FALSE));
       llOwnerSay(test("abcdefg", ["b"], ["b"], TRUE));
       llOwnerSay(test("abcdefg", ["b"], ["bc"], TRUE));
       llOwnerSay(test("abcdefg", ["bc"], ["b"], TRUE));
       llOwnerSay(test("abcdefg", ["b"], ["ab"], TRUE));
       llOwnerSay(test("abcdefg", ["b", "g"], ["ab"], TRUE));
       llOwnerSay(test("abcdefg", ["b"], ["ab", "g"], TRUE));
       llOwnerSay("----------------  " + (string)llGetFreeMemory());
   }

}</lsl>

See also: Script Library - Separate Words