Changeset 1231 for cpp/frams/genetics/f4/f4_general.cpp

Timestamp:

05/02/23 01:36:15 (2 years ago)

Author:

Maciej Komosinski

Message:

• Thanks to r1230, it is possible to detect (and repair=remove) junk trailing genes that are left after successful parsing (after last '>')
• The validate() function may attempt to repair a genotype where earlier it would give up
• Stricter parsing of the '#' gene

File:

• cpp/frams/genetics/f4/f4_general.cpp (modified) (8 diffs)

cpp/frams/genetics/f4/f4_general.cpp

@@ -640 +640 @@
 
 
-f4_Cells::f4_Cells(f4_Node *genome, int nrepair)
-{
-        // create ancestor cell
+f4_Cells::f4_Cells(f4_Node *genome, bool nrepair)
+{
         repair = nrepair;
         errorcode = GENOPER_OK;
@@ -650 +649 @@
         repair_insert = NULL;
         tmpcel = NULL;
-        f4rootnode = NULL;
+
+        // create ancestor cell
         C[0] = new f4_Cell(this, 0, genome, genome, NULL, 0, GeneProps::standard_values);
         cell_count = 1;
@@ -656 +656 @@
 
 
-f4_Cells::f4_Cells(SString & genome, int nrepair)
-{
-        repair = nrepair;
-        errorcode = GENOPER_OK;
-        errorpos = -1;
-        repair_remove = NULL;
-        repair_parent = NULL;
-        repair_insert = NULL;
-        tmpcel = NULL;
-        f4rootnode = NULL;
-
-        // transform geno from string to nodes
-        f4rootnode = new f4_Node();
-        int _ = 0;
-        int res = f4_processRecur(genome.c_str(), _, f4rootnode);
-        if (res || (f4rootnode->childCount() != 1))
-        {
-                errorcode = GENOPER_OPFAIL;
-                errorpos = -1;
-        }
-
-        // create ancestor cell
-        C[0] = new f4_Cell(this, 0, f4rootnode->child, f4rootnode->child, NULL, 0, GeneProps::standard_values);
-        cell_count = 1;
-}
 
 f4_Cells::~f4_Cells()
 {
         // release cells
-        int i;
         if (cell_count)
         {
-                for (i = cell_count - 1; i >= 0; i--)
+                for (int i = cell_count - 1; i >= 0; i--)
                         delete C[i];
                 cell_count = 0;
         }
-        if (f4rootnode)
-                delete f4rootnode;
 }
 
@@ -1130 +1102 @@
 int f4_Node::childCount()
 {
-        if (child != NULL)
-        {
-                if (child2 != NULL) return 2;
-                else return 1;
-        }
-        else
-        {
-                if (child2 != NULL) return 1;
-                else return 0;
-        }
+        return int(child != NULL) + int(child2 != NULL); //0, 1 or 2
 }
 
@@ -1256 +1219 @@
         len = out.length();
         if (len > 1)
-                if (out[len - 1] == '>') { (out.directWrite())[len - 1] = 0; out.endWrite(); }; //Macko 2023-04 TODO "can be omitted", but should we remove it as a rule even in generated genotypes? see if I can somehow detect junk characters after top-level '>' ends properly: /*4*/<X>N:N>whatever
+                if (out[len - 1] == '>') { (out.directWrite())[len - 1] = 0; out.endWrite(); }; //Macko 2023-04 "can be omitted", but it is removed as a rule even in generated genotypes :)
         // copy back to string
         // if new is longer, reallocate buf
@@ -1305 +1268 @@
         while (pos_inout < (int)strlen(genot))
         {
-                //#define PRINT_PARSING_LOCATION
+//#define PRINT_PARSING_LOCATION
 #ifdef PRINT_PARSING_LOCATION
                 printf("%s\n", genot);
@@ -1342 +1305 @@
                 case '#':
                 {
-                        // repetition marker, 1 by default
-                        ExtValue val;
-                        const char* end = val.parseNumber(genot + pos_inout + 1, ExtPType::TInt);
-                        //TODO end==NULL? -> error!
-                        int reps = (end == NULL) ? 1 : val.getInt();
+                        // repetition marker
+                        ExtValue reps;
+                        const char* end = reps.parseNumber(genot + pos_inout + 1, ExtPType::TInt);
+                        if (end == NULL)
+                                return pos_inout + 1; //error
                         f4_Node *node = new f4_Node("#", par, pos_inout);
-                        node->reps = reps;
+                        node->reps = reps.getInt();
                         // skip number
                         pos_inout += end - (genot + pos_inout);
@@ -1455 +1418 @@
 }
 
+int f4_process(const char *genot, f4_Node *root)
+{
+        int pos = 0;
+        int res = f4_processRecur(genot, pos, root);
+        if (res > 0)
+                return res; //error
+        else if (genot[pos] == 0) //parsed until the end - OK!
+                return 0;
+        else return pos + 1; //junk, unparsed genes after successful parsing, for example /*4*/<X>N:N>whatever or /*4*/<X>X>>>
+}
+
 const char* parseConnection(const char *fragm, int& relfrom, double &weight)
 {