updates

include/amsstring4/amsstring4.hpp

@@ -41,6 +41,12 @@ static const ams_chartype ams_char_nt = (ams_chartype) '\0'; //null terminator
 static const ams_chartype ams_char_dq = (ams_chartype) '"'; //doublequote
 static const ams_chartype ams_char_sq = (ams_chartype) '\''; //singlequote
 
+ams_chartype tolower(ams_chartype c);
+ams_chartype toupper(ams_chartype c);
+bool amsstring_isspace(ams_chartype c);
+int amsstring_strlen(const ams_chartype *c);
+bool amsstring_ischarnumeric(const ams_chartype c);
+
 
 class amsstring
 {
@@ -191,12 +197,12 @@ amsstring select_between(const amsstring &s, const ams_chartype delimitleft, con
 
 //selects between the outermost set of delimitleft and delimitright strings
 //if delimitleft or delimitright strings are empty, selects the left or right end of the array
-amsstring select_between(const amsstring &s, const ams_chartype *delimitleft, const ams_chartype *delimitright); //
+amsstring select_between(const amsstring &s, const ams_chartype *delimitleft, const ams_chartype *delimitright); 
 
 amsstring concatenate(const amsarray<amsstring> &slist);
 
 //adds \n between each line in the concatenated string
-amsstring concat_lines(const amsarray<amsstring> &lns);//
+amsstring concat_lines(const amsarray<amsstring> &lns);
 
 
 

include/amsstring4/amsstring4_tests.hpp

@@ -20,6 +20,7 @@ namespace ams
     void amsstring4_test_concatenation_operators();
     
     void amsstring4_test_convenience1a();
+    void amsstring4_test_convenience1b();
 
 };
 

src/amsstring4/amsstring4_class.cpp

@@ -3,7 +3,20 @@
 namespace ams
 {
 
-    ams_chartype tolower(ams_chartype c)
+    //isspace() is locale dependent. For determinism's sake, I am going to implement my own checks.
+    /* Returns true for the 6 ASCII white-space characters: \t \n \v \f \r ' '. */
+    bool amsstring_isspace(const ams_chartype c)
+    {
+      bool ret = 0;
+      //isspace(c)||c==' '||c=='\t'||c=='\n'||c=='\r'||c=='\f'||c=='\v'||(int)c==9||(int)c==10||(int)c==11||(int)c==12
+      //ret = (c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r' || c == ' ');
+      
+      ret = ((c == '\t') || (c == '\n') || (c == '\v') || (c == '\f') || (c == '\r') || (c == ' ') || ((int)c==10) || ((int)c==11) || ((int)c==12) );
+      
+      return ret;
+    }
+
+    ams_chartype tolower(const ams_chartype c)
     {
       unsigned char c2 = (unsigned char) c;
       ams_chartype ret;
@@ -18,7 +31,7 @@ namespace ams
       return ret;
     }
     
-    ams_chartype toupper(ams_chartype c)
+    ams_chartype toupper(const ams_chartype c)
     {
       unsigned char c2 = (unsigned char) c;
       ams_chartype ret;
@@ -33,7 +46,17 @@ namespace ams
       return ret;
     }
 
-    static int amsstring_strlen(const ams_chartype *c)
+    //isspace() is locale dependent. For determinism's sake, I am going to implement my own checks.
+    //returns true if a character is of a type used forming a decimal number
+    bool amsstring_ischarnumeric(const ams_chartype c)
+    {
+      ams_chartype cl = ams::tolower(c);
+      bool ret = 0;
+      ret = ret || ( ((int)cl>=48) && ((int)cl<=57) ) || cl=='.' || cl=='e' || '-';
+      return ret;
+    }
+    
+    int amsstring_strlen(const ams_chartype *c)
     {
       int I;
       if(c!=NULL)
@@ -1138,13 +1161,15 @@ namespace ams
         while(qf==0)
         {
           c = s->cstring[I];
-          if(!(isspace(c)||c==' '||c=='\t')&&mode==0)
+          //if(!(isspace(c)||c==' '||c=='\t')&&mode==0)
+          if(!(amsstring_isspace(c))&&mode==0)
           {
             mode = 1;
             ind1 = I;
             ind1 = ams::max<int>(0,ind1);
           }
-          if((isspace(c)||c==' '||c=='\t'||c=='\0')&&mode==1)
+          //if((isspace(c)||c==' '||c=='\t'||c=='\0')&&mode==1)
+          if((amsstring_isspace(c))&&mode==1)
           {
             mode = 0;
             ind2 = I;
@@ -1179,7 +1204,8 @@ namespace ams
         for(I=0;I<s->length;I++)
         {
           c = s->cstring[I];
-          if(!(isspace(c)||c==' '||c=='\t'))
+          //if(!(isspace(c)||c==' '||c=='\t'))
+          if(!(amsstring_isspace(c)))
           {
             ind1 = I;
             break;
@@ -1188,7 +1214,8 @@ namespace ams
         for(I=s->length-1;I>=0;I--)
         {
           c = s->cstring[I];
-          if(!(isspace(c)||c==' '||c=='\t'))
+          //if(!(isspace(c)||c==' '||c=='\t'))
+          if(!(amsstring_isspace(c)))
           {
             ind2 = I+1;
             break;
@@ -1219,7 +1246,8 @@ namespace ams
       for(I=0;I<s->length;I++)
       {
         c = s->cstring[I];
-        if(!(isspace(c)||c==' '||c=='\t'||c=='\n'||c=='\r'||c=='\f'||c=='\v'||(int)c==9||(int)c==10||(int)c==11||(int)c==12))
+        //if(!(isspace(c)||c==' '||c=='\t'||c=='\n'||c=='\r'||c=='\f'||c=='\v'||(int)c==9||(int)c==10||(int)c==11||(int)c==12))
+        if(!(amsstring_isspace(c)))
         {
           q.cstring[J] = s->cstring[I];
           J = J + 1;

src/amsstring4/amsstring4_convenience1.cpp

@@ -94,22 +94,22 @@ amsarray<amsstring> split(const amsstring &s, const ams_chartype delimitchar)
     return ret;
 }
 
-static int amsstring_strlen(const ams_chartype *c)
-{
-  int I;
-  if(c!=NULL)
-  {
-    I = 0;
-    while(c[I]!='\0')
-    {
-      I++;
-    }
-    return I;
-    //return strlen((const char*) c);
-  }
-  else
-    return 0;
-}
+// static int amsstring_strlen(const ams_chartype *c)
+// {
+//   int I;
+//   if(c!=NULL)
+//   {
+//     I = 0;
+//     while(c[I]!='\0')
+//     {
+//       I++;
+//     }
+//     return I;
+//     //return strlen((const char*) c);
+//   }
+//   else
+//     return 0;
+// }
 
 amsarray<amsstring> split(const amsstring &s, const ams_chartype *delimitstr)
 {
@@ -255,13 +255,15 @@ amsarray<amsstring> splitwhitespace(const amsstring &s)
     while(qf==0)
     {
         c = s.cstring[I];
-        if(!(isspace(c)||c==' '||c=='\t')&&mode==0)
+        //if(!(isspace(c)||c==' '||c=='\t')&&mode==0)
+        if(!(amsstring_isspace(c))&&mode==0)
         {
             mode = 1;
             ind1 = I;
             ind1 = ams::max<int>(0,ind1);
         }
-        if((isspace(c)||c==' '||c=='\t'||c=='\0')&&mode==1)
+        //if((isspace(c)||c==' '||c=='\t'||c=='\0')&&mode==1)
+        if((amsstring_isspace(c))&&mode==1)
         {
             mode = 0;
             ind2 = I;
@@ -295,7 +297,8 @@ amsstring stripwhitespace(const amsstring &s)
     for(I=0;I<s.length;I++)
     {
         c = s.cstring[I];
-        if(!(isspace(c)||c==' '||c=='\t'))
+        //if(!(isspace(c)||c==' '||c=='\t'))
+        if(!(amsstring_isspace(c)))
         {
             ind1 = I;
             break;
@@ -304,7 +307,8 @@ amsstring stripwhitespace(const amsstring &s)
         for(I=s.length-1;I>=0;I--)
         {
         c = s.cstring[I];
-        if(!(isspace(c)||c==' '||c=='\t'))
+        //if(!(isspace(c)||c==' '||c=='\t'))
+        if(!(amsstring_isspace(c)))
         {
             ind2 = I+1;
             break;
@@ -335,7 +339,8 @@ amsstring stripallwhitespace(const amsstring &s)
     for(I=0;I<s.length;I++)
     {
       c = s.cstring[I];
-      if(!(isspace(c)||c==' '||c=='\t'||c=='\n'||c=='\r'||c=='\f'||c=='\v'||(int)c==9||(int)c==10||(int)c==11||(int)c==12))
+      //if(!(isspace(c)||c==' '||c=='\t'||c=='\n'||c=='\r'||c=='\f'||c=='\v'||(int)c==9||(int)c==10||(int)c==11||(int)c==12))
+      if(!(amsstring_isspace(c)))
       {
         ret.cstring[J] = s.cstring[I];
         J = J + 1;
@@ -426,6 +431,32 @@ amsstring select_between(const amsstring &s, const ams_chartype delimitleft, con
     return ret;
 }
 
+//selects between the outermost set of delimitleft and delimitright chars
+amsstring select_between(const amsstring &s, const ams_chartype *delimitleft, const ams_chartype *delimitright)
+{
+    amsstring ret;
+
+    int ind1,ind2;
+    int slen1,slen2;
+
+    slen1 = amsstring_strlen(delimitleft);
+    slen2 = amsstring_strlen(delimitleft);
+    
+    ind1 = -1;
+    ind2 = -1;
+    if(slen1>0)
+      ind1 = s.find(delimitleft);
+    if(slen2>0)
+      ind2 = s.findright(delimitright);
+
+    if(ind1<0) ind1=0;
+    if(ind2<0) ind2=s.length;
+
+    ret = s.substring(ind1,ind2);
+
+    return ret;
+}
+
 amsstring concatenate(const amsarray<amsstring> &slist)
 {
     amsstring ret;
@@ -440,6 +471,60 @@ amsstring concatenate(const amsarray<amsstring> &slist)
     return ret;
 }
 
+amsstring concat_lines(const amsarray<amsstring> &slist)
+{
+    amsstring ret;
+    amsstring s;
+    int I;
+    ret.resize(0);
+    for(I=0;I<slist.length;I++)
+    {
+        s = slist[I];
+        s=s+"\n"; //maybe check to see if a newline already exists for this line?
+        ret.append(s);
+    }
+
+    return ret;
+}
+
+
+//splits a string into contiguous sets of alphanumeric characters delimited by any non-alphanum chars
+//for eventual conversion of space,tab,comma separated numbers into numeric tuples and arrays
+//This step does not check that each string is a valid number.
+amsarray<amsstring> splitalphanum(const amsstring &s)
+{
+  amsarray<amsstring> sa;
+  int I0,I1,I;
+  int mode;
+  ams_chartype c;
+  amsstring s2;
+
+  mode = 0;
+  for(I=0;I<s.length;I++)
+  {
+    c = s[I];
+    if(mode==0 && amsstring_ischarnumeric(c))
+    {
+      mode = 1;
+      I0 = I;
+    }
+
+    if(mode==1 && (!amsstring_ischarnumeric(c) || I==length-1)
+    {
+      I1 = I;
+      s2 = s.substring(I0,I1);
+      sa.append(s2);
+      mode = 0;
+    }
+  }
+
+  return sa;
+}
+
+/////////////
+// Testing //
+/////////////
+
 void amsstring4_test_findright()
 {
   amsstring s1,s2;
@@ -594,6 +679,10 @@ void amsstring4_test_convenience1a()
   s1 = "/some/other/path";
   sa = splitpath(s1);
   printf("splitpath('%s') = ['%s','%s']\n",s1.cstring,sa[0].cstring,sa[1].cstring);
+}   
+
+void amsstring4_test_convenience1b()
+{
 
 
 }

src/main.cpp

@@ -22,7 +22,8 @@ int main(int argc, char* argv[])
 
     // amsstring4_test_concatenation_operators(); 
 
-    amsstring4_test_convenience1a();
+    //amsstring4_test_convenience1a();
+    amsstring4_test_convenience1b();
     
     return ret;
 }