got parsing working using PCRE; updated testinput.i

parse-input.cc

@@ -43,6 +43,8 @@ bool parse_input(char * buff, int size)
     sections["tokens"] = tokens;
     sections["rules"] = rules;
     Section section = none;
+    string line;
+    bool append_line = false;
 
     for (int i = 0; i < sizeof(exprs)/sizeof(exprs[0]); i++)
     {
@@ -64,16 +66,36 @@ bool parse_input(char * buff, int size)
         {
             line_length--;
         }
-        string line(input, line_length);
+        if (append_line)
-        if (pcre_exec(empty, NULL, line.c_str(), line.size(), 0, 0,
-                    ovector, ovec_size) == 0)
-            continue;
-        if (pcre_exec(comment, NULL, line.c_str(), line.size(), 0, 0,
-                    ovector, ovec_size) == 0)
-            continue;
-        if (pcre_exec(section_name, NULL, line.c_str(), line.size(), 0, 0,
-                    ovector, ovec_size) == 0)
         {
+            line += string(input, line_length);
+        }
+        else
+        {
+            line = string(input, line_length);
+        }
+        if (line.size() > 0 && line[line.size()-1] == '\\')
+        {
+            line[line.size()-1] = ' ';
+            append_line = true;
+        }
+        else
+        {
+            append_line = false;
+        }
+        if ( append_line
+          || (pcre_exec(empty, NULL, line.c_str(), line.size(),
+                  0, 0, ovector, ovec_size) >= 0)
+          || (pcre_exec(comment, NULL, line.c_str(), line.size(),
+                  0, 0, ovector, ovec_size) >= 0)
+           )
+        {
+            /* nothing */;
+        }
+        else if (pcre_exec(section_name, NULL, line.c_str(), line.size(),
+                    0, 0, ovector, ovec_size) >= 0)
+        {
+            sn = string(input, ovector[2], ovector[3] - ovector[2]);
             if (sections.find(sn) != sections.end())
             {
                 section = sections[sn];
@@ -83,16 +105,17 @@ bool parse_input(char * buff, int size)
                 cerr << "Unknown section name '" << sn << "'!" << endl;
                 return false;
             }
-            continue;
         }
+        else
+        {
             switch (section)
             {
                 case none:
                     cerr << "Unrecognized input on line " << lineno << endl;
                     return false;
                 case tokens:
-                if (pcre_exec(token, NULL, line.c_str(), line.size(), 0, 0,
+                    if (pcre_exec(token, NULL, line.c_str(), line.size(),
-                            ovector, ovec_size) == 0)
+                                0, 0, ovector, ovec_size) >= 0)
                     {
                         string name(line, ovector[2], ovector[3] - ovector[2]);
                         string definition(line,
@@ -112,8 +135,8 @@ bool parse_input(char * buff, int size)
                     }
                     break;
                 case rules:
-                if (pcre_exec(rule, NULL, line.c_str(), line.size(), 0, 0,
+                    if (pcre_exec(rule, NULL, line.c_str(), line.size(),
-                            ovector, ovec_size) == 0)
+                                0, 0, ovector, ovec_size) >= 0)
                     {
                         string name(line, ovector[2], ovector[3] - ovector[2]);
                         string definition(line,
@@ -127,7 +150,10 @@ bool parse_input(char * buff, int size)
                     }
                     break;
             }
+        }
         input = newline + 1;
         lineno++;
     }
+
+    return true;
 }

testinput.i

@@ -1,13 +1,13 @@
 
 [tokens]
 
-ASSIGN     := ":="
+ASSIGN     :=
-DASSIGN    := ":=="
+DASSIGN    :==
-IDENTIFIER := "[a-zA-Z_][a-zA-Z_0-9]*"
+IDENTIFIER [a-zA-Z_][a-zA-Z_0-9]*
 
 [rules]
 
 Assignment := IDENTIFIER ASSIGN Expression
 
-Expression := IDENTIFIER
+Expression := IDENTIFIER \
             | Assignment