From 68a4bf6240a1ddf7b75f8b61d1f000be5d0102b1 Mon Sep 17 00:00:00 2001
From: Josh Holtrop <jholtrop+git@gmail.com>
Date: Thu, 1 Apr 2010 22:33:30 -0400
Subject: [PATCH] added unistring::operator+=(), working on parser more

---
 parse-input.cc | 68 +++++++++++++++++++++++++++++++++++++++++++++++++-
 unicode.cc     |  6 +++++
 unicode.h      |  1 +
 3 files changed, 74 insertions(+), 1 deletion(-)

diff --git a/parse-input.cc b/parse-input.cc
index 1a449aa..3062cd0 100644
--- a/parse-input.cc
+++ b/parse-input.cc
@@ -14,7 +14,8 @@ using namespace std;
 
 void parse_input(refptr< vector<unichar_t> > ucs)
 {
-    enum State { INITIAL, SECTION_NAME, RULES };
+    enum State { INITIAL, SECTION_NAME, RULES, RULE_NAME, RULE_COLON,
+        RULE_EQUALS, RULE_RHS };
     State state = INITIAL;
     int lineno = 1;
     int colno = 1;
@@ -69,6 +70,71 @@ void parse_input(refptr< vector<unichar_t> > ucs)
                         break;
                 }
                 break;
+            case RULES:
+                if (isspace(c))
+                {
+                }
+                else if (  ('a' <= c && c <= 'z')
+                        || ('A' <= c && c <= 'Z')
+                        || (c == '_') )
+                {
+                    build_str = "";
+                    build_str += c;
+                    state = RULE_NAME;
+                }
+                else
+                {
+                    SET_ERROR("Unexpected character");
+                }
+                break;
+            case RULE_NAME:
+                if (   ('a' <= c && c <= 'z')
+                    || ('A' <= c && c <= 'Z')
+                    || ('0' <= c && c <= '9')
+                    || (c == '_') )
+                {
+                    build_str += c;
+                }
+                else if (isspace(c))
+                {
+                    state = RULE_COLON;
+                }
+                else
+                {
+                    SET_ERROR("Expected ':='");
+                }
+                break;
+            case RULE_COLON:
+                if (isspace(c))
+                {
+                }
+                else if (c == ':')
+                {
+                    state = RULE_EQUALS;
+                }
+                else
+                {
+                    SET_ERROR("Expected ':='");
+                }
+                break;
+            case RULE_EQUALS:
+                if (c == '=')
+                {
+                    state = RULE_RHS;
+                }
+                else
+                {
+                    SET_ERROR("Expected '='");
+                }
+                break;
+            case RULE_RHS:
+                break;
+        }
+
+        if (error)
+        {
+            cerr << errstr << endl;
+            break;
         }
     }
 }
diff --git a/unicode.cc b/unicode.cc
index bff58b5..480cdd5 100644
--- a/unicode.cc
+++ b/unicode.cc
@@ -13,6 +13,12 @@ unistring & unistring::operator=(const char * ascii_str)
     return *this;
 }
 
+unistring & unistring::operator+=(const unichar_t c)
+{
+    chars.push_back(c);
+    return *this;
+}
+
 bool unistring::operator==(const char * ascii_str)
 {
     int len = chars.size();
diff --git a/unicode.h b/unicode.h
index e161819..57acdf2 100644
--- a/unicode.h
+++ b/unicode.h
@@ -10,6 +10,7 @@ class unistring
 {
     public:
         unistring & operator=(const char * ascii_str);
+        unistring & operator+=(const unichar_t c);
         bool operator==(const char * ascii_str);
 
     protected: