lexer: Reimplement for better testability and internationalization.

[pspp-builds.git] / src / data / identifier.c

diff --git a/src/data/identifier.c b/src/data/identifier.c
index 1e149b8dc3ea9673c801be8ed7cb0010dbb546b5..f1c22ef1b567223579586ca8ac6f24fd3f4e3722 100644 (file)
--- a/src/data/identifier.c
+++ b/src/data/identifier.c
@@ -1,5 +1,5 @@
 /* PSPP - a program for statistical analysis.
-   Copyright (C) 1997-9, 2000, 2005, 2009, 2010 Free Software Foundation, Inc.
+   Copyright (C) 1997-9, 2000, 2005, 2009, 2010, 2011 Free Software Foundation, Inc.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -23,7 +23,6 @@
 
 #include "data/identifier.h"
 
-#include <assert.h>
 #include <string.h>
 #include <unictype.h>
 
@@ -31,6 +30,124 @@
 
 #include "gl/c-ctype.h"
 
+#include "gettext.h"
+#define _(msgid) gettext (msgid)
+
+/* Tokens. */
+
+/* Returns TYPE as a string, e.g. "ID" for T_ID. */
+const char *
+token_type_to_name (enum token_type type)
+{
+  switch (type)
+    {
+#define TOKEN_TYPE(TYPE) case T_##TYPE: return #TYPE;
+      TOKEN_TYPES
+#undef TOKEN_TYPE
+    case TOKEN_N_TYPES:
+    default:
+      return "unknown token type";
+    }
+}
+
+/* Returns an ASCII string that yields TOKEN if it appeared in a syntax file,
+   as a statically allocated constant string.  This function returns NULL for
+   tokens that don't have any fixed string representation, such as identifier
+   and number tokens. */
+const char *
+token_type_to_string (enum token_type token)
+{
+  switch (token)
+    {
+    case T_ID:
+    case T_POS_NUM:
+    case T_NEG_NUM:
+    case T_STRING:
+    case T_STOP:
+      return NULL;
+
+    case T_ENDCMD:
+      return ".";
+
+    case T_PLUS:
+      return "+";
+
+    case T_DASH:
+      return "-";
+
+    case T_ASTERISK:
+      return "*";
+
+    case T_SLASH:
+      return "/";
+
+    case T_EQUALS:
+      return "=";
+
+    case T_LPAREN:
+      return "(";
+
+    case T_RPAREN:
+      return ")";
+
+    case T_LBRACK:
+      return "[";
+
+    case T_RBRACK:
+      return "]";
+
+    case T_COMMA:
+      return ",";
+
+    case T_AND:
+      return "AND";
+
+    case T_OR:
+      return "OR";
+
+    case T_NOT:
+      return "NOT";
+
+    case T_EQ:
+      return "EQ";
+
+    case T_GE:
+      return ">=";
+
+    case T_GT:
+      return ">";
+
+    case T_LE:
+      return "<=";
+
+    case T_LT:
+      return "<";
+
+    case T_NE:
+      return "~=";
+
+    case T_ALL:
+      return "ALL";
+
+    case T_BY:
+      return "BY";
+
+    case T_TO:
+      return "TO";
+
+    case T_WITH:
+      return "WITH";
+
+    case T_EXP:
+      return "**";
+
+    case TOKEN_N_TYPES:
+      NOT_REACHED ();
+    }
+
+  NOT_REACHED ();
+}
+
 /* Recognizing identifiers. */
 
 static bool
@@ -197,20 +314,3 @@ lex_id_to_token (struct substring id)
 
   return T_ID;
 }
-
-/* Returns the name for the given keyword token type. */
-const char *
-lex_id_name (enum token_type token)
-{
-  const struct keyword *kw;
-
-  for (kw = keywords; kw < &keywords[keyword_cnt]; kw++)
-    if (kw->token == token)
-      {
-        /* A "struct substring" is not guaranteed to be
-           null-terminated, as our caller expects, but in this
-           case it always will be. */
-        return ss_data (kw->identifier);
-      }
-  NOT_REACHED ();
-}