lexer: Reimplement for better testability and internationalization.

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

diff --git a/src/data/identifier.h b/src/data/identifier.h
index 352ef2eafc1b3b381a82d7ee55e96b4dc344623e..7f2f904239167f1c5c4200a570621dd7d83cf66a 100644 (file)
--- a/src/data/identifier.h
+++ b/src/data/identifier.h
@@ -1,5 +1,5 @@
 /* PSPP - a program for statistical analysis.
-   Copyright (C) 1997-9, 2000 Free Software Foundation, Inc.
+   Copyright (C) 1997-9, 2000, 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
@@ -19,43 +19,73 @@
 
 #include <ctype.h>
 #include <stdbool.h>
-#include <sys/types.h>
-#include <libpspp/str.h>
+#include <unitypes.h>
+#include "libpspp/str.h"
+
+#define TOKEN_TYPES                                                     \
+    TOKEN_TYPE(ID)                         /* Identifier. */            \
+    TOKEN_TYPE(POS_NUM)                    /* Positive number. */       \
+    TOKEN_TYPE(NEG_NUM)                    /* Negative number. */       \
+    TOKEN_TYPE(STRING)                     /* Quoted string. */         \
+    TOKEN_TYPE(STOP)                       /* End of input. */          \
+                                                                        \
+    TOKEN_TYPE(ENDCMD)                     /* . */                      \
+    TOKEN_TYPE(PLUS)                       /* + */                      \
+    TOKEN_TYPE(DASH)                       /* - */                      \
+    TOKEN_TYPE(ASTERISK)                   /* * */                      \
+    TOKEN_TYPE(SLASH)                      /* / */                      \
+    TOKEN_TYPE(EQUALS)                     /* = */                      \
+    TOKEN_TYPE(LPAREN)                     /* ( */                      \
+    TOKEN_TYPE(RPAREN)                     /* ) */                      \
+    TOKEN_TYPE(LBRACK)                     /* [ */                      \
+    TOKEN_TYPE(RBRACK)                     /* ] */                      \
+    TOKEN_TYPE(COMMA)                      /* , */                      \
+                                                                        \
+    TOKEN_TYPE(AND)                        /* AND */                    \
+    TOKEN_TYPE(OR)                         /* OR */                     \
+    TOKEN_TYPE(NOT)                        /* NOT */                    \
+                                                                        \
+    TOKEN_TYPE(EQ)                         /* EQ */                     \
+    TOKEN_TYPE(GE)                         /* GE or >= */               \
+    TOKEN_TYPE(GT)                         /* GT or > */                \
+    TOKEN_TYPE(LE)                         /* LE or <= */               \
+    TOKEN_TYPE(LT)                         /* LT or < */                \
+    TOKEN_TYPE(NE)                         /* NE or ~= */               \
+                                                                        \
+    TOKEN_TYPE(ALL)                        /* ALL */                    \
+    TOKEN_TYPE(BY)                         /* BY */                     \
+    TOKEN_TYPE(TO)                         /* TO */                     \
+    TOKEN_TYPE(WITH)                       /* WITH */                   \
+                                                                        \
+    TOKEN_TYPE(EXP)                        /* ** */
 
 /* Token types. */
-enum
+enum token_type
   {
-    T_ID = 256, /* Identifier. */
-    T_POS_NUM, /* Positive number. */
-    T_NEG_NUM, /* Negative number. */
-    T_STRING,  /* Quoted string. */
-    T_STOP,    /* End of input. */
+#define TOKEN_TYPE(TYPE) T_##TYPE,
+    TOKEN_TYPES
+    TOKEN_N_TYPES
+#undef TOKEN_TYPE
+  };
 
-    T_AND,     /* AND */
-    T_OR,      /* OR */
-    T_NOT,     /* NOT */
+const char *token_type_to_name (enum token_type);
+const char *token_type_to_string (enum token_type);
 
-    T_EQ,      /* EQ */
-    T_GE,      /* GE or >= */
-    T_GT,      /* GT or > */
-    T_LE,      /* LE or <= */
-    T_LT,      /* LT or < */
-    T_NE,      /* NE or ~= */
+/* Tokens. */
+bool lex_is_keyword (enum token_type);
 
-    T_ALL,     /* ALL */
-    T_BY,      /* BY */
-    T_TO,      /* TO */
-    T_WITH,    /* WITH */
+/* Validating identifiers. */
+#define ID_MAX_LEN 64          /* Maximum length of identifier, in bytes. */
 
-    T_EXP,     /* ** */
-  };
-
-/* Tokens. */
-bool lex_is_keyword (int token);
+bool id_is_valid (const char *id, const char *dict_encoding, bool issue_error);
+bool id_is_plausible (const char *id, bool issue_error);
 
 /* Recognizing identifiers. */
 bool lex_is_id1 (char);
 bool lex_is_idn (char);
+bool lex_uc_is_id1 (ucs4_t);
+bool lex_uc_is_idn (ucs4_t);
+bool lex_uc_is_space (ucs4_t);
 size_t lex_id_get_length (struct substring);
 
 /* Comparing identifiers. */
@@ -64,7 +94,4 @@ bool lex_id_match_n (struct substring keyword, struct substring token,
                      size_t n);
 int lex_id_to_token (struct substring);
 
-/* Identifier names. */
-const char *lex_id_name (int);
-
 #endif /* !data/identifier.h */