X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fdata%2Fidentifier.c;h=db20010464cab1e3f3a1f42cc1e7a4c012bf8c1b;hb=3b6384c0f3f35fb0b280cb20ddaa01b50912f4b8;hp=4b613bb480edb5555cb1532176d745182345c0c4;hpb=1b3322acf30d531cefe3cdbf7287ec8cde601bcd;p=pspp

diff --git a/src/data/identifier.c b/src/data/identifier.c
index 4b613bb480..db20010464 100644
--- 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, 2012, 2013 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
@@ -23,15 +23,12 @@
 
 #include "data/identifier.h"
 
-#include <assert.h>
 #include <string.h>
-#include <unictype.h>
 #include <unistr.h>
+#include <unictype.h>
 
 #include "libpspp/assertion.h"
 #include "libpspp/cast.h"
-#include "libpspp/i18n.h"
-#include "libpspp/message.h"
 
 #include "gl/c-ctype.h"
 
@@ -193,7 +190,13 @@ lex_is_idn (char c)
 bool
 lex_uc_is_id1 (ucs4_t uc)
 {
-  return is_ascii_id1 (uc) || (uc >= 0x80 && uc_is_property_id_start (uc));
+  return (uc < 0x80
+          ? is_ascii_id1 (uc)
+          : (uc_is_general_category_withtable (uc,
+                                               UC_CATEGORY_MASK_L |
+                                               UC_CATEGORY_MASK_M |
+                                               UC_CATEGORY_MASK_S)
+             && uc != 0xfffc && uc != 0xfffd));
 }
 
 /* Returns true if Unicode code point UC may be a character in an identifier
@@ -201,8 +204,14 @@ lex_uc_is_id1 (ucs4_t uc)
 bool
 lex_uc_is_idn (ucs4_t uc)
 {
-  return (is_ascii_id1 (uc) || isdigit (uc) || uc == '.' || uc == '_'
-          || (uc >= 0x80 && uc_is_property_id_continue (uc)));
+  return (uc < 0x80
+          ? is_ascii_id1 (uc) || isdigit (uc) || uc == '.' || uc == '_'
+          : (uc_is_general_category_withtable (uc,
+                                               UC_CATEGORY_MASK_L |
+                                               UC_CATEGORY_MASK_M |
+                                               UC_CATEGORY_MASK_S |
+                                               UC_CATEGORY_MASK_N)
+             && uc != 0xfffc && uc != 0xfffd));
 }
 
 /* Returns true if Unicode code point UC is a space that separates tokens. */
@@ -225,15 +234,21 @@ lex_uc_is_space (ucs4_t uc)
 size_t
 lex_id_get_length (struct substring string)
 {
-  size_t length = 0;
-  if (!ss_is_empty (string) && lex_is_id1 (ss_first (string)))
+  const uint8_t *s = CHAR_CAST (const uint8_t *, string.string);
+  size_t len = string.length;
+  size_t ofs;
+  int mblen;
+
+  for (ofs = 0; ofs < string.length; ofs += mblen)
     {
-      length = 1;
-      while (length < ss_length (string)
-             && lex_is_idn (ss_at (string, length)))
-        length++;
+      ucs4_t uc;
+
+      mblen = u8_mbtouc (&uc, s + ofs, len - ofs);
+      if (!(ofs == 0 ? lex_uc_is_id1 (uc) : lex_uc_is_idn (uc)))
+        break;
     }
-  return length;
+
+  return ofs;
 }
 
 /* Comparing identifiers. */
@@ -319,20 +334,3 @@ 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 ();
-}