Merge 'master' into 'psppsheet'.

[pspp] / src / libpspp / str.c

diff --git a/src/libpspp/str.c b/src/libpspp/str.c
index 7e722c17e6b92602798725e3bdd618327a46691f..44e4a1da28ee7183a98eb1e819a755bc55759c0f 100644 (file)
--- a/src/libpspp/str.c
+++ b/src/libpspp/str.c
@@ -1,5 +1,5 @@
 /* PSPP - a program for statistical analysis.
-   Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011 Free Software Foundation, Inc.
+   Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2012 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
@@ -491,11 +491,15 @@ bool
 ss_tokenize (struct substring ss, struct substring delimiters,
              size_t *save_idx, struct substring *token)
 {
+  bool found_token;
+
   ss_advance (&ss, *save_idx);
   *save_idx += ss_ltrim (&ss, delimiters);
   ss_get_bytes (&ss, ss_cspan (ss, delimiters), token);
-  *save_idx += ss_length (*token) + 1;
-  return ss_length (*token) > 0;
+
+  found_token = ss_length (*token) > 0;
+  *save_idx += ss_length (*token) + found_token;
+  return found_token;
 }
 
 /* Removes the first CNT bytes from SS. */
@@ -1544,6 +1548,13 @@ ds_put_byte_multiple (struct string *st, int ch, size_t cnt)
   memset (ds_put_uninit (st, cnt), ch, cnt);
 }
 
+/* Appends Unicode code point UC to ST in UTF-8 encoding. */
+void
+ds_put_unichar (struct string *st, ucs4_t uc)
+{
+  ds_extend (st, ds_length (st) + 6);
+  st->ss.length += u8_uctomb (CHAR_CAST (uint8_t *, ds_end (st)), uc, 6);
+}
 
 /* If relocation has been enabled, replace ST,
    with its relocated version */