Refactor a little.

charset.c

@@ -557,33 +557,19 @@ is_utf8_well_formed(ss, slen)
 }
 
 /*
- * Return number of invalid UTF-8 sequences and binary chars found in a buffer.
+ * Skip bytes until a UTF-8 lead byte (11xxxxxx) or ASCII byte (0xxxxxxx) is found.
  */
- public int
-utf_bin_count(data, len)
- char *data;
- int len;
+ public void
+utf_skip_to_lead(pp, limit)
+ char **pp;
+ char *limit;
 {
- int bin_count = 0;
- char *edata = data + len;
- while (data < edata)
- {
- if (is_utf8_well_formed(data, edata-data))
- {
- if (bin_char_in_string(&data, edata))
- bin_count++;
- } else /* invalid UTF-8 */
- {
- /* Skip to next lead byte. */
- bin_count++;
- do {
- ++data;
- } while (data < edata && !IS_UTF8_LEAD(*data & 0377) && !IS_ASCII_OCTET(*data));
- }
- }
- return (bin_count);
+ do {
+ ++(*pp);
+ } while (*pp < limit && !IS_UTF8_LEAD((*pp)[0] & 0377) && !IS_ASCII_OCTET((*pp)[0]));
 }
 
+
 /*
  * Get the value of a UTF-8 character.
  */

filename.c

@@ -460,7 +460,7 @@ bin_file(f)
  int bin_count = 0;
  char data[256];
  char* p;
- char* pend;
+ char* edata;
 
  if (!seekable(f))
  return (0);
@@ -469,15 +469,19 @@ bin_file(f)
  n = read(f, data, sizeof(data));
  if (n <= 0)
  return (0);
- if (utf_mode)
+ edata = &data[n];
+ for (p = data; p < edata; )
  {
- bin_count = utf_bin_count(data, n);
- } else
- {
- pend = &data[n];
- for (p = data; p < pend; )
+ if (utf_mode && !is_utf8_well_formed(p, edata-data))
+  {
+  bin_count++;
+  utf_skip_to_lead(&p, edata);
+ } else 
  {
- if (bin_char_in_string(&p, pend))
+ LWCHAR c = step_char(&p, +1, edata);
+ if (ctldisp == OPT_ONPLUS && IS_CSI_START(c))
+ skip_ansi(&p, edata);
+ else if (binary_char(c))
  bin_count++;
  }
  }
@@ -488,26 +492,6 @@ bin_file(f)
  return (bin_count > 5);
 }
 
-/*
- * Determine if the next char in a string is binary.
- */
- public int
-bin_char_in_string(pp, limit)
- char **pp;
- constant char *limit;
-{
- LWCHAR c = step_char(pp, +1, limit);
- if (ctldisp == OPT_ONPLUS && IS_CSI_START(c))
- {
- /* Skip the CSI sequence. */
- do {
- c = step_char(pp, +1, limit);
- } while (*pp < limit && is_ansi_middle(c));
- } else if (binary_char(c))
- return (1);
- return (0);
-}
-
 /*
  * Try to determine the size of a file by seeking to the end.
  */

line.c

@@ -571,7 +571,7 @@ is_ansi_end(ch)
 }
 
 /*
- *
+ * Can a char appear in an ANSI escape sequence, before the end char?
  */
  public int
 is_ansi_middle(ch)
@@ -584,6 +584,23 @@ is_ansi_middle(ch)
  return (strchr(mid_ansi_chars, (char) ch) != NULL);
 }
 
+/*
+ * Skip past an ANSI escape sequence.
+ * pp is initially positioned just after the CSI_START char.
+ */
+ public void
+skip_ansi(pp, limit)
+ char **pp;
+ constant char *limit;
+{
+ LWCHAR c;
+ do {
+ c = step_char(pp, +1, limit);
+ } while (*pp < limit && is_ansi_middle(c));
+ /* Note that we discard final char, for which is_ansi_middle is false. */
+}
+
+
 /*
  * Append a character and attribute to the line buffer.
  */