Re: [PATCH v2] fetch: align new ref summary printout in UTF-8 locales

On Thu, Sep 6, 2012 at 2:20 AM, Torsten Bögershausen [off-list ref] wrote:

On 09/05/2012 08:15 PM, Torsten Bögershausen wrote:

quoted

On 04.09.12 12:39, Nguyễn Thái Ngọc Duy wrote:

quoted

+/* return the number of columns of string 's' in current locale */
+int gettext_width(const char *s)
+{
+       static int is_utf8 = -1;
+       if (is_utf8 == -1)
+               is_utf8 = !strcmp(charset, "UTF-8");
+
+       return is_utf8 ? utf8_strwidth(s) : strlen(s);

Will that work for non-ASCII encodings?
For ISO-8859-x we can say strlen() == strwidth(),
but for other encodings using multibytes that doesn't work, does it?

No it does not. I think I mentioned that in the first version that I
was only interested in utf-8. Others can extend the function for their
favourite encodings.

(Sorry the message went out before completely written)
Something like that:

int gettext_width(const char *s) {

  static int is_utf8 = -1;

  if (is_utf8 == -1)

    is_utf8 = !strcmp(charset, "UTF-8");

  if (is_utf8)
    return utf8_strwidth(s);
  else  {
    char *s_utf = reencode_string(s, "UTF-8", charset);
    if (s_utf) {
      witdh = utf8_strwidth(s_utf);
      free(s_utf);
    } else
      width = strlen(s);

    return width;
}

Yes, something like that, assuming that column information is intact
after the conversion. Maybe you can make that a new function, int
strwidth(const char *str, const char *charset), and make
gettext_strwidth() a thin wrapper:

int gettext_strwidth(const char *s)
{
   return strwidth(s, charset);
}
-- 
Duy