[Vm-dev] [commit][3255] squeak -help says that the default encoding for external text is UTF-8, make it so.

[Eliot Miranda]

Strange but true ;-).  The reason is that strcmp answers a - b fir the first difference in the string.  In which case it should have been called strdiff.  Ugh...  

Eliot (phone)

On Feb 13, 2015, at 5:13 AM, Wolfgang Eder <edw at generalmagic.at> wrote:

> strcmp answers zero if the strings are the same.
> zero, interpreted as a boolean, is false.
> so, !strcmp() means that the strings are equal.
> kind regards wolfgang
> 
>> Am 13.02.2015 um 14:02 schrieb Henrik Johansen <henrik.s.johansen at veloxit.no>:
>> 
>> 
>> 
>>> On 13 Feb 2015, at 4:00 , commits at squeakvm.org wrote:
>>> 
>>> 
>>> Revision: 3255
>>> Author:   lewis
>>> Date:     2015-02-12 19:00:34 -0800 (Thu, 12 Feb 2015)
>>> Log Message:
>>> -----------
>>> squeak -help says that the default encoding for external text is UTF-8, make it so.
>>> 
>>> Modified Paths:
>>> --------------
>>>  trunk/platforms/unix/vm/sqUnixMain.c
>>> 
>>> Modified: trunk/platforms/unix/vm/sqUnixMain.c
>>> ===================================================================
>>> --- trunk/platforms/unix/vm/sqUnixMain.c    2015-02-12 19:20:06 UTC (rev 3254)
>>> +++ trunk/platforms/unix/vm/sqUnixMain.c    2015-02-13 03:00:34 UTC (rev 3255)
>>> @@ -99,7 +99,7 @@
>>>      int    uxDropFileCount=    0;    /* number of dropped items    */
>>>      char **uxDropFileNames=    0;    /* dropped filenames        */
>>> 
>>> -       int    textEncodingUTF8= 0;    /* 1 if copy from external selection uses UTF8 */
>>> +       int    textEncodingUTF8= 1;    /* 1 if copy from external selection uses UTF8 */
>>> 
>>> #if defined(IMAGE_DUMP)
>>> static int    dumpImageFile=    0;    /* 1 after SIGHUP received */
>>> @@ -1082,10 +1082,11 @@
>>>      len= strlen(buf);
>>>      for (i= 0;  i < len;  ++i)
>>>        buf[i]= toupper(buf[i]);
>>> -      if ((!strcmp(buf, "UTF8")) || (!strcmp(buf, "UTF-8")))
>>> -        textEncodingUTF8= 1;
>>> -      else
>>> -        setEncoding(&uxTextEncoding, buf);
>>> +      if (strcmp(buf, "UTF8") && strcmp(buf, "UTF-8"))
>>> +        {
>>> +          textEncodingUTF8= 0;
>>> +          setEncoding(&uxTextEncoding, buf);
>>> +        }
>>>      free(buf);
>>>      return 2;
>>>    }
>> 
>> Wait, wat?
>> They way I read that, it says, "If buf is both "UTF8" and "UTF-8", set the text encoding to not be utf8"?
>> I don't get how the initial version could work either (the if/else blocks look reversed, for my interpretation "if buf is neither UTF8 nor UTF-8"... ), but does anyone fluent in C have time to explain to me how the new version works / how I'm reading wrong?
>> 
>> Cheers,
>> Henry
>