<div dir="ltr"><div>openCharEncodubgConvert should inside file charEncoding.c (in module modul localization). I am currently not before my own desk. To see if I need to fix it few days later.</div>
<div> </div>
<div>May be you don't remove the other charEncoding.c,charEncoding.h (in modules windows_tool )</div>
<div> </div>
<div>YungLee<br><br></div>
<div class="gmail_quote">2008/8/17 Sylvestre Ledru <span dir="ltr"><<a href="mailto:sylvestre.ledru@scilab.org">sylvestre.ledru@scilab.org</a>></span><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Hello YungLee,<br><br>Seems a nice and useful work! I am trying to test it.<br>However, it seems that a file is missing in your diff.<br>
For example, I don't have the function "openCharEncodingConverter"<br>(maybe you forgot the "svn add" on this file).<br><br>Thanks again,<br>Sylvestre<br><br>Le samedi 16 août 2008 à 02:03 +0800, Yung-Jang Lee a écrit :<br>
<div>
<div></div>
<div class="Wj3C7c">> Sylvestre :<br>><br>> I have finished the console display and input/outfile encoding<br>> fix. Also included is my zh_TW translation.<br>><br>> In this version, Scilab5 is internal UTF encoding both in Java and<br>
> C/C++ contex. This way Java and TCL interface need only handle UTF<br>> string.<br>><br>> System locale is used in 1. Dos/Shell terminal input/output 2.<br>> External file read/write.<br>><br>> In current implementation, C-style output function fail to<br>
> handle UTF characters is also fixed in this patch.<br>><br>> I have test my patch with the attached script. My patch still fail<br>> to past the Fortran-Style output test, because this need old-fashion<br>
> fortran programming skill and I am not qualified about it. But the<br>> C-style output encoding is fixed.<br>><br>> Input encoding for command such as 'exec"scriptFile"' is changed<br>
> to system locale, to consistent with output encoding.<br>><br>> I have a business trip from next monday for a week. Just leave<br>> comments or suggestions to my email, I will response to this topic<br>> when I back to my desk.<br>
><br>><br>> WBR,<br>><br>> YungLee<br>><br>><br>><br>><br>> ------------------------Test<br>> Script-------------------------------------<br>> mode(-1);<br>> text='中文測試'; // test string for Windows CP950 (zh_TW) and Linux<br>
> (zh_TW.BIG5)<br>> //<br>> //test 1. terminal string output error, display ok for Chinese, fail<br>> for Russian<br>> //<br>> // reason : an UTF sting convert to UTF once again<br>> //<br>> disp(text) //<br>
> mprintf(text)<br>> //test 2. about() command fail to display Chinese/Russian<br>> //<br>> // reason :gettext fununction should return string in UTF rather than<br>> // system locale, this influence about() command<br>
> //<br>> x_message(gettext('&File')) // gettext inside x_message<br>> about();<br>> //<br>> // Bug 3. string output in file, must encoded in system locale<br>> // in current SVN in the output string is in UTF encoding.<br>
> //<br>> // Fortran-Style<br>> output<---------------------------------------------------------------------- not easy to fix,<br>> u=file('open','resultsA','unknown') //open the result file<br>
> write(u,text,'(a)') // encoding wrong, not in system locale<br>> file('close',u)<br>> // C style output<br>> u=mopen('resultsB','w') //open the result file<br>> mfprintf(u,text+" plus %s\n",text) // file encoding wrong, not in<br>
> system locale<br>> mclose(u);<br>> //<br>> // Bug5. formatted output to a string<br>> //<br>> printf(text+" with aonther %s\n",text) // output encoding wrong, not<br>> in system locale<br>
> //<br>> msprintf(text+" plus %s\n",text) // error in terminal mode,ok for java<br>> console<br>> //<br>> // Text Drawing test : test to see any side effect for text drawing<br>> //<br>> title('title test'+text);<br>
> t=0:0.1:2*%pi;<br>> plot2d(t,[sin(t'),cos(t')],[-1,2]);<br>> legends(['legends test'+text ;'text B '+text],[-1,2], opt=3 )<br>> ---------------------------------------------------------<br>
><br><br></div></div></blockquote></div><br></div>