Entering unicode or extended ascii characters

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   Hi Pete,

Could you answer some basic questions on entering characters in either non-standard codepages, or in Unicode?

Essentially, I would like to enter some characters with macrons - that is a horizontal bar over the letter.  I need these on the vowel characters. Windows 1257 includes these glyphs (as does ISO8859-4).

In character mode, if I set the font to Deja Vu Sans Mono, and the character set to Windows - Baltic (Windows 1257), I can print all the required characters (e.g. CHAR(194) or CHAR(226)). However, if I try to ENTER the characters using Alt-194 or Alt-226, then AccuTerm displays ^?. The received data is actually correct, because if I print it out again, it is the correct character.

In GUI mode, I am using the Verdana font. This is Unicode aware. But if I try to enter the Unicode values (Alt-0256 or Alt-0257), nothing appears at all. If I try Alt-keys in the normal range, then the character set being displayed appears to be the OEM set. (I haven't tried displaying the characters via GUI because it is unclear whether I should be sending Unicode or some (which) code page).

I have tried numerous combinations of font and character sets, and have yet to find something that works in a way that a user will recognise. [Curiously, Verdana displays the characters I want in Word using the above Alt-codes, but Excel does not. Not sure if that is relevant, but it may help.]

Do you have any suggestions of how to actually enter these characters ... AND have them displayed as they entered (rather than inputting ^? and outputting as the proper glyph).

AT 7.1b sp2 (7.1.2012) on UniVerse 11.1.1/Windows

Thanks,

Brian

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
Shrek59 Members Profile Find Members Posts Senior Member Joined: December 04 2006 Location: New Zealand Status: Offline Points: 208	Post Options Post Reply Quote Shrek59 Report Post Thanks(0) Quote Reply Topic: Entering unicode or extended ascii characters Posted: September 28 2014 at 3:30pm
	Hi Pete, Could you answer some basic questions on entering characters in either non-standard codepages, or in Unicode? Essentially, I would like to enter some characters with macrons - that is a horizontal bar over the letter. I need these on the vowel characters. Windows 1257 includes these glyphs (as does ISO8859-4). In character mode, if I set the font to Deja Vu Sans Mono, and the character set to Windows - Baltic (Windows 1257), I can print all the required characters (e.g. CHAR(194) or CHAR(226)). However, if I try to ENTER the characters using Alt-194 or Alt-226, then AccuTerm displays ^?. The received data is actually correct, because if I print it out again, it is the correct character. In GUI mode, I am using the Verdana font. This is Unicode aware. But if I try to enter the Unicode values (Alt-0256 or Alt-0257), nothing appears at all. If I try Alt-keys in the normal range, then the character set being displayed appears to be the OEM set. (I haven't tried displaying the characters via GUI because it is unclear whether I should be sending Unicode or some (which) code page). I have tried numerous combinations of font and character sets, and have yet to find something that works in a way that a user will recognise. [Curiously, Verdana displays the characters I want in Word using the above Alt-codes, but Excel does not. Not sure if that is relevant, but it may help.] Do you have any suggestions of how to actually enter these characters ... AND have them displayed as they entered (rather than inputting ^? and outputting as the proper glyph). AT 7.1b sp2 (7.1.2012) on UniVerse 11.1.1/Windows Thanks, Brian

wwf_admin Members Profile Find Members Posts Admin Group Joined: October 01 2003 Location: United States Status: Offline Points: 22	Post Options Post Reply Quote wwf_admin Report Post Thanks(0) Quote Reply Posted: September 29 2014 at 4:02am
	Hi Brian - Internationalization is a tricky subject. Since you are using Universe, do you have NLS enabled? If so, you should be able to use UTF-8 with AccuTerm. I suspect that you do not have NLS enabled since the echo for Alt+194 is ^?. On my D3 machine, with xcs-on in effect, I can enter Alt+194 from the keyboad and the appropriate character (depends on codepage) is displayed. AccuTerm (text screen) does not support direct keyboard entry of arbitrary Unicode characters. You can paste from the clipboard - use the Character Map app to copy to the clipboard. You mention that you are using codepage 1257 in AccuTerm. Is your Windows machine configured to use codepage 1257 as well? That is important if you want to use the 8-bit character set with a GUI app. Again, the clipboard & Character Map work with GUI apps, but direct keyboard entry of Unicode characters using Alt + "+" + hex (in conjunction with the EnableHexNumpad registry key) does not appear to work. This may be a bug (the text screen is not a bug, just a limitation, as it conflicts with other keyboard functions). As far as I know, Windows does not support entering any non-8-bit characters using ALT+0xxx. Thanks, Pete

Shrek59 Members Profile Find Members Posts Senior Member Joined: December 04 2006 Location: New Zealand Status: Offline Points: 208	Post Options Post Reply Quote Shrek59 Report Post Thanks(0) Quote Reply Posted: September 29 2014 at 8:55am
	Thanks, Pete. At this stage, I'm just dipping my toe in the water ... and going to full NLS support is more like jumping in the deep end! Using the character map in GUI mode, I could paste the required characters into a text box - but they appeared to be stripped out before going into guiargs. Of course, this could be the code page issue you were referring to. Chcp reports the server is set to code page 850 - and those characters don't exist in that code page. Overall, this is probably going to sit in the "too hard" basket for a while longer. Cheers, Brian

PSchellenbach Members Profile Find Members Posts Admin Group Moderator Joined: December 15 2003 Location: United States Status: Offline Points: 2150	Post Options Post Reply Quote PSchellenbach Report Post Thanks(0) Quote Reply Posted: September 29 2014 at 9:17am
	Hi Brian - The correct way to handle this is to jump in the deep end and turn on NLS and use UTF-8 in AccuTerm. I have not tried this with UniVerse, but I know it works fine in QM. If/when you take this route, if you encounter any issues with Universe NLS, let me know. FYI: I think chcp is only for the DOS command window, not for Windows apps. Thanks, Pete