Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!cernvax!chx400!ugun2b!ugsc2a!fisher From: fisher@sc2a.unige.ch Newsgroups: comp.binaries.ibm.pc.d Subject: Re: ALED ... does *not* mask the 8-th bit! Message-ID: <1991Jun18.104523.484@sc2a.unige.ch> Date: 18 Jun 91 08:45:23 GMT References: <3245@krafla.rhi.hi.is> <4089@sixhub.UUCP> <3258@krafla.rhi.hi.is> Organization: University of Geneva, Switzerland Lines: 34 In article <3258@krafla.rhi.hi.is>, einari@rhi.hi.is (Einar Indridason) writes: > I don't know how that ALED reads the keyboard, but it seem to me to be > something like this: > int 16h > cmp al,7fh > ja error ; Nobody uses the 8th bit, so why should I??? > .... > This is clearly the Wrong way of reading the keyboard. This is not correct. Using the swiss-french keyboard, I can enter without problem chars such as ('e), (`a), (,c) or (`e). This means that the 8-th bit is *not* masked. On the other hand, keys such as ("u) or ("a) return an error message "Macro not defined", the ("o) gets me an ascii table (this I find really funny) and (~n) displays the time in the top right corner... Even the non-US AltGr key works on some keys (@), (#), etc. The AltGr key is often confused by US software with a regular Alt... I suspect the main problem with ALED is the macro feature. I also suspect that this behavior is due to a bug with the NULL character. Any normal key returns a character between 32 and 254. Control keys and special keys like Tab return characters between 1 and 31. Shift keys return nothing, only an internal shift-state table is updated. All others return a NULL character followed by a unique code. It seems that the input routine of ALED sometimes confuses "normal" characters with an "extended" code. Another bug would be to mask the 8-th bit when checking for control characters, but this seems unlikely. BTW: the display of an ASCII table or of a clock are both mapped to Alt-keys, which makes the first therory more likely. Thus: ALED does *not* mask the 8-th bit! It *has* a serious bug involving extended codes and possibly the Alt-key, making the program useless for non-english users. Markus G. Fischer fisher@sc2a.unige.ch