Hmmm, very interesting. Although, in emacs, speakup still doesn't speak characters or lines as you move the application cursor. Even if you enable the experimental cursor peaking, it still does very poorly and repeats charactors twice in a row. It still is better here then in pico, because you here the charactor you just backspaced over. Since this doesn't seem to be an alltogether speakup problem, I guess I'll get into the 1.0 boat. Greg On Fri, Sep 28, 2001 at 06:56:45PM -0400, Charles Hallenbeck wrote: > Hi Greg > > On Fri, 28 Sep 2001, Gregory Nowak wrote: > > > Is there a particuar factor anyway that would make speakup speak > > cursor movements in one editing enviornment, and not another? > > I do not know the details of the source code well enough to say, but > evidently pico and nano handle screen output very differently, so that one > makes speakup say the char under the cursor when the cursor moves, and the > other does not. But you have to distinguish between cursor tracking and > speaking the current text as the cursor moves. Perhaps it has to do with > how an application chooses to show you the cursor. If it shows you the > cursor by brightening the char the cursor is on, perhaps by rewriting that > char to the screen, that should be easy to catch. > > The situation in DOS was relatively easy compared to Linux. DOS has system > calls that tell you where the cursor is and what char is there. What I did > was for every key press to record the current cursor position befpre > [assomg tje cjar tp tje suste,. tjem waiting an arbitrary time and > checking the position again. If it moved from its original position, then > I did one of three things: (1) if it moved only one position, I pronounce > the char at the new location; (2) if it moved horizontally more than one > position, I read enough chars from the screen to isolate a 'word' and > pronounce that word; and (3) if it moved vertically, I read the line to > which it has moved. That was a good plan, but it meant capturing the key > stroke before the system gets it, as well as capturing data sent to the > screen, and coordinating the two. > > I think there is room for improvement in the way speakup speaks the cursor > movements, since at present the way it does it varies from one application > to another, and it shouldn't do that. I am not an emacs user like Kirk is, > but I am told that emacs behaves even more friendly than nano does in that > respect. Hey Kirk, are you reading this? How about switching to pico for a > day or two so you can see how a weak application works? <smile> > > Chuck > > > > Visit me at http://www.mhonline.net/~chuckh > "They that can give up essential liberty to obtain a little temporary > safety deserve neither liberty nor safety." -- Benjamin Franklin, 1759. > > > _______________________________________________ > Speakup mailing list > Speakup at braille.uwo.ca > http://speech.braille.uwo.ca/mailman/listinfo/speakup
