Hi Isaac, glad you are now able to get it working. Interested to know what you found 'strange' about the railwaycat version. I didn't notice much along these lines and found overall it worked far better. Note that one of the things I do like about it is that you can set an option to turn off the voiceover screen reader while running in a terminal, which is really useful as it means you don't get the conflicts you mention and you don't have to turn off voiceOver. Tim Isaac Leonard via Emacspeak <emacspeak(a)emacspeak.org> writes: > Hi Tim and Victor > I went with the second version and it seems to be working fine now thank > you. > I tried out the railwaycat version however it was a bit strange in both > GUI mode and terminal mode. > Reverting back to the normal version and making the edits to emacspeak > seem to work well for now. > Thank you for the help. > BTW I use the GUI version launched from the terminal as it seems to conflict with voiceover and tdsr > (terminal screen reader) the least while not disabling them. > Tim Cross via Emacspeak <emacspeak(a)emacspeak.org> writes: > >> Good point Victor - I forgot about that issue. I was never able to >> reproduce it using the railwaycat version. >> >> One additional step to take if Isaac decides to modify the Emacspeak >> sources is to ensure you do a make clean, make config and make. While >> just running make will probably work, the make clean; make config and >> make workflow is much more reliable and a good habit to develop. >> >> Victor Tsaran via Emacspeak <emacspeak(a)emacspeak.org> writes: >> >>> Hi Isaac! >>> We have discussed this issue on this list a few months ago >>> There are two possible ways to get around this crash. >>> 1. Install the RailwayCat version of Emacs. You will need to add another tap to your homebrew installation. Just search for >>> RailwayCat Emacs and the installation instructions will be there. >>> 2. Modify the lisp/emacspeak.el file and replace every instance of >>> make-thread >>> with >>> funcall >>> Then, you will have to "make emacspeak" again to recompile the modified file. >>> >>> I think Raman explained in his original message that this switches Emacspeak from multi-threaded into single-threaded mode >>> (at least this is the way I understood it then). >>> >>> If you are running Emacspeak from source, you will have to remember to "git stash" and then "git stash pop" these changes >>> during the the "pull" process. This is, of course, assuming that emacspeak.el does not get modified in the repository. >>> >>> I myself tried both approaches but ended up with step 2 in the end. >>> >>> Hope this helps. >>> Victor >>> >>> On Tue, Aug 31, 2021 at 9:24 PM Isaac Leonard via Emacspeak <emacspeak(a)emacspeak.org> wrote: >>> >>> Hi Tim >>> I'm studying at UON (University of Newcastle) in NSW. >>> I'm running emacs >>> 27.2 installed from homebrew. >>> I installed emacspeak from source with the git repo. >>> I'm running MacOS Mojave 10.14.6 on a macbook pro 2017 model. >>> I'd think it was emacs itself too however reverting emacspeak to >>> version 52.0 fixes the issue. >>> I've had no other memery issues in the past with emacs >>> and none with any other apps so I don't think its a hardware problem. >>> I load emacspeak in my init.el with: >>> (load-file "~/emacspeak/lisp/emacspeak-setup.el") >>> which is the way specified in the docs. >>> Tim Cross via Emacspeak <emacspeak(a)emacspeak.org> writes: >>> >>> > Hi Isaac, >>> > >>> > welcome to the list. >>> > >>> > What version of Emacs and how did you install it on the mac? >>> > >>> > What version of macOS are you running? >>> > >>> > That error would indicate a problem in Emacs rather than Emacspeak. >>> > There are various different ways of installing Emacs on the mac. I've >>> > found the best results are with the railwaycat mac-port homebrew recipe >>> > of Emacs. The mac-port version is a modified version of Emacs >>> > specifically built for the mac. One of its nice features is that it has >>> > a variable you can set to turn off voiceOver when running Emacs inside a >>> > terminal. I would also ensure your using Emacs 27 and not current >>> > development version 28. >>> > >>> > It does look like a memory allocation problem. My biggest concern is >>> > that this type of error is often associated with a hardware issue i.e. >>> > bad membory. That may not be the case, but is a possibility. >>> > >>> > BTW which University are you studying at? >>> > >>> > Tim >>> > >>> > Isaac Leonard via Emacspeak <emacspeak(a)emacspeak.org> writes: >>> > >>> >> Hi all, I am new to the mailing list and to mailing lists in general. >>> >> I saw that I should introduce myself so here goes >>> >> I am a 21 year old programmer from australia, I am totally blind and >>> >> have used emacspeak for around 2 years now. >>> >> I am currently studying computer science at university and mainly use >>> >> emacs for all of my writing when possible. >>> >> >>> >> I've recently had an issue with emacspeak crashing when I start emacs. >>> >> A few months ago when version 53 was released I tried to upgrade however >>> >> it caused emacs to crash on startup around 50% of the time. >>> >> I reverted to version 52.0 until the other day when I decided to try >>> >> upgrade again to version 54.0 and just deal with the crashes. >>> >> Yesterday however I attempted to open emacs 4 times in a row with >>> >> it crashing each time so decided to follow it up further. >>> >> This only happens with emacspeak53.0 or 54.0, previous versions have >>> >> never crashed before. >>> >> Here is the errors printed to the terminal from when it crashed >>> >> those 4 times : >>> >> $emacs >>> >> Emacs-x86_64-10_14(2360,0x700009d2e000) malloc: *** error for object >>> >> 0x70000a5307e8: pointer being f >>> >> reed was not allocated >>> >> Fatal error 11: Segmentation fault >>> >> Abort trap: 6 >>> >> $emacs >>> >> Emacs-x86_64-10_14(2433,0x700003241000) malloc: *** error for object >>> >> 0x7ffee6db5200: pointer being f >>> >> reed was not allocated >>> >> Emacs-x86_64-10_14(2433,0x10a8415c0) malloc: *** error for object >>> >> 0x7ffee6db5310: pointer being free >>> >> d was not allocated >>> >> Emacs-x86_64-10_14(2433,0x700003241000) malloc: *** set a breakpoint >>> >> in malloc_error_break to debug >>> >> Emacs-x86_64-10_14(2433,0x10a8415c0) malloc: *** set a breakpoint in >>> >> malloc_error_break to debug >>> >> Fatal error 6: Abort trap >>> >> Abort trap: 6 >>> >> $emacs >>> >> Emacs-x86_64-10_14(2497,0x70000ce9d000) malloc: *** error for object >>> >> 0x7ffee91b45a0: pointer being f >>> >> reed was not allocated >>> >> Emacs-x86_64-10_14(2497,0x70000ce9d000) malloc: *** set a breakpoint >>> >> in malloc_error_break to debug >>> >> Fatal error 4: Illegal instruction >>> >> Abort trap: 6 >>> >> $emacs >>> >> Emacs-x86_64-10_14(2576,0x700002fcd000) malloc: *** error for object >>> >> 0x7ffee07fcac8: pointer being f >>> >> reed was not allocated >>> >> Emacs-x86_64-10_14(2576,0x700002fcd000) malloc: *** set a breakpoint >>> >> in malloc_error_break to debug >>> >> Fatal error 4: Illegal instruction >>> >> Abort trap: 6 >>> >> >>> >> Do you know if there is anything I can do to fix this or is it a bug in >>> >> emacspeak? >>> >> It seems to me as though it is attempting to access a freed object. >>> >> Has anyone else had this issue? >>> > _______________________________________________ >>> > Emacspeak mailing list -- emacspeak(a)emacspeak.org >>> > To unsubscribe send an email to emacspeak-leave(a)emacspeak.org >>> >>> -- >>> Best regards Isaac >>> _______________________________________________ >>> Emacspeak mailing list -- emacspeak(a)emacspeak.org >>> To unsubscribe send an email to emacspeak-leave(a)emacspeak.org >>> >>> _______________________________________________ >>> Emacspeak mailing list -- emacspeak(a)emacspeak.org >>> To unsubscribe send an email to emacspeak-leave(a)emacspeak.org >> _______________________________________________ >> Emacspeak mailing list -- emacspeak(a)emacspeak.org >> To unsubscribe send an email to emacspeak-leave(a)emacspeak.org
|May 1995 - Last Year|Current Year|
If you have questions about this archive or had problems using it, please contact us.