emacs-vr-mode-users Mailing List for Emacs VR Mode (Page 2)
Brought to you by:
grifgrif
You can subscribe to this list here.
2002 |
Jan
|
Feb
(6) |
Mar
(13) |
Apr
|
May
(3) |
Jun
|
Jul
(11) |
Aug
|
Sep
|
Oct
(1) |
Nov
(2) |
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2003 |
Jan
(1) |
Feb
|
Mar
(1) |
Apr
(1) |
May
|
Jun
(1) |
Jul
(1) |
Aug
|
Sep
(4) |
Oct
|
Nov
|
Dec
|
2005 |
Jan
|
Feb
|
Mar
(2) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(2) |
Dec
|
2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(3) |
2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(6) |
From: Patrik J. <pa...@uc...> - 2002-10-25 04:56:05
|
Hey, Sorry for taking so long to get back to you, your message disappeared into the depths of my inbox... The failed assertion usually (I think) occurs when you start dictating before Emacs has finished executing a *voice commander* command, because at that time they disagree on which buffer is active. Unless... which version of Emacs are you using? I'm still plagued by what I believe is multi-byte problems with version 21. I thought I had fixed this but I realized it's still there. If you could tell me what the VR modes process window says when it happens it would help me figure out if you have the same problem. "cannot create voice menu" happens when VR mode doesn't exit properly. If it doesn't unload the command objects you get this error when you try to started again since the voice menu is still there. (If it happens in connection to the above problem, "ignore" the failed assertion and then do escape-x vr-mode in Emacs so the VR mode sub process exits cleanly. After that you can restart it again.) the last one happens sometimes (though very rarely for me). It's usually because Emacs has silently deleted the contents of the minibuffer so the character position of your mode wants to insert at doesn't exist. Try using the "resynchronize" voice command, it usually fixes these problems. I think this also happens more often with Emacs 21, but I don't have the slightest clue why that would be... About your question regarding the VR mode selection, there is an overlay "vr-select-overlay" which contains the selection. You should be able to just read that variable. Regards, /Patrik At 10:41 AM 9/23/2002 +0100, you wrote: >Dear Patrik: > >Thanks for continuing to support VR mode. Unfortunately, I found that >008 beta wasn't stable and had to go back to the older version. Here >are some of the errors I have received. The first one arose far too >often. > >Cheers, >Uday > > > >Assertion failed! >Program: ...\vr.exe >File: c:\program files\pbv kit\vrmode\src\buffer.cpp >Line: 248 > >Expression: p > >--------------- > >Cannot create voice menu: 0x80040626 > >------------------ > >Args out of range: 36,36 > (when trying to dictate a command in the minibuffer) ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Ulrike S. <ul...@go...> - 2002-07-15 22:54:38
|
I also want to thank you very, very much. With the newest release the commands on the voice reference card work fine now and I have to learn more about the others ... Uli Pat Mecheski schrieb: > > Eureka! > > I downloaded the latest build that was recently released (release 8 beta) > and it works! I made no modifications to the VR mode files and left the > following statement in PBVKit's _emacs as is. > > (setq vr-voice-command-list > '( > vr-cachepad-command-list > vr-else-command-list > vr-emacs-command-list > vr-python-command-list > vr-setnu-command-list > vr-translate-mode-command-list > vr-voicegrip-command-list > vr-vr-command-list > )) > > Thanks for the help David and thanks for the new build Patrik! > > Pat > > ------------------------ Yahoo! Groups Sponsor ---------------------~--> > Save on REALTOR Fees > http://us.click.yahoo.com/Xw80LD/h1ZEAA/Ey.GAA/saFolB/TM > ---------------------------------------------------------------------~-> > > Community email addresses: > Post message: Voi...@on... > Subscribe: Voi...@on... > Unsubscribe: Voi...@on... > List owner: Voi...@on... > > Shortcut URL to this page: > http://www.onelist.com/community/VoiceCoder > > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ |
From: Pat M. <pa...@op...> - 2002-07-11 23:27:55
|
Eureka! I downloaded the latest build that was recently released (release 8 beta) and it works! I made no modifications to the VR mode files and left the following statement in PBVKit's _emacs as is. (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) Thanks for the help David and thanks for the new build Patrik! Pat |
From: Pat M. <pa...@op...> - 2002-07-11 18:58:46
|
Hello Ben, I am also having problems with this (see email list emacs-vr-mode-users subject: [Emacs-vr-mode-users] Getting VR Mode to activate (was: pbvki t emacs vr problems). Dave pointed out that there is a problem in vr.el that prevents lists of command lists from working properly. The patch is to replace vr-startup in vr.el with the following (defun vr-startup () "Initialize any per-execution state of the VR Mode subprocess." (let ((l (lambda (x) (cond ((eq x 'vr-default-voice-commands) (mapcar l vr-default-voice-command-list)) ; ; Patch to fix lists of lists. ; Dave Fifield March 2002 ; ((condition-case e (listp (symbol-value x)) ('error nil)) (mapcar l (symbol-value x))) ((symbolp x) (vr-send-cmd (concat "define-command " (vr-strip-dash x) "|" (symbol-name x)))) ((and (listp x) (eq (car x) 'list)) (vr-send-cmd (format "define-list %s|%s" (nth 1 x) (nth 2 x)))) ((and (consp x) (vectorp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (symbolp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (stringp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) (t (error "Unknown vr-voice-command-list element %s" x)) ) ))) (mapcar l (if (eq vr-voice-command-list t) vr-default-voice-command-list vr-voice-command-list))) ;; don't set up these hooks until after initialization has succeeded (add-hook 'post-command-hook 'vr-post-command) (add-hook 'minibuffer-setup-hook 'vr-enter-minibuffer) (add-hook 'kill-buffer-hook 'vr-kill-buffer) (vr-maybe-activate-buffer (current-buffer)) (run-hooks 'vr-mode-startup-hook) ) I tried this patch, but have not been able to get the commands to be defined. There appears to be a problem with the following code in PBVKit's _emacs ; load lists of voice commands for emacs and all of the pbv kit components. ; and turn on vr-mode so the commands are available immediately ; (comment out of vr-voice-command-list the components that you do not use.) (load-library (substitute-in-file-name "$PBVKITDIR/elisp/vr-mode-commands")) (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) If you comment the above setq vr-voice-command-list code then the command defined by vr.el work just fine, but then you don't get the other pbvkit commands. I have been distracted by more pressing problems like work ;-), when I get back to it I will post a solution if I find one. If anyone is successfully using pbvkit with the latest vrmode, please let us know what you had to change. Thanks, pat -----Original Message----- From: Ben Staniford [mailto:be...@st...] Sent: Thursday, July 11, 2002 3:27 AM To: voice coder Subject: [VoiceCoder] PBV-KIT with the latest version of vr mode I have been using the pbv kit for a little while , I'm using NaturallySpeaking 6 and Windows 2000. The VR mode has been quite buggy so I installed the latest version from source forge, after I installed this I found that select' n' say work much better and VR mode did not crash. I also found all the emacs commands that came with the pbv kit stopped working. Can anyone tell me why emacs is not loading the commands? I'm not that familiar with emacs. My _emacs file looks like this: ; VR Mode ;(setq vr-win-class "emacs") (setq vr-win-title "emacs") ; turn on vr mode automatically for certain file types (setq vr-activation-list '("\\.c$" "\\.cpp$" "\\.cxx$" "\\.h$" "\\.hxx$" "\\.pl$" "\\.py$" "\\.el$" "\\.java$" "^\*scratch\*$" "\\.txt$")) ;(setq vr-voice-command-list '(vr-default-voice-command-list)) ; load lists of voice commands for emacs and all of the pbv kit components. ; and turn on vr-mode so the commands are available immediately ; (comment out of vr-voice-command-list the components that you do not use.) (load-library (substitute-in-file-name "$PBVKITDIR/elisp/vr-mode-commands")) (setq vr-voice-command-list '( vr-default-voice-command-list vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) |
From: Dave F. <da...@al...> - 2002-07-10 01:58:02
|
Pat, Is the following line in your _emacs file? (load-library (substitute-in-file-name "$PBVKITDIR/elisp/vr-mode-commands")) It needs to come before: (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) Here's my entire _emacs file for comparison: ============================== ; Sample .emacs file for the PBV Kit ; IMPORTANT: This file will be deleted if you uninstall the PBV Kit, part 2. ; If you want to make changes to this file you should copy this file outside ; of the PBV Kit directory structure, and set your HOME environment variable ; to point to the new directory. See the readme.htm file for details. ; initialize the pbv extensions (load-file (substitute-in-file-name "$PBVKITDIR/elisp/pbvkit-start.el")) ; CachePad ; use a vertical CachePad window along right margin (setq cache-horizontal-split t) (if cache-horizontal-split (progn (setq cache-column-width (if (< (x-display-pixel-width) 800) 16 21)) (setq cache-num-columns 1) (setq cache-num-rows 20) )) ; uncomment the following line if you want a CachePad window to be ; displayed at the start of your emacs session. ;(create-cache-window) ; VR Mode ; turn on vr mode automatically for certain file types (setq vr-activation-list '("\\.c$" "\\.cpp$" "\\.cxx$" "\\.h$" "\\.hxx$" "\\.pl$" "\\.py$" "\\.el$")) ; load lists of voice commands for emacs and all of the pbv kit components. ; and turn on vr-mode so the commands are available immediately ; (comment out of vr-voice-command-list the components that you do not use.) (load-library (substitute-in-file-name "$PBVKITDIR/elisp/vr-mode-commands")) (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) (vr-mode 1) ;; ;; Emacs Niceties ;; ;; The following aren't necessary for programming by voice but they ;; make for a nice Emacs environment. ;; Show Paren mode highlights matching parenthesis. (show-paren-mode 1) ;; Font Lock mode does syntax highlighting. (global-font-lock-mode t) (setq font-lock-maximum-decoration 1) ; no angry fruit salad! ;; Autofill mode will automatically wrap and fill lines. ;; It is intelligent about comments. (add-hook 'c-mode-hook 'turn-on-auto-fill) (add-hook 'emacs-lisp-mode-hook 'turn-on-auto-fill) (add-hook 'perl-mode-hook 'turn-on-auto-fill) (add-hook 'python-mode-hook 'turn-on-auto-fill) (add-hook 'text-mode-hook 'turn-on-auto-fill) (add-hook 'visual-basic-mode-hook 'turn-on-auto-fill) ;; CUA mode makes Emacs follow windows conventions with respect to the home and ;; end keys, extending the selection by holding down shift with arrow keys, ;; and ctrl+c/x/v/z for cut, copy, paste, and undo. (load (substitute-in-file-name "$PBVKITDIR/elisp/cua-mode")) (CUA-mode t) (put 'erase-buffer 'disabled nil) ============================================== dave -- Dave Fifield da...@cs... |
From: Pat M. <pa...@op...> - 2002-07-09 23:58:39
|
I think I am getting closer. The following lines in PBV Kit's _emacs file don't seem to define the commands in pbvkit/elisp/vr-mode-commands.el with VR mode version 6 (works fine with version 4). Even with Dave's patch it the commands don't get defined. (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) If I comment these lines out, the commands defined in vr.el get defined. Any ideas on how to define the commands in pbvkit/elisp/vr-mode-commands.el? Thanks, pat -----Original Message----- From: Dave Fifield [mailto:da...@al...] Sent: Monday, July 08, 2002 6:24 PM To: ema...@li... Subject: RE: [Emacs-vr-mode-users] Getting VR Mode to activate (was: pbvki t emacs vr problems) Pat, Patrik is right, there is a problem in VR Mode version 6 in the handling of vr-voice-command-list. This worked fine in version 4 but was broken in version 6. The problem is that command lists made up of other command lists do not get parsed properly. To fix this problem, replace the vr-startup function definition in vr.el with the one I have included below. In my vr.el, it starts online 1273. The only difference is the addition of the two lines after my comment (see below). I patched it in March of 2002 and sent a copy to Patrik ( Patrik, did you get it?) and it has been working fine ever since. Anyway here's the patch. ================================================= (defun vr-startup () "Initialize any per-execution state of the VR Mode subprocess." (let ((l (lambda (x) (cond ((eq x 'vr-default-voice-commands) (mapcar l vr-default-voice-command-list)) ; ; Patch to fix lists of lists. ; Dave Fifield March 2002 ; ((condition-case e (listp (symbol-value x)) ('error nil)) (mapcar l (symbol-value x))) ((symbolp x) (vr-send-cmd (concat "define-command " (vr-strip-dash x) "|" (symbol-name x)))) ((and (listp x) (eq (car x) 'list)) (vr-send-cmd (format "define-list %s|%s" (nth 1 x) (nth 2 x)))) ((and (consp x) (vectorp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (symbolp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (stringp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) (t (error "Unknown vr-voice-command-list element %s" x)) ) ))) (mapcar l (if (eq vr-voice-command-list t) vr-default-voice-command-list vr-voice-command-list))) ;; don't set up these hooks until after initialization has succeeded (add-hook 'post-command-hook 'vr-post-command) (add-hook 'minibuffer-setup-hook 'vr-enter-minibuffer) (add-hook 'kill-buffer-hook 'vr-kill-buffer) (vr-maybe-activate-buffer (current-buffer)) (run-hooks 'vr-mode-startup-hook) ) ======================================== Cheers, Dave -- Dave Fifield da...@cs... -----Original Message----- From: ema...@li... [mailto:ema...@li...]On Behalf Of Patrik Jonsson Sent: July 8, 2002 5:48 PM To: Pat Mecheski; 'ul...@go...'; 'ema...@li...' Subject: RE: [Emacs-vr-mode-users] Getting VR Mode to activate (was: pbvki t emacs vr problems) hi again, It does indeed look like you only have some weird subset of commands defined.I don't know where this is set, I suggest you grep the pbv kit files, looking for where the variable vr-voice-command-list is set. This variable is set by VR Mode to the default command list in vr.el, but it's probably being overridden in some other file. The simplest thing for you to do is probably to comment out things in the pbv kit start up files and instead just change the command list in VR.el Good luck, /Patrik P.S. Are you using Emacs voice commander as well? That should enable you to use most emacs commands by voice, without explicitly defining them as VR Mode commands. There are some advantages to use VR Mode commands, mostly that VR Mode can't see changes to the buffer that occur as a result of a voice commander command, and sometimes when using Emacs over a network connection, the voice commander commands don't seem to be sent correctly. VR Mode commands are also slightly faster. On the other hand, voice commander has a mode-specific mechanism for enabling commands, which VR Mode doesn't. At 12:30 PM 7/8/2002 -0700, Pat Mecheski wrote: >Hello Patrik! > >Thanks for responding! I am still stuck. > >Below is the output of my VR Mode Process Status window. > >I have only modified _emacs to properly set vr-win-class and vr-win-title. >All other files installed with the pbv-kit have remained unchanged. > >Note that selecting by voice works. I am just unable to use the commands >like "find file" and "quit emacs". It looks like they are not being loaded. >0 -> define-command vr cachepad command list|vr-cachepad-command-list >0 -> define-command vr else command list|vr-else-command-list >0 -> define-command vr emacs command list|vr-emacs-command-list >0 -> define-command vr python command list|vr-python-command-list >0 -> define-command vr setnu command list|vr-setnu-command-list >0 -> define-command vr translate mode command >list|vr-translate-mode-command-lis... >0 -> define-command vr voicegrip command list|vr-voicegrip-command-list >0 -> define-command vr vr command list|vr-vr-command-list ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users |
From: Patrik J. <pa...@uc...> - 2002-07-09 01:44:47
|
>I patched it in March of 2002 and sent a copy to Patrik ( >Patrik, did you get it?) and it has been working fine ever since. yes, I did get it but then it fell between the cracks. Thanks for reminding me, it'll make it into the next update! /Patrik ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Dave F. <da...@al...> - 2002-07-09 01:21:36
|
Pat, Patrik is right, there is a problem in VR Mode version 6 in the handling of vr-voice-command-list. This worked fine in version 4 but was broken in version 6. The problem is that command lists made up of other command lists do not get parsed properly. To fix this problem, replace the vr-startup function definition in vr.el with the one I have included below. In my vr.el, it starts online 1273. The only difference is the addition of the two lines after my comment (see below). I patched it in March of 2002 and sent a copy to Patrik ( Patrik, did you get it?) and it has been working fine ever since. Anyway here's the patch. ================================================= (defun vr-startup () "Initialize any per-execution state of the VR Mode subprocess." (let ((l (lambda (x) (cond ((eq x 'vr-default-voice-commands) (mapcar l vr-default-voice-command-list)) ; ; Patch to fix lists of lists. ; Dave Fifield March 2002 ; ((condition-case e (listp (symbol-value x)) ('error nil)) (mapcar l (symbol-value x))) ((symbolp x) (vr-send-cmd (concat "define-command " (vr-strip-dash x) "|" (symbol-name x)))) ((and (listp x) (eq (car x) 'list)) (vr-send-cmd (format "define-list %s|%s" (nth 1 x) (nth 2 x)))) ((and (consp x) (vectorp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (symbolp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) ((and (consp x) (stringp (cdr x))) (vr-send-cmd (format "define-command %s|%s" (car x) (cdr x)))) (t (error "Unknown vr-voice-command-list element %s" x)) ) ))) (mapcar l (if (eq vr-voice-command-list t) vr-default-voice-command-list vr-voice-command-list))) ;; don't set up these hooks until after initialization has succeeded (add-hook 'post-command-hook 'vr-post-command) (add-hook 'minibuffer-setup-hook 'vr-enter-minibuffer) (add-hook 'kill-buffer-hook 'vr-kill-buffer) (vr-maybe-activate-buffer (current-buffer)) (run-hooks 'vr-mode-startup-hook) ) ======================================== Cheers, Dave -- Dave Fifield da...@cs... -----Original Message----- From: ema...@li... [mailto:ema...@li...]On Behalf Of Patrik Jonsson Sent: July 8, 2002 5:48 PM To: Pat Mecheski; 'ul...@go...'; 'ema...@li...' Subject: RE: [Emacs-vr-mode-users] Getting VR Mode to activate (was: pbvki t emacs vr problems) hi again, It does indeed look like you only have some weird subset of commands defined.I don't know where this is set, I suggest you grep the pbv kit files, looking for where the variable vr-voice-command-list is set. This variable is set by VR Mode to the default command list in vr.el, but it's probably being overridden in some other file. The simplest thing for you to do is probably to comment out things in the pbv kit start up files and instead just change the command list in VR.el Good luck, /Patrik P.S. Are you using Emacs voice commander as well? That should enable you to use most emacs commands by voice, without explicitly defining them as VR Mode commands. There are some advantages to use VR Mode commands, mostly that VR Mode can't see changes to the buffer that occur as a result of a voice commander command, and sometimes when using Emacs over a network connection, the voice commander commands don't seem to be sent correctly. VR Mode commands are also slightly faster. On the other hand, voice commander has a mode-specific mechanism for enabling commands, which VR Mode doesn't. At 12:30 PM 7/8/2002 -0700, Pat Mecheski wrote: >Hello Patrik! > >Thanks for responding! I am still stuck. > >Below is the output of my VR Mode Process Status window. > >I have only modified _emacs to properly set vr-win-class and vr-win-title. >All other files installed with the pbv-kit have remained unchanged. > >Note that selecting by voice works. I am just unable to use the commands >like "find file" and "quit emacs". It looks like they are not being loaded. >0 -> define-command vr cachepad command list|vr-cachepad-command-list >0 -> define-command vr else command list|vr-else-command-list >0 -> define-command vr emacs command list|vr-emacs-command-list >0 -> define-command vr python command list|vr-python-command-list >0 -> define-command vr setnu command list|vr-setnu-command-list >0 -> define-command vr translate mode command >list|vr-translate-mode-command-lis... >0 -> define-command vr voicegrip command list|vr-voicegrip-command-list >0 -> define-command vr vr command list|vr-vr-command-list ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users |
From: Patrik J. <pa...@uc...> - 2002-07-08 20:18:37
|
hi again, It does indeed look like you only have some weird subset of commands defined.I don't know where this is set, I suggest you grep the pbv kit files, looking for where the variable vr-voice-command-list is set. This variable is set by VR Mode to the default command list in vr.el, but it's probably being overridden in some other file. The simplest thing for you to do is probably to comment out things in the pbv kit start up files and instead just change the command list in VR.el Good luck, /Patrik P.S. Are you using Emacs voice commander as well? That should enable you to use most emacs commands by voice, without explicitly defining them as VR Mode commands. There are some advantages to use VR Mode commands, mostly that VR Mode can't see changes to the buffer that occur as a result of a voice commander command, and sometimes when using Emacs over a network connection, the voice commander commands don't seem to be sent correctly. VR Mode commands are also slightly faster. On the other hand, voice commander has a mode-specific mechanism for enabling commands, which VR Mode doesn't. At 12:30 PM 7/8/2002 -0700, Pat Mecheski wrote: >Hello Patrik! > >Thanks for responding! I am still stuck. > >Below is the output of my VR Mode Process Status window. > >I have only modified _emacs to properly set vr-win-class and vr-win-title. >All other files installed with the pbv-kit have remained unchanged. > >Note that selecting by voice works. I am just unable to use the commands >like "find file" and "quit emacs". It looks like they are not being loaded. >0 -> define-command vr cachepad command list|vr-cachepad-command-list >0 -> define-command vr else command list|vr-else-command-list >0 -> define-command vr emacs command list|vr-emacs-command-list >0 -> define-command vr python command list|vr-python-command-list >0 -> define-command vr setnu command list|vr-setnu-command-list >0 -> define-command vr translate mode command >list|vr-translate-mode-command-lis... >0 -> define-command vr voicegrip command list|vr-voicegrip-command-list >0 -> define-command vr vr command list|vr-vr-command-list ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Pat M. <pa...@op...> - 2002-07-08 19:26:47
|
Hello Patrik! Thanks for responding! I am still stuck. Below is the output of my VR Mode Process Status window. I have only modified _emacs to properly set vr-win-class and vr-win-title. All other files installed with the pbv-kit have remained unchanged. Note that selecting by voice works. I am just unable to use the commands like "find file" and "quit emacs". It looks like they are not being loaded. Any ideas appreciated, thanks! pat Emacs VR Mode, Copyright (C) 1999 Barry Jaspan This is free software, and you are welcome to redistribute it under certain conditions; see the accompanying file COPYING.txt <- (listening 1094) 0 <- (connected) 0 -> initialize Emacs|nil|393606 0 <- (initialize succeeded) 0 <- (mic-state on) Wnd (393606) Emacs/Emacs (Lisp Interaction) - matches Emacs/* 0 <- (frame-activated 393606) 0 -> define-command vr cachepad command list|vr-cachepad-command-list 0 -> define-command vr else command list|vr-else-command-list 0 -> define-command vr emacs command list|vr-emacs-command-list 0 -> define-command vr python command list|vr-python-command-list 0 -> define-command vr setnu command list|vr-setnu-command-list 0 -> define-command vr translate mode command list|vr-translate-mode-command-lis... 0 -> define-command vr voicegrip command list|vr-voicegrip-command-list 0 -> define-command vr vr command list|vr-vr-command-list 0 -> deactivate-buffer 0 <- (recognition begin) 0 <- (recognition end) -----Original Message----- From: Patrik Jonsson [mailto:pa...@uc...] Sent: Monday, July 08, 2002 11:43 AM To: ul...@go...; ema...@li... Subject: [Emacs-vr-mode-users] Getting VR Mode to activate (was: pbvkit emacs vr problems) hi, There are two possible (OK, likely, there are many possible) sources to this problem. Look in the VR Mode protocol trace window. When you select the Emacs window, you should see something like this: Wnd (722020) Emacs/emacs at griffin @lisp-interaction-mode@lisp-mode@||||||||||||@vr-mode@font-lock-mode@abbrev- mode matches Emacs/emacs at griffin after the number in parentheses is the win-class and win-title of the Emacs window, after the "matches" is your VR-win-class/VR-win-title. If they are set incorrectly, you will see "doesn't match" instead of "matches". If you pass this test, make sure that the commands are actually loading. look at the protocol trace when you first start VR Mode. You should see something like this: Emacs VR Mode, Copyright (C) 1999 Barry Jaspan This is free software, and you are welcome to redistribute it under certain conditions; see the accompanying file COPYING.txt <- (listening 2100) 0 <- (connected) 0 -> initialize Emacs|emacs at griffin|722020 0 <- (initialize succeeded) 0 <- (mic-state on) Wnd (722020) Emacs/emacs at griffin @lisp-interaction-mode@lisp-mode@||||||||||||@vr-mode@font-lock-mode@abbrev- mode matches Emacs/emacs at griffin 0 <- (frame-activated 722020) 0 -> define-list 0to20|0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 -> define-command activate buffer|vr-add-to-activation-list 0 -> define-command quit|[7] 0 -> define-command undo|undo ... and a lot more like this the define-commands tell you which commands VR Mode will recognize. I think there is some incompatibility in the pbv kit start up files and the current version of VR Mode, so the commands you think are actually set may not be so. if you still have problems, let me know. /Patrik At 06:18 PM 7/8/2002 +0900, ul...@go... wrote: >Pat Mecheski wrote: > > > I fixed the emacs frame window handle problem by adding the following lines > > to pbvkit's _emacs file > > > > (setq vr-win-class '"Emacs") > > (setq vr-win-title '"") > > > > I still can't seem to get vr-mode to recognize basic commands like "find > > file" or "quit emacs". > >I have the same problem. The only, possibly irrelevant, difference might be >that put the line > > (setq vr-win-title "Emacs") > >into pbvkit-start.el > >I did not see any reply to Pat's question in the list-archive. Is is it >possible to solve that problem? > >TIA, >Uli ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users |
From: Patrik J. <pa...@uc...> - 2002-07-08 18:43:18
|
hi, There are two possible (OK, likely, there are many possible) sources to this problem. Look in the VR Mode protocol trace window. When you select the Emacs window, you should see something like this: Wnd (722020) Emacs/emacs at griffin @lisp-interaction-mode@lisp-mode@||||||||||||@vr-mode@font-lock-mode@abbrev-mode matches Emacs/emacs at griffin after the number in parentheses is the win-class and win-title of the Emacs window, after the "matches" is your VR-win-class/VR-win-title. If they are set incorrectly, you will see "doesn't match" instead of "matches". If you pass this test, make sure that the commands are actually loading. look at the protocol trace when you first start VR Mode. You should see something like this: Emacs VR Mode, Copyright (C) 1999 Barry Jaspan This is free software, and you are welcome to redistribute it under certain conditions; see the accompanying file COPYING.txt <- (listening 2100) 0 <- (connected) 0 -> initialize Emacs|emacs at griffin|722020 0 <- (initialize succeeded) 0 <- (mic-state on) Wnd (722020) Emacs/emacs at griffin @lisp-interaction-mode@lisp-mode@||||||||||||@vr-mode@font-lock-mode@abbrev-mode matches Emacs/emacs at griffin 0 <- (frame-activated 722020) 0 -> define-list 0to20|0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 -> define-command activate buffer|vr-add-to-activation-list 0 -> define-command quit|[7] 0 -> define-command undo|undo ... and a lot more like this the define-commands tell you which commands VR Mode will recognize. I think there is some incompatibility in the pbv kit start up files and the current version of VR Mode, so the commands you think are actually set may not be so. if you still have problems, let me know. /Patrik At 06:18 PM 7/8/2002 +0900, ul...@go... wrote: >Pat Mecheski wrote: > > > I fixed the emacs frame window handle problem by adding the following lines > > to pbvkit's _emacs file > > > > (setq vr-win-class '"Emacs") > > (setq vr-win-title '"") > > > > I still can't seem to get vr-mode to recognize basic commands like "find > > file" or "quit emacs". > >I have the same problem. The only, possibly irrelevant, difference might be >that put the line > > (setq vr-win-title "Emacs") > >into pbvkit-start.el > >I did not see any reply to Pat's question in the list-archive. Is is it >possible to solve that problem? > >TIA, >Uli ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: <ul...@go...> - 2002-07-08 09:27:02
|
Pat Mecheski wrote: > I fixed the emacs frame window handle problem by adding the following lines > to pbvkit's _emacs file > > (setq vr-win-class '"Emacs") > (setq vr-win-title '"") > > I still can't seem to get vr-mode to recognize basic commands like "find > file" or "quit emacs". I have the same problem. The only, possibly irrelevant, difference might be that put the line (setq vr-win-title "Emacs") into pbvkit-start.el I did not see any reply to Pat's question in the list-archive. Is is it possible to solve that problem? TIA, Uli |
From: Pat M. <pa...@op...> - 2002-05-17 23:26:48
|
I fixed the emacs frame window handle problem by adding the following lines to pbvkit's _emacs file (setq vr-win-class '"Emacs") (setq vr-win-title '"") I still can't seem to get vr-mode to recognize basic commands like "find file" or "quit emacs". I have no problem using these voice commands with the version of vr-mode (version 4) that came with pbvkit. When I upgraded pbvkit's vrmode subdirectory to version 7b, vrmode stopped recognizing the commands. Are there other changes that must be made to pbvkit's _emacs file? Thanks, pat -----Original Message----- From: Pat Mecheski [mailto:pa...@op...] Sent: Thursday, May 16, 2002 11:48 AM To: 'Voi...@ya...' Subject: [VoiceCoder] pbvkit emacs vr problems Hello, I installed the PBVKit and upgraded the vrmode subdirectory with release 7. When I start emacs, I get the following message at the bottom of the window. VR Mode: 4260484 is not an Emacs frame window handle; ignored. Is this a problem? The mini buffer reports VR-on. When I say a command like "find file", the message at the bottom of the window flashes "Waiting for voice recognition..." and the command is not recognized just the text is written into the buffer. I suspect something is wrong with my emacs defaults file. I am using the default _emacs files supplied with the PBVKit, the entry for vr-mode follows below. Any help is greatly appreciated. I am running Windows 2000 with Dragon NaturallySpeaking Professional 6. Thanks, Pat ; VR Mode ; turn on vr mode automatically for certain file types (setq vr-activation-list '("\\.c$" "\\.cpp$" "\\.cxx$" "\\.h$" "\\.hxx$" "\\.pl$" "\\.py$" "\\.el$")) ; load lists of voice commands for emacs and all of the pbv kit components. ; and turn on vr-mode so the commands are available immediately ; (comment out of vr-voice-command-list the components that you do not use.) (load-library (substitute-in-file-name "$PBVKITDIR/elisp/vr-mode-commands")) (setq vr-voice-command-list '( vr-cachepad-command-list vr-else-command-list vr-emacs-command-list vr-python-command-list vr-setnu-command-list vr-translate-mode-command-list vr-voicegrip-command-list vr-vr-command-list )) (vr-mode 1) Yahoo! Groups Sponsor ADVERTISEMENT <http://rd.yahoo.com/M=226020.2031982.3521154.1971030/D=egroupweb/S=17050067 64:HM/A=1073422/R=0/*http://www.qksrv.net/click-1084497-610300?url=http://ww w.tv/en/landing.shtml&HHcode=yhc3my2> <http://us.adserver.yahoo.com/l?M=226020.2031982.3521154.1971030/D=egroupmai l/S=1705006764:HM/A=1073422/rand=559408887> Community email addresses: Post message: Voi...@on... Subscribe: Voi...@on... Unsubscribe: Voi...@on... List owner: Voi...@on... Shortcut URL to this page: http://www.onelist.com/community/VoiceCoder <http://www.onelist.com/community/VoiceCoder> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service <http://docs.yahoo.com/info/terms/> . |
From: Stuart N. <st...@cs...> - 2002-05-17 00:48:47
|
I happen to know that the most recent maintainer of VR-mode, Patrik Jonsson, has just left for a conference in Hawaii (poor guy). I'm sure he'll have some ideas when he returns, but you might have to wait until the end of the month, unless someone else has some ideas... Stuart From: Glen Harris <gle...@ar...> Date: Thu, 16 May 2002 18:47:16 -0500 I am a beginner user of VR mode. I am having a problem similar to that described by Uday S. Reddy (03/13/2002). The mode line in my emacs window states "VR-on", and not "VR:on", which I believe (from Patrik Jonsson's posting) is indicative of a problem although I am not sure what or how to solve. The VR Mode Process Status window shows that I have properly set the vr-win-class & vr-win-title variables. I see this in the Process Status when I click in my Emacs window: Wnd (4195656) Emacs/emacs@AUS-PORT26 matches Emacs/emacs@AUS-PORT26 Apparently I am using NaturalText(?) to dictate to emacs window instead of VR mode, because I see "recognition begin/end" in the VR mode process status window (and text does indeed appear in the emacs buffer), however I cannot use any of the select or correction features. Thanks for any assistance. Glen ------------------------------------------------------------------- Glen Harris Advanced RISC Machines email: gle...@ar... Austin Design Center phone: +1 (512) 314 1064 Building 3, Suite 560 fax: +1 (512) 314 1078 1250 Capital of Texas Hwy Austin, Texas 78746 ------------------------------------------------------------------- _______________________________________________________________ Have big pipes? SourceForge.net is looking for download mirrors. We supply the hardware. You get the recognition. Email Us: ban...@so... _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users |
From: Glen H. <gle...@ar...> - 2002-05-16 23:46:29
|
I am a beginner user of VR mode. I am having a problem similar to that described by Uday S. Reddy (03/13/2002). The mode line in my emacs window states "VR-on", and not "VR:on", which I believe (from Patrik Jonsson's posting) is indicative of a problem although I am not sure what or how to solve. The VR Mode Process Status window shows that I have properly set the vr-win-class & vr-win-title variables. I see this in the Process Status when I click in my Emacs window: Wnd (4195656) Emacs/emacs@AUS-PORT26 matches Emacs/emacs@AUS-PORT26 Apparently I am using NaturalText(?) to dictate to emacs window instead of VR mode, because I see "recognition begin/end" in the VR mode process status window (and text does indeed appear in the emacs buffer), however I cannot use any of the select or correction features. Thanks for any assistance. Glen ------------------------------------------------------------------- Glen Harris Advanced RISC Machines email: gle...@ar... Austin Design Center phone: +1 (512) 314 1064 Building 3, Suite 560 fax: +1 (512) 314 1078 1250 Capital of Texas Hwy Austin, Texas 78746 ------------------------------------------------------------------- |
From: Chris G. <chr...@mo...> - 2002-03-28 08:52:21
|
Finally I found a specific problem: If I use vr-mode with a document containing a specific character the memory allocation problems occur. This character looks the same as a space in Emacs. The source was a web page I cut and pasted from. I found this character was in the buffer when I tried to save the file. Emacs highlighted the offending character and complained that I need to choose a coding system. When I deleted the offending character vr-mode was happy again. I have attached a one line text document with the offending character marked. With this document I don't actually get a message about the enormous memory allocation, but I do get temporary freezes and messages like: 0 <- (get-buffer-info "nastycharacter.txt" 2) 0 -> selection start FAILED: -1, 0 reading data timeout waiting for selection start 0 <- (make-changes 0 0 "" 0 10) Thank you for your patience, Chris Gaskett chr...@mo... www.geocities.com/~chris_gaskett --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.336 / Virus Database: 188 - Release Date: 2002/03/11 |
From: Chris G. <chr...@mo...> - 2002-03-27 01:21:15
|
Oh dear. The mystery is not solved. Turning off the Japanese settings gave me some respite from the enormous memory allocation problem, but it is still occurring. As before, it only occurs with certain files, and changes to the file sometimes stop it from occurring. Seems to me that certain patterns of characters with particular byte alignments cause the problem. I have not looked at how vr-mode detects the end of the stream, but I wonder if it can be a improved. Otherwise, perhaps a limit could be imposed on the length of stream read or the ensuing memory allocation as a harm minimisation strategy. Thank you as always for your help, Chris Gaskett chr...@mo... www.geocities.com/~chris_gaskett --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.336 / Virus Database: 188 - Release Date: 2002/03/11 |
From: Darren R. <dra...@ya...> - 2002-03-17 01:34:59
|
Hello, There's an error in the readme.txt file that had me scratching my head for a few minutes. In the "Custom voice commands" section, it says that you can include the default commands in your command list by using 'vr-default-voice-command-list'. This is incorrect, it should be 'vr-default-voice-commands' instead. Darren __________________________________________________ Do You Yahoo!? Yahoo! Sports - live college hoops coverage http://sports.yahoo.com/ |
From: Patrik J. <pa...@uc...> - 2002-03-14 02:37:52
|
Hi Uday, It seems vr mode does not consider your buffer active. If the mode line indeed says "VR:on", and not "VR-on", the most likely issue is that NaturallySpeaking does not recognize the Emacs window. When you select the Emacs window, you should see a line like Wnd (1114596) Emacs/emacs at griffin @c++-mode@ccmode-mode@||||||||||||@else-mode@vr-mode@font-lock-mode@abbrev-mode matches Emacs/emacs at griffin it should say "matches", if it says "doesn't match", you haven't set the variables vr-win-class and vr-win-title to the correct values. They should be set to something that uniquely identifies the Emacs window running vr mode. On the line above the word between the number and the/(in this case, Emacs), is the win-class, and the thing between the/and "matches" is the win-title. The corresponding strings after "matches" is what your vr-win-class and vr-win title are set to. If it actually does say "matches", there's something strange going on. It should work, get back to me if it doesn't. The way you tell whether you're dictating using vr mode or using NaturalText (Which Just Sends Keystrokes to the Emacs Window), Is to look for communication like "make-changes number number number "what you just said" in the process status window. This is NaturallySpeaking actually communicating with Emacs through vr mode. If you just see "recognition begin/end", but you can still dictate, it's most likely just sending keystrokes, in which case "Select Something" won't work. Hope that helps, /Patrik At 12:59 AM 3/14/2002 +0000, Uday S. Reddy wrote: >I have just installed VR mode, and am running into a problem. I hope >this is a common problem and others can tell me what might be going >wrong. > >I can dictate into an Emacs buffer fine. The modeline says "VR:on". >But, I am not able to do any select-n-say functions. Select <word> as >well as Correct <word> both do nothing. Beetween saying select and ><word>, I get a message in the minibuffer to the effect of "waiting >for dictation...". The VR Mode Process Status window spews out >(recognition begin) and (recognition end). So, as far as I can see, >the utterances are going to the right place. But, nothing happens. > >What can I do? > >Cheers, >Uday > > > >_______________________________________________ >Emacs-vr-mode-users mailing list >Ema...@li... >https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Uday S. R. <U....@cs...> - 2002-03-14 00:58:28
|
I have just installed VR mode, and am running into a problem. I hope this is a common problem and others can tell me what might be going wrong. I can dictate into an Emacs buffer fine. The modeline says "VR:on". But, I am not able to do any select-n-say functions. Select <word> as well as Correct <word> both do nothing. Beetween saying select and <word>, I get a message in the minibuffer to the effect of "waiting for dictation...". The VR Mode Process Status window spews out (recognition begin) and (recognition end). So, as far as I can see, the utterances are going to the right place. But, nothing happens. What can I do? Cheers, Uday |
From: Patrik J. <pa...@uc...> - 2002-03-13 06:54:20
|
weird! Well, as long as it works... :-) /Patrik At 09:59 AM 3/13/2002 +0900, Chris Gaskett wrote: >Mystery soved! > >... and it's my fault as usual. > >Several months ago I made some changes to my .emacs file so that I could cut >and paste in Japanese, including cutting and pasting from other MS Windows >applications. > >I added the following lines: >(set-language-environment "Japanese") >(set-default-coding-systems 'japanese-shift-jis) >(set-selection-coding-system 'japanese-shift-jis) > >I tested vr-mode a bit at the time, and everything seemed okay. But now it >seems that some text, given a particular alignment, causes failure. > >Today I tried just: >(set-language-environment "Japanese") > >And this is also a problem, just with different errors. > >So for the moment I have a workaround. I don't know anything about how these >different coding systems work, or how to detect the end of the stream in a >more fail safe manner. >I think it's not a high priority to try to fix it at the moment. > >Sorry again to waste peoples time, at least if this happens to someone else >we have a good idea where to start looking! > >Chris Gaskett >chr...@mo... >www.geocities.com/~chris_gaskett > >--- >Outgoing mail is certified Virus Free. >Checked by AVG anti-virus system (http://www.grisoft.com). >Version: 6.0.323 / Virus Database: 180 - Release Date: 2002/02/08 > > > >_______________________________________________ >Emacs-vr-mode-users mailing list >Ema...@li... >https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Chris G. <chr...@mo...> - 2002-03-13 00:58:50
|
Mystery soved! ... and it's my fault as usual. Several months ago I made some changes to my .emacs file so that I could cut and paste in Japanese, including cutting and pasting from other MS Windows applications. I added the following lines: (set-language-environment "Japanese") (set-default-coding-systems 'japanese-shift-jis) (set-selection-coding-system 'japanese-shift-jis) I tested vr-mode a bit at the time, and everything seemed okay. But now it seems that some text, given a particular alignment, causes failure. Today I tried just: (set-language-environment "Japanese") And this is also a problem, just with different errors. So for the moment I have a workaround. I don't know anything about how these different coding systems work, or how to detect the end of the stream in a more fail safe manner. I think it's not a high priority to try to fix it at the moment. Sorry again to waste peoples time, at least if this happens to someone else we have a good idea where to start looking! Chris Gaskett chr...@mo... www.geocities.com/~chris_gaskett --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.323 / Virus Database: 180 - Release Date: 2002/02/08 |
From: Chris G. <chr...@mo...> - 2002-03-12 09:17:43
|
Hi Patrik (and others), If I put the text below (without the hyphens above and below) into a file called test.txt the memory allocation error occurs on the first utterance after the buffer is activated: ----------------------------------------------------- The use of Advantage Learning prevents the \myq-values from diverging to 1 but does not prevent policy degradation. The increased variance introduced by Advantage Learning with small $k$ could be a problem. Following the by \citeasnoun{lin92} that the experience replay process can cause unlearning if precautions are not taken, several to preventing unlearning are tested: \begin{enumerate} \item disposing of experiences after a time limit or after they have been replayed several times; \item maintaining separate data bases for reward = 0 and reward = -1 to try to maintain a balance of these experiences; \item various numbers of hidden nodes; \item various fixed learning rates; and \item regularising the neural network weights. \end{enumerate} These measures are ineffective in this case. This experiment confirms that the convergence problem is not limited to finding a good policy, but also includes maintaining a good policy. Use of the new off-policy TD($\lambda$) algorithm proposed by \citeasnoun{off-policy-eligibility} is suggested as a fruitful avenue for investigation as it may improve stability by reducing the reliance on bootstrapping. ------------------------------------------------------ There should be linebreaks only after "problem.", "tested:", each item in the list (includeing {enumerate}), "case." and "bootstrapping." Strangely, I can't isolate any particular part or the text that causes the problem. If I remove any lines there is no memory allocation error. I removed some words from the original text and the error still occurs, but if I change large chunks the memory allocation error stops occurring. If I remove enough to stop the error, run vr-mode, then undo the changes it doesn't give an error ... until I do vr-resychronize. I did some editing on the files that don't cause a freeze but require constant resynchronisation. These files also produce memory allocation errors. I did all these tests on a fresh "out of the box" vr-mode to make sure it wasn't something I changed. Thanks for your help, Chris Gaskett -----Original Message----- From: ema...@li... [mailto:ema...@li...]On Behalf Of Patrik Jonsson Sent: Tuesday, March 12, 2002 4:23 PM To: Chris Gaskett; ema...@li... Subject: Re: [Emacs-vr-mode-users] enormous memory allocation and freeze At 03:58 PM 3/12/2002 +0900, Chris Gaskett wrote: >0 -> selection start FAILED: out of memory allocating 1696625512 bytes >timeout waiting for selection start Christ! I have never, ever, seen that, and I can't even imagine what it's trying to do! that error means that it seems to be reading garbage from the socket connected to Emacs... has this always happened? If not, when did it start? Can you isolate the conditions that makes it happen? I'm completely out of my league here, I don't know anything about how these things work... :-( Followed by the entire file I am editing, consisting of many screens of text >(1 chapter, 38kb). I expected to only see one screen worth of text being >communicated. I didn't mark or select the text or scroll around to cause >this to occur, nor is all the text on 1 enormous line. The freezes occur >whether or not I have vr-mode logging on and status window open. this is standard. Vr mode downloads the entire buffer, and from that point on uses differential updates until it realizes (or you tell it) that it needs to resynchronize again. Whether this is the best way can be discussed... /Patrik ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... _______________________________________________ Emacs-vr-mode-users mailing list Ema...@li... https://lists.sourceforge.net/lists/listinfo/emacs-vr-mode-users --- Incoming mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.323 / Virus Database: 180 - Release Date: 2002/02/08 --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.323 / Virus Database: 180 - Release Date: 2002/02/08 |
From: Patrik J. <pa...@uc...> - 2002-03-12 07:24:13
|
At 03:58 PM 3/12/2002 +0900, Chris Gaskett wrote: >0 -> selection start FAILED: out of memory allocating 1696625512 bytes >timeout waiting for selection start Christ! I have never, ever, seen that, and I can't even imagine what it's trying to do! that error means that it seems to be reading garbage from the socket connected to Emacs... has this always happened? If not, when did it start? Can you isolate the conditions that makes it happen? I'm completely out of my league here, I don't know anything about how these things work... :-( Followed by the entire file I am editing, consisting of many screens of text >(1 chapter, 38kb). I expected to only see one screen worth of text being >communicated. I didn't mark or select the text or scroll around to cause >this to occur, nor is all the text on 1 enormous line. The freezes occur >whether or not I have vr-mode logging on and status window open. this is standard. Vr mode downloads the entire buffer, and from that point on uses differential updates until it realizes (or you tell it) that it needs to resynchronize again. Whether this is the best way can be discussed... /Patrik ============================================================ Patrik Jonsson (831) 459-3809 Department of Astronomy & Astrophysics University of California, Santa Cruz, CA 95064 This message has been written using a voice recognition system. Words that don't make sense or not the fault of the author... |
From: Chris G. <chr...@mo...> - 2002-03-12 06:59:31
|
Dear friends, After having some success with the vr-mode I'm having trouble again, but only with some files. As usual, I'm having trouble finding the pattern. I am using vr-mode v6 patch 1, Dragon NaturallySpeaking preferred version 5, under Windows 2000. I'm getting some freezes just after making an utterance. The status window shows: 0 <- (recognition begin) 0 <- (get-buffer-info "consequences.tex" 779) 0 -> selection start FAILED: out of memory allocating 1696625512 bytes timeout waiting for selection start 0 <- (make-changes 18 0 "continuous state reinforcement learning " 57 0) 0 -> tick FAILED: out of memory allocating 1870097952 bytes timeout waiting for tick 0 <- (recognition end) In the *vr* buffer on opening the document I get: -> (get-buffer-info "consequences.tex" -1) get-buffer-info: current buffer: consequences.tex vr-buffer:consequences.tex #<overlay in no buffer> nil <- r 328 <- r 328 <- r 0 <- r 977 <- r 1 modified <- r 779 <- r 38260 <- r \chapter{Issues for Continuous \myq-Learning}\label{sec:issues} \begin{Quote} A monk asked Followed by the entire file I am editing, consisting of many screens of text (1 chapter, 38kb). I expected to only see one screen worth of text being communicated. I didn't mark or select the text or scroll around to cause this to occur, nor is all the text on 1 enormous line. The freezes occur whether or not I have vr-mode logging on and status window open. Editing some other files is sometimes okay, some of which are twice the size of the file above, but I lose synchronisation often. Often M-x vr-resynchronise does not fix the synchronisation for me, even temporarily. I welcome any suggestions. Yours sincerely, Chris Gaskett chr...@mo... www.geocities.com/~chris_gaskett --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.323 / Virus Database: 180 - Release Date: 2002/02/08 |