Re: problem with left-handed mouse on X - is GPM to blame or is it X?
On Mon, May 10, 2004 at 07:28:57PM -0400, Zephaniah E. Hull wrote: Please see the last message, where I covered just this. Raw mode repeats byte for byte, it does not attempt to do ANYTHING to them. In addition, he is repeating a PS2 protocol, with all the problems involved in that. I will say it once more, this is a case of user error. The proper configuration for this is having gpm not repeat, and both X and gpm reading from /dev/input/mice, with X doing some button remapping in whatever manner is now suggested. Okay. I checked the manpage for this information and did not find it. I also found that the gpm(8) manpage is ugly. So I did something about it. Please see two attached files; one is a diff which patches the manpage to simply document these two minor points. The other one overhauls the manpage completely. I looked into the source package, but the method used to generate both Texinfo and Roff document source from a single file is truly grody. If 'xmodmap -e pointer = 3 2 1' is no longer the suggested means, what is? That's a good question, and I don't have the answer at present. -- G. Branden Robinson| You are not angry with people when Debian GNU/Linux | you laugh at them. Humor teaches [EMAIL PROTECTED] | them tolerance. http://people.debian.org/~branden/ | -- W. Somerset Maugham --- gpm.8~ 2004-05-20 15:05:38.0 -0500 +++ gpm.8 2004-05-20 15:09:34.0 -0500 @@ -145,6 +145,10 @@ normal sequence, `321' can be used by left-handed people, and `132' can be useful with two-button mice (especially within Emacs). All the button permutations are allowable. +.IP +.B Note: +This option is ignored if the \(oq\-R\(cq option is used with the parameter +\(oqraw\(cq. .TP -d \fIdelta\fP Set the delta value. When a single motion event @@ -231,6 +235,9 @@ an easy way to manage those stupid dual-mode mice which force you to keep the middle button down while changing video mode. The option is forced on by the `-M' option. +.IP +.B Note: The only repeat type reliably supported by version 4.3 or later of +.B the XFree86 X server is \(oqms3\(cq. .TP -s \fInumber\fP Set the sample rate for the mouse device. .TH gpm 8 20 May 2004 GPM General Purpose Mouse 1.19.6 .SH NAME gpm \- cut\-and\-paste utility and mouse server for virtual consoles .SH SYNOPSIS .B gpm [ .I options ] .br .SH DESCRIPTION GPM (General Purpose Mouse) tries to be a useful mouse server for applications running on the Linux console. The information below is extracted from the Texinfo documentation, which is the preferred source of information. .PP GPM makes it easy to select and copy regions of text on the virtual console, much as is done in graphical programs such as word processors. Text regions are highlighted with the mouse in an area called the \(oqselection buffer\(cq, or just \(oqselection\(cq. The term \(oqcut\-and\-paste\(cq can be a little misleading, because GPM never deletes text from the console. GPM is actually more of a .RI \(oq copy \-and\-paste\(cq utility. .PP The .B gpm executable is meant to act like a daemon (thus, \(oqgpmd\(cq would be a better name for it). .SS Operation To select text press and hold the left mouse button and drag the mouse. .PP On three\-button mice, click the middle button to paste text in the same or another console. The right button is used to extend the selection, like in .BR xterm (1x). .PP Two\-button mice use the right button to paste text. .PP Double\- and triple\-clicks of the left button select whole words and whole lines. Use of the .B \-p option is recommended for best visual feedback. Which characters constitute part of a \(oqword\(cq can be changed with the .B \-l option. .PP If a trailing space after the visible content of a line is highlighted, and if there is no other text on the remainder of the line, the rest of the line will be selected automatically. If a number of lines are selected, highlighted trailing spaces on each line will be removed from the selection buffer. .PP Any output on the virtual console holding the selection will clear the highlighted selection from the screen, to maintain integrity of the display, although the contents of the selection buffer will be unaffected. .PP The selection mechanism is disabled if the controlling virtual console is placed in graphics mode (for example when running an X server), and is re\-enabled when text mode is resumed \(em but see the \(oqBUGS\(cq section below. .SS Special Commands GPM has the ability to execute special commands via mouse actions under certain circumstances. The special commands default to rebooting and halting the system, but the user can specify his or her personal choice. The ability to invoke commands using only the mouse can be a handy one, because it allows you to shut the system down cleanly when the keyboard is locked and no network is available to restore the system to a sane
Re: problem with left-handed mouse on X - is GPM to blame or is it X?
On Fri, Apr 30, 2004 at 10:14:32AM +0300, Martin-Éric Racine wrote: 1) The information about using GPM's -B option is incorrect. As you can see from the enclosed configs, I already use -B 321, but this is ignored by X. Zeph, can you comment on this? Is -B 321 really broken, or only broken for the raw repeat type? I think I can imagine why it might not work for the raw repeat type. 2) xmodmap has been broken since XFree 4.3.0. Any xmodmap receipe I had left from 4.2.1 either results in broken keyboard maps or in is simply ignored. This is because the XKB configuration data has been overhauled. xmodmap has been deprecated for years. Honnestly, the correct action would be for X to follow the button order configured for GPM. I really wonder why that fails. If you use the raw repeat type, there may be no way it *can* work. Zeph, can you help clear up this issue? -- G. Branden Robinson|It was a typical net.exercise -- a Debian GNU/Linux |screaming mob pounding on a greasy [EMAIL PROTECTED] |spot on the pavement, where used to http://people.debian.org/~branden/ |lie the carcass of a dead horse. signature.asc Description: Digital signature
Re: problem with left-handed mouse on X - is GPM to blame or is it X?
On Mon, May 10, 2004 at 04:10:18PM -0500, Branden Robinson wrote: On Fri, Apr 30, 2004 at 10:14:32AM +0300, Martin-?ric Racine wrote: 1) The information about using GPM's -B option is incorrect. As you can see from the enclosed configs, I already use -B 321, but this is ignored by X. Zeph, can you comment on this? Is -B 321 really broken, or only broken for the raw repeat type? I think I can imagine why it might not work for the raw repeat type. Honnestly, the correct action would be for X to follow the button order configured for GPM. I really wonder why that fails. If you use the raw repeat type, there may be no way it *can* work. Zeph, can you help clear up this issue? Please see the last message, where I covered just this. Raw mode repeats byte for byte, it does not attempt to do ANYTHING to them. In addition, he is repeating a PS2 protocol, with all the problems involved in that. I will say it once more, this is a case of user error. The proper configuration for this is having gpm not repeat, and both X and gpm reading from /dev/input/mice, with X doing some button remapping in whatever manner is now suggested. If 'xmodmap -e pointer = 3 2 1' is no longer the suggested means, what is? -- 1024D/E65A7801 Zephaniah E. Hull [EMAIL PROTECTED] 92ED 94E4 B1E6 3624 226D 5727 4453 008B E65A 7801 CCs of replies from mailing lists are requested. Mercury Knghtbrd: Eww, find a better name, the movie sucked.. G Knghtbrd Mercury: The engine is better than the movie
Re: problem with left-handed mouse on X - is GPM to blame or is it X?
[Zeph Hull dropped from Cc.] On Thu, Apr 29, 2004 at 12:29:35PM +0300, Martin-Éric Racine wrote: Greetings, Given how everyone in this family is left-handed, I recently got around inverting the order of the mouse buttons in GPM; works great in console. Since GPM is setup as a repeater for X, I thought that X would also end up defaulting to inverted button order, but it doesn't. I haven't found any option in X to configure the button order and Googling didn't tell me anything relevant about perhaps selecting a different protocol in GPM to get everything working in X the same way it does on console. Would anybody happen to have a cure for this? There is information in the Debian X FAQ about this. /usr/share/doc/xfree86-common/FAQ.gz http://necrotic.deadbeast.net/xsf/XFree86/trunk/debian/local/FAQ Please let us know if it is insufficient. -- G. Branden Robinson|Religion consists in a set of Debian GNU/Linux |things which the average man thinks [EMAIL PROTECTED] |he believes and wishes he was http://people.debian.org/~branden/ |certain of. -- Mark Twain signature.asc Description: Digital signature
Re: problem with left-handed mouse on X - is GPM to blame or is it X?
On Thu, Apr 29, 2004 at 12:29:35PM +0300, Martin-?ric Racine wrote: Greetings, Given how everyone in this family is left-handed, I recently got around inverting the order of the mouse buttons in GPM; works great in console. Since GPM is setup as a repeater for X, I thought that X would also end up defaulting to inverted button order, but it doesn't. repeat_type=raw type=imps2 Section InputDevice Option Device/dev/gpmdata Option Protocol ImPS/2 EndSection impolite rant removed This is user error. The raw repeat type is something that I am beginning to think is more of a bug then a feature, it does a byte for byte repeat, making no changes to the byte stream. Further more, it seems to be most commonly used with the PS/2 protocols, which, is another rant about two-way communications on a FIFO. Kill the repeat type, have X read from /dev/input/mice as well, and use xmodmap in your ~/.xsession. -- 1024D/E65A7801 Zephaniah E. Hull [EMAIL PROTECTED] 92ED 94E4 B1E6 3624 226D 5727 4453 008B E65A 7801 CCs of replies from mailing lists are requested. VOICE MODE=Pitr So, you are thinking am Communist ? Deal, Comerade ! /VOICE -- Chris on ASR. signature.asc Description: Digital signature
problem with left-handed mouse on X - is GPM to blame or is it X?
Greetings, Given how everyone in this family is left-handed, I recently got around inverting the order of the mouse buttons in GPM; works great in console. Since GPM is setup as a repeater for X, I thought that X would also end up defaulting to inverted button order, but it doesn't. I haven't found any option in X to configure the button order and Googling didn't tell me anything relevant about perhaps selecting a different protocol in GPM to get everything working in X the same way it does on console. Would anybody happen to have a cure for this? Thanks! - # /etc/gpm.conf - configuration file for gpm(1) # device=/dev/input/mice responsiveness=15 repeat_type=raw type=imps2 append=-B 321 sample_rate= - - # /etc/X11/XF86Config-4 Section InputDevice Identifier Configured Mouse Driver mouse Option CorePointer Option Device/dev/gpmdata Option Protocol ImPS/2 Option Emulate3Buttons true Option ZAxisMapping 4 5 EndSection - -- Martin-Éric Racine, ICT Consultant http://www.pp.fishpool.fi/~q-funk/
