Re: FVWM: fvwm3? [on Wayland]
- Are people using FVWM for its looks? (Themability) Looks/functionallity: I want the MWM look/functionallity. Yes, I too like a simple mwm-like look, but even more I like the complete and easy customizability. I think there are proofs on the net of completely different looks. AND functionality ! It's not just a matter of changing a wallpaper. - Are people using FVWM for just being light-weight? YES! I want as lightweight a window manager as posible. I also do appreciate a lightweight wm (it usually runs at 1% CPU), or better I deprecate how heavy are some DE like KDE. - Are people using FVWM for its binding/scripting support? Yes, I added some of my own widgets, see reference below. And I would add a fourth reason ]] - I use FVWM because its customization (even if it has a steep learning curve) is all concentrated in a single text file, so with a minimum of self-documentation one will always know where to go (unless those DE with infinite branches of GUIs) These discussion instigated me to do sometyhing I lept pending for a couple of years, i.e. add some new screenshots to my FVWM doc page referring to my current fvwrc on Ubuntu, which adds to a pluridecennial experience http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html -- Cosi' per li gran savi si confessa / Che la Fenice muore e poi rinasce, / Quando al cinquecentesimo anno appressa. (Dante, Inferno, XXIV, 106-108) Thus the great wise testify / That the Phoenix dies and is born again, / When the five-hundredth year gets close.
Re: FVWM: fvwm3?
On Fri, 2 Feb 2024, Robert Heller wrote: Afterall, no one needs more then one computer... I suppose there is a smiley missing after the sentence :-) My usual way of working (post-COVID, from home) involves usually one or two ssh sessions on two different remote work machines. Quite occasionally I also activate a VNC viewer with a remote session of a VNC server on one of the work machines, and run X stuff there. Occasionally I also run a point-to-point VPN work-home and NFS mount over it, I rsync stuff from work to home, work on it at home, and rsync back when finished. Very rarely I edit work files over remote X, and even more rarely over remote NFS, but that's because I think my connection is low for that. At work I used regularly edit across machines over NFS, ssh and occasionally remote X (but all machines are on a LAN ... some on different VLANs and I have even expect gatewayed scripts to bypass that) Independently of all that, I've never considered switching to wayland, and do not think any colleague does. Our recent standard installation at work is Xubuntu (it used to be OpenSuse), and I had it mimicked on both the home desktop (20.04) and home laptop (22.04) ... first thing I did was to imstall also fvwm, and then run my good old .fvwmrc with the Minimum Necessary Change. I hope to be able to go on with Xorg until I live.
Re: FVWM: emulating mouse button 2 on HP touchpad
On Mon, 15 May 2023, Lucio Chiappetti wrote: I'll be reading the libinput synaptics and synclient manpages. I've done this exercise and some tests ELAN076D: 00 04F3:3244 Mouse id=10 [slave pointer (2)] ELAN076D: 00 04F3:3244 Touchpadid=11 [slave pointer (2)] (not sure why it reports twice the same device as touchpad and mouse, there is no physical mouse attached) I've no idea, I suppose they are both loaded in case a mouse is attached. Synaptics driver and libinput (probably the correct, "modern" option). If I do xinput --list-props for device 11 (touchpad) it returns Synaptics and for device 10 (Mouse) it returns libinput. So I am assuming I have to use synclient to get my status. The current one has just the touchpad, but the lower part of it is "depressable". I believe this is what is technically called a "clickpad". The synaptics man page says that in principle one could set EmulateMidButtonTime to non-zero to do MB2 with simultaneous MB1-MB3 (as I was used on the old touchpad-with-buttons) but ClickPads do not support middle mouse button emulation ... so I gabe up on that. However looking at the synclient output I discovered parameters RTCornerButton = 2 RBCornerButton = 3 LTCornerButton = 0 LBCornerButton = 0 so this mean that TAPPING (not pressing-releasing as for the lower area) in the right-top corner of the touchpad is MB2. Since I found somehmoiw difficult to locate the exact area of such corner, and the tap somehow "hard" I decided to ... ... simply SWAP the actions for MB2 and MB3 with xinput set-button-map 11 1 3 2 4 5 6 7 8 9 10 11 12 This way I can click in the lower right corner to do MB2 (which I use often in any window to do quick paste, and more seldom in the root window to call IconMan), while I tap in the upper right corner to do MB3 (which I use seldom, I can think only of a browser window to get a context dependent menu) I'll keep this along with PointerKey assignment to right-control (which works except in xterms). If I'm satisfied I'll make it permanent. Thanks in oarticular to Mandar Mitra for addressing me to synaptics and xinput. -- Lucio Chiappetti
Re: FVWM: emulating mouse button 2 on HP touchpad
On Mon, 15 May 2023, Mandar Mitra wrote: You mention your laptop is new, so it is likely that the touchpad supports multitouch and other goodies. yes, brand new. So the OS (Xubuntu 22.04) Synaptics driver and libinput (probably the correct, "modern" option). xinput returns among others (when directly on the laptop, if ssh-ing from th3 desktop it returns the desktop stuff, which I suppose is correct) ELAN076D:00 04F3:3244 Mouse id=10 [slave pointer (2)] ELAN076D:00 04F3:3244 Touchpadid=11 [slave pointer (2)] (not sure why it reports twice the same device as touchpad and mouse, there is no physical mouse attached) Synaptics driver and libinput (probably the correct, "modern" option). If I do xinput --list-props for device 11 (touchpad) it returns Synaptics and for device 10 (Mouse) it returns libinput. What does it mean ? Which one is actually used ? It should be possible to configure the touchpad itself to send, for example, MB2 on a two finger tap. (the quite older - and now junked - laptop had two buttons below the touchpad, for MB1 and MB3, and simultaneous press of them got MB2). The current one has just the touchpad, but the lower part of it is "depressable". Depressing and releasing in the lower left part is MB1, in the lower right part is MB3. The exact boundary seems not reproducible unless one uses positions very close to the extrema. The damn thing is far too sensitive ! I'll be reading the libinput synaptics and synclient manpages. Any suggestion welcome
Re: FVWM: emulating mouse button 2 on HP touchpad
On Sun, 14 May 2023, Thomas Adam wrote: On Sun, 14 May 2023 at 22:12, Lucio Chiappetti wrote: So the problem is an incompatibility between xterm and fvwm syntethic events (or a syntax problem with PointerKey ?) No. By default generating synthetic events is considered a security risk. which is why so many applications now no longer support them. For my education ... what are *exactly* synthetic events ? - those generated by a command like xdotool click 2 ? (if executed after a sleep calling it into another window, it works in an xterm) - those generated by fvwm as actions of a Key or PointerKey directive ? I tried (from manpage) Key (xterm) V A C Echo ctrl-V-in-Term and it works and writes the message to .xsession-errors So what could be an "action" to be put at the end of a Key or PointerKey directive to do a X11 native "paste from clipboard" (other that xdotool click 2 which introduces a level of indirection) ? I am pursuing another suggestion (more to come), otherwise as last resort I will be forced to replace (on the laptop) usage of xterm with urxvt (I'll have to look for all my colours and font size settings) ... I do not like the newer console for xfce terminals.
Re: FVWM: emulating mouse button 2 on HP touchpad
On Sat, 13 May 2023, Lucio Chiappetti wrote: In particular (independently of fvwm ?) I am used to do quick copy-and-paste of text selecting it with mouse button 1 (MB1) and pasting with mouse button 2 (MB2). Or shift-MB1 and shift-MB2 in my favourite editor (Hessling ... Editor), due to a peculiarity of the editor. Now I've got a new HP Probook laptop which has a touchpad without buttons. I wonder whether there is a solution (either fvwm-specific or X11-generic) to achieve a decent emulation of the paste function. I have found a *partially* satisafactory solution. But it does no work in xterm. I inserted near the end of my .fvwmrc PointerKey Control_R WR A Exec exec /usr/bin/xdotool click 2 - this way if I press the right-control key while in the root menu, it invokes my FvwmIconMan WinLIst emulation. - If I press the right-control key in the xclipboard, or in a uxrvt or in a browser, this pastes my last MB1 selection. - If I press shift-right-control in my THE editor window it also pastes my last MB1 selection. - But in an xterm nothing happens. The fvwm manpage suggests to place in .Xdefaults XTerm*allowSendEvents: true This seems not to be honoured (as shown by ctl-MB1: allowsends is not set). However most of my xterms are started with class XTermNew. So I placed *allowSendEvents: true in ~/XTermNew (and also in ~/XTerm). This way allowsends appear to be set, but the control-right key does nothing. This has nothing to do with xdotool. If in ANOTHER terminal I do sleep 5; xdotool click 2 and then qjuickly enough move the mouse to an xterm, the MB1 selection *is* pasted. But of course this is a bit clumsy. So the problem is an incompatibility between xterm and fvwm syntethic events (or a syntax problem with PointerKey ?) Is there a solution ?
Re: FVWM: emulating mouse button 2 on HP touchpad
On Sat, 13 May 2023, E Frank Ball wrote: On Sat, May 13, 2023 at 11:44:36PM +0200, Lucio Chiappetti wrote: > In particular (independently of fvwm ?) I am used to do quick > copy-and-paste of text selecting it with mouse button 1 (MB1) and pasting > with mouse button 2 (MB2). Or shift-MB1 and shift-MB2 in my favourite > editor (The Hessling Editor), due to a peculiarity of the editor. > I wonder whether there is a solution (either fvwm-specific or X11-generic) > to achieve a decent emulation of the paste function. to paste. Unfortunately this seems not to work at all on the HP Probook laptop. It does work (though would not be optimal for me) on the desktop, where I don't need it :-) On the laptop is assigned to the f10 key (if used without the fn prefix). Despite the fact xev returns keycode 118 (keysym 0xff63, Insert), when I press the key on the laptop I get a tilde, not paste. (I'd say the soultion, even if working, would be not optimal, since while I could use MB1 to select and MB2 or to paste from a normal (e.g. terminal or browser) window into a normal window, I'd have to use shift-MB2 to paste into a THE (Hessling editor) window ... and how could I do ?) Sorry ...
FVWM: emulating mouse button 2 on HP touchpad
I am since ages a keen fvwm user (2.6.8 bundled with ubuntu), mostly on a desktop PC, with a personal configuration described in http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html. In particular (independently of fvwm ?) I am used to do quick copy-and-paste of text selecting it with mouse button 1 (MB1) and pasting with mouse button 2 (MB2). Or shift-MB1 and shift-MB2 in my favourite editor (The Hessling Editor), due to a peculiarity of the editor. Now I've got a new HP Probook laptop which has a touchpad without buttons. Apparently tapping near the bottom left and bottom right corners of the touchpad works as MB1 and MB3 ... but there is no MB2. I wonder whether there is a solution (either fvwm-specific or X11-generic) to achieve a decent emulation of the paste function. I found a sequence like this xkbset m ; xkbset exp 5 -m ; xmodmap -e "keycode 105 = Pointer_Button2" can temporarily assign the MB2-click (for paste) to my right-control key (I never use such key). The reason it has to be temporary is that xkbset m., enabling the mousekeys, disables the numeric keypad (but I make a lot of use of the numeric keypad - with Num Lock off - for specific functions in my editor configuration ... actually I quite insisted on having an "English International" keyboard to have the num keypad) I also found mention of a command like "xdotool click 2". Would it make sense to assign it with PointerKey in .fvwmrc ? Or are there simpler or better solutions " Thanks -- A middle rank researcher at end career is not rich but is in the top 5% of the Italian income tax taxpayers. Does it not sound strange ?
Re: FVWM: sticky windows on given display
On Wed, 6 Apr 2022, Brian A wrote: There must be a way to specify in .fvwm/config Start Function area the "StaysOnTop" parameter? Or Do I have to setup a separatet Style with Layers? AFAIK, or at least as I use it, StaysOnTop is defined in the Style of the widget. This is an extract of my .fvwm2rc for the relevant widgets Style MyBox Sticky,NoTitle,WindowListSkip,StaysOnTop,HandleWidth 1 Style SubPagerSticky,NoTitle,WindowListSkip,StaysOnTop,HandleWidth 1 Style SubMeterSticky,NoTitle,WindowListSkip,StaysOnTop,HandleWidth 1 Style FvwmPager SkipMapping,!Title, Sticky,WindowListSkip,StaysOnTop Style "ProcMeter V3.5b" SkipMapping,!Title, Sticky,WindowListSkip,StaysOnTop,HandleWidth 1 Style rclockSticky, StaysOnTop, StartIconic,NoIcon,TitleAtBottom Style xdaliclockSticky, WindowListSkip, StaysOnTop, NoTitle Style FvwmBanner!Title, Sticky, StaysOnTop, WindowListSkip
Re: FVWM: FvwmIconMan does not update new windows (?)
On Mon, 3 Jan 2022, João Oliveira wrote: I recently update to Ubuntu 20.04 and fvwm 2.6. I am quite happy but for a problem with FvwmIconMan: it does not update when new windows are opened or closed. When StartFunction is called ... I also had to use FvwmIconMan under Ubuntu 20.04, but I was not using it before on OpenSuse, where I had FvwmTaskBar and FvwmWinList ... so I'm using it (two instances of it) as a replacement for both. I do not fully understand your reference to StartFunction. I do not call FvwmIconMan directly from StartFunction as a "permanently open window". The task bar is called from a container with other items (clock, pager, perfmeter etc.), or better in the container there is a button which toggles the task bar on and off. Default is off ! In this case I verified that a "new window" shows up when the task bar is displayed. A new window in the current desktop. FvwmIconMan itself does not appear UNLESS the desktop has no windows in it. The second instance of FvwmIconMan (replacing FvwmWinList) is invoked as Transient from a mouse button. This shows all windows in all desktops. Being transient I cannot test what happens when a new window is created, because it disappears as soon as I go to a window ... ... no, even here it does show new windows (I tried this with a "(sleep 5; urxvt)" in a window, then calling the winlist from the mouse, and waiting for the rest of the 5 sec ... the new windows shows up). Could supply my .fvwmrc if requested, but beware it's quite complex. Please send any advice to olive...@inf.pucrs.br ! Complying with your request (but why did you post from a different address ?). Best regards LC
Re: FVWM: Move windows to a new desk preserving pages
On Mon, 1 Mar 2021, Ethan Raynor wrote: i quite often find myself wanting to move windows between desks to suit my workflow. what i would like is the ability to move windows as they are on desk X to screen Y without them losing which page they are on. I usually have 7 desks (not that I use them all all the times), and usually each is a single page (I have the possibility to transform each desk in a 1x2, 2x1 or 2x2 arrangement, but I have really never used it). Also I have entries in the window menu to "occupy desk n" (or even "ocpy page n"), as well as other accelerators, but at the end it happens to me that the way I use almost all the times is simply to make a given window sticky or unsticky. Typical use, I keep mail in one desk, the browser in another desk, and work in the other. If I have to paste an URL (or a piece of web page) in a mail or in a text editor winodw, I just make such window sticky, go to the browser desk, do the copy, go back and unstick. Screenshot and .fvwrc samples in http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "Can you see Enrico Fermi punching a time clock? There are effective ways to measure scientific productivity; times clocks are not the way." (Leon M. Lederman to INFN)
Re: FVWM: Draw text in background
On Mon, 21 Sep 2020, Martin wrote: Now I would like to draw text in the upper left corner (output of khal, showing my calendar). ... does not show up in background mode. Not sure what you mean by "background mode", on the root window ? I guess it could be obtained with some Style options. About putting text labels, I have some, which are realized via FvwmButtons. Namely I have one called MyLabel which puts a motto in a random font on the bottom left of each page, and one called MyDesk which puts the desktop name in the bottom right corner (and updates automatically when I switch desk) I have forgottem how I did it, but you can get a feeling of the look in http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html#tre and further down there is a link to the config file which does that http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/fvwmrc.now Feel free to inspire ... but expect no more support :-) Hope it helps -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "Can you see Enrico Fermi punching a time clock? There are effective ways to measure scientific productivity; times clocks are not the way." (Leon M. Lederman to INFN)
Re: FVWM: Calling Old Timers! Fvwm FTP mirrors, anyone?
On Wed, 2 Sep 2020, Thomas Adam wrote: So, dust off your spinning rust disks, your DATs, and carrier pidgeons, and let me know if there's anything out there worthy of salvaging. Sorry not to be able to help, I have been using fvwm since about 2004 (when I moved from Unices to Linux, namely Suse 9.2, and I always used the OS bundled binaries. My sympathy and encouragement for your "enterprise" -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html#uno
Re: FVWM: FVWM on a high density screen -> FvwmForm
On Sat, 15 Aug 2020, Dan Espen wrote: Lucio Chiappetti writes: Yes, it's called Regina, and is required if one wants to use THE (The Hessling Editor) as editor, which is a very good Unix clone of the IBM VM/CMS XEDIT editor. I remember XEDIT, from a shop that was using CMS for a while. I started it up and it looked like hell. Most distracting was a column scale right across the middle of the screen. Fortunately, I didn't run away and hide. (SET) SCALE OFF (better if in one's profile) the scale is generally annoying, but can be useful if one is editing a very wide file and wants to limit visibility and/or editing to columns n-m (SET VERIFY n m) ... what other editors off the possibility to limit editing to columns or sets of lines with or without a given pattern or to select and remove entire columns ? I gave it another try and found all the hooks and toggles to customize it to my liking Yes, it is in my list of favourite easily user-customizable s/w (like THE, fvwm, alpine ...)
Re: FVWM: FVWM on a high density screen -> FvwmForm
On Fri, 14 Aug 2020, Dan Espen wrote: Yep, there is a Unix version of REXX and it's a reasonably good scripting language. Yes, it's called Regina, and is required if one wants to use THE (The Hessling Editor) as editor, which is a very good Unix clone of the IBM VM/CMS XEDIT editor. Back when using IBMs I never got proficient with REXX (I was accustomed to EXEC2, and we abandoned mainframes quite early). Just the bit enough to maintain my THE macros http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/myxedit.html
Re: FVWM: Style ... SkipMapping
On Wed, 20 Nov 2019, Dr Rainer Woitok wrote: Well, in my case it's just another page in the same desktop. I'm not sure whether or not that matters. But above of that it's perhaps time to compare Fvwm and perhaps also Firefox versions. What versions do you use on what distribution? On OpenSuse Leap 42.3, fvwm 2.6.6 Mozilla Firefox 60.0.2 (as I said, I do not routinely use FF, I vastly prefer Pale Moon !)
Re: FVWM: Style ... SkipMapping
On Wed, 20 Nov 2019, Dr Rainer Woitok wrote: No idea why you start an application you always want NOT from InitFunction Simply because I do NOT always want it :-) of course ... your choice. I have many things (most, editors and terminals) I do not start at login (these are essentially just browser, mail, plus fvwm-related things like pager, perfmeter etc. ... there is even a network monitor I start at login but I dismiss soon) I even tried now starting firefox (not my palemoon browser which I left running) with /usr/bin/firefox & and it went straight to the desktop designates for browsers (due to the vestigial style). That's exactly what I'm observing here, too, and what I simply don't want. I want Fvwm NOT to change desktops when I start Firefox during login. Maybe, I just didn't yet get my point accross ... No, sorry, I was not clear. It is the browser which goes straight to the desktop designated, but I remain in the primary desktop ! Which I guess is what you'd like. And what I get. If you start a program using "nohup", this command ... I knew about it but it's a long long time since I felt the need to use it. -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "Can you see Enrico Fermi punching a time clock? There are effective ways to measure scientific productivity; times clocks are not the way." (Leon M. Lederman to INFN)
Re: FVWM: Style ... SkipMapping
On Mon, 18 Nov 2019, Dr Rainer Woitok wrote: Style Firefox StartsOnDesk 1, SkipMapping, NoPPosition + "I" Test (Init) Exec exec /usr/bin/palemoon myhomepage And it works? Well, I have "FixedPPosition" rather than "NoPPosition". Sincerely I do not even remember why I have NoPPosition. I see I have it for all four "browser-type" entries, so it should be something dating a long time ago ... ... but I see I have it also globally in the front section together with this lot (which I can't remember when I inserted, I propagate .fvwrc from upgrade to upgrade) Style * SloppyFocus Style * DecorateTransient Style * SmartPlacement Style * NoPPosition Style * ManualPlacementIgnoresStartsOnPage, ActivePlacementIgnoresStartsOnPage BorderStyle Inactive -- HiddenHandles NoInset Concerning InitFunction vs started from command line I replied in the previous mail -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
Re: FVWM: Style ... SkipMapping
On Mon, 18 Nov 2019, Dr Rainer Woitok wrote: No, as could be clearly seen, it's not in the "InitFunction" but rather in a shell script named "firefox-start-up.sh" which contains the command "nohup firefox > /dev/null 2>&1 &". No idea why you start an application you always want NOT from InitFunction But perhaps that's the cause? Does "SkipMapping" only work for applic- ations started from within the "fvwm" configuration file? But it should not matter. I sometimes have to kill my browser and then I restart it from a terminal, and goes straight to the desktop my .fvmrc (refer to my previous mail) designates for browsers, I even tried now starting firefox (not my palemoon browser which I left running) with /usr/bin/firefox & and it went straight to the desktop designates for browsers (due to the vestigial style). If you are concerned of the runtime messages the browser may issue while running (is that why you redirect stderr ?), also palemoon has that problem when started from a terminal. But I simply kill such terminal to get rid of them ! -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "Can you see Enrico Fermi punching a time clock? There are effective ways to measure scientific productivity; times clocks are not the way." (Leon M. Lederman to INFN)
Re: FVWM: Style ... SkipMapping
On Sat, 16 Nov 2019, Dr Rainer Woitok wrote: During the login process on my laptop I want to fire up Firefox using a page different from the default page where my terminal window is opened. This way I can start entering commands without having to wait for Firefox loading all its tabs. I ceased using Firefox long ago (now using Palemoon) but I guess what I do is so similar for all browsers and was what I did in the past with FF. The only point is do you really mean "page" or "desktop" ? I have 7 desktops each one usually of one page (I can toggle to 2 or 4 pages to extend off screen but I rarely do that). All browsers should go to the second desktop (number 1, named "web") while the first one (number 0, named "mail") is the one where I usually login. I have this (actually also a vestigial FF line) Style "Pale Moon" StartsOnDesk 1, SkipMapping, NoPPosition Style Firefox StartsOnDesk 1, SkipMapping, NoPPosition and in my StartFunction + "I" Test (Init) Exec exec /usr/bin/palemoon myhomepage I have also a (vestigial ?) efent handler which I used to force the browser to a fixed non-default window size, but I guess it does not matter anymore. You could get my present (and some past) .fvwmrc from http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html 3. What happens: Firefox is starting in its own page, but this page is also the one shown on the screen rather than the page containing my terminal wind- ow. I noticed a funny behaviour sporadically when starting libreoffice (with a slow splash screen) and changing desks while it is coming up. libreoffice may come up in the wrong desktop. -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "Can you see Enrico Fermi punching a time clock? There are effective ways to measure scientific productivity; times clocks are not the way." (Leon M. Lederman to INFN)
Re: FVWM: two questions about icons
On Wed, 5 Jun 2019, Ronald F. Guilmette wrote: The default theme for fvwm includes a big box that appears in the lower right hand corner of the screen. Is there a name for that whole thing? I guess it is some instance of a FvwmButtons (there can be more than one, even nested, for instance I have three or five: a control panel, a pager and a performance meter, and a couple of labels ... the control one is always present in the top right corner, a label with the desktop name in the bottom right, and a motto in the bottom left ... the last two items of my control FvwmButtons are two buttons which toggle the visibility of the pager and the performance meter, they appear just below the control panel and partly overlap (usually the pager hides the top items in the performance meter, that's why I can toggle them on/off). I posted an URL with screendumps in an earlier posting. Anyway, that wide but not very tall box includes several different parts: the configuration of the FvwmButtons is a sort of grid, and can contain containers. If you add/replace an item in the FvwmButtons you have to take into account its position and geometry, and coordinate this geometry with the individual tool geometry. For instance my control panel is defined as having 9 rows: row 1 is a container with two buttons side by side (toggle task bar) row 2 contains the current date rows 3 and 4 contains a digital clock (xdaliclock) and this is defined to be double height !! row 5 is a contained with 4 buttons side by side (toggles desktop as 1 page, 1x2 pages, 2x1 pages or 2x2 pages0 row 6 contains indicators whether other machines on my LAN are on or off row 7 is an indicator on packets lost on the wide area network connection rows 8 and 9 are the buttons to toggle pager and performance meter the user *does not* see a system load graph, by default, but *does* see the current date & time in the lower right hand corner of his/her screen, by default. As you see, I agree with the need to have date and time. I never liked xload graphs, but do like to keep info on things like CPU load, network traffic etc. in numeric form, and I use procmeter3 for that (my performance meter FvwmButtons is a list of 17 items with the most useful at bottom ... so they show up even if the top ones are hidden behind the pager. Concerning icons of "closed" applications, I make little use of them, they go on bottom right (sort of), may be hidden by the pager and performance meters (that's why these are toggleable). After all screens have a finite size :-) As you can see fvwm is *completely* customizable (alas, at the expense of a learning curve, but fully worthwhile for me) -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "All that is google does not glitter Nor all who use alpine/procmail are lost"
Re: FVWM: FvwmForums: Change of engine and help with theming
On Wed, 5 Jun 2019, Thomas Adam wrote: On Tue, Jun 04, 2019 at 07:46:02PM -0400, Dan Espen wrote: Sorry, no inclination to follow a forum. Wish we'd go back to exclusively email. I personally don't disagree -- but you know what the old maxim says: "Give the people what they want". It seems the y00fh of today prefer forums. God knows why, but they do. Think that an usenet (nntp) newsgroup would give both the benefits of e-mail and forums (in terms of asynchronous access, and of each user choosing the interface one prefers ... which could mimic e-mail in full). I follow/ed several usenet newsgroup (usually once per day, looking only at new posts), several mailing lists (usually in MIME digest mode, one digest per day ... the fvwm list is the only one not offering digest mode) and three forums only, all phpBB (one thing I hate of forums is that each has its own interface ... "you are lost in a twisting maze of little forums all different"). We could shunt the wiki over to Github as well I'd never found Github good for "support" in the way mailing lists and forums do (at least for the packages for which I had to go to Github). It is a repository, or perhaps a collaborative tool for developers, but I found no satisfactory interactivity for questions users may address to developers. -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "All that is google does not glitter Nor all who use alpine/procmail are lost"
Re: FVWM: two questions about icons
On Tue, 4 Jun 2019, Ronald F. Guilmette wrote: Grrr. I can't imagine why this would be so hard. I didn't think what I wanted to do would be so complex. Again, I just want a digital xclock to appear in the blank space below the xbiff thingy in the default theme. AFAIK there is no such thing as a "default theme" (valid for any fvwm installation anywhere). There could be a "system default" bundled in your OS distro (different for each distro). You have to locate that, copy it to your ~/.fvwm/.fvwm2rc (so that all the system stuff is now "yours") and try to play with that adding (patiently) one thing at a time.
Re: FVWM: appearance
On Wed, 2 Jan 2019, Martin Cermak wrote: I don't think there is any difference in how fvwm looks in your screenshots. What differs is the appearance of your File Manager. I tend to agree. In fact I was puzzled by your screen dumps, since they contained only part of the screen with a specific (non-fvwm) application (a file manager which as such is unknown to me). The only things which are responsibility of the window manager (fvwm) are the windows decors (title bar and borders) which look pretty standard (one is light gray and the other is almost black, but this may be because one of the windows was focused when taking the dump). If you are interested in fvwm customization, you should read the man pages of fvwm (long and instructive but not easy) and of your file manager whatever it is called, and find out WHERE their configuration file is stored. For fvwm, the usual personal configuration file is ~/.fvwm/.fvwm2rc but if you do not find one for you, it may be using some "system default". On my opensuse this should be /etc/X11/fvwm2/system.fvwm2rc, which I believe is what one sees for a fresh new account. fvwm is highly customizable but luckily all configuration can reside in a SINGLE file (you can get some ideas of my experience with fvwm in http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html) Concerning other applications, I cannot predict. If it were a good-behaved X11 application (which nowadays somebody may consider old fashioned), its configuration will be controlled by a number of resources with syntaxes like Rxvt*cursorColor: yellow the.normalFont: -*-Terminal-Bold-R-Narrow--*-140-*-*-C-*-ISO8859-1 the.colorCyan: #00fafa *VT100*wideFont1: -misc-fixed-medium-r-normal-*-18-120-100-100-c-180-iso10646-1 *ScrollBar:on These resources can reside in a file ~/.Xdefaults common to all applications (in this case the resource is usually prefixed with the capitalized name of the application) or in an application specific file (whose name depends on the application but usually is either a dot-file with the name of the application, e.g. ~/.rclock for rclock, or the capitalized name of the application, e.g. ~/XTerm for xterm). It is also possible that there is no user file and it uses some system configuration in some system place (application dependent, see man page). The user file can be created only with the items which you want to override. It is also possible to have alternate user configuration files, and "pass" them to a particular instance of the application. This might in general occur in two ways: - one is to pass the name of the file as an argument to the application in a way which is application dependent, e.g. xterm -class XTermLin will read from the file XTermLin - another one is to use the general resource manager to read the specific user file before invoking the application, e.g. edit:aliased to xrdb -merge ~/edit0.x ; xthe !* & edit1: aliased to xrdb -merge ~/edit1.x ; xthe !* & edit or edit1 invoke the same editor but they pre-load a different resource file (in this case with different background colours etc.) If the application is not well-behaved (somebody will say modern), it might use some more complex hierarchy of files, and maybe the files are not just plain, easily editable, text files with "resource: value" lines, but some more complex things (maybe xml or json or whatever else). They will be probably be some dot-files or dot-directories under your home, like .macromedia .mozilla or whatever. If you start with a fresh account and, after running an application for the first time, you see a new dot file appearing, that will be it. -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "All that is google does not glitter Nor all who use alpine/procmail are lost"
Re: FVWM: bookmark/unbookmark multiple windows
On Wed, 19 Dec 2018, Peter Holm wrote: To select a window for moving there is afaik two methods. Either you make the window sticky, and then unstick it - or you bookmark the window. Do you mean moving across desks/pages ? You can stick/unstick or define a menu button which moves the window to a selectd desk/page (or drag and drop across pages of same desk). Although I have extensive MoveToDesk MoveToPage menus, it ends that most of the cases I use the stick/unstick method. -- Lucio Chiappetti - INAF/IASF - via Corti 12 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html "All that is google does not glitter Nor all who use alpine/procmail are lost"
Re: FVWM: [Draft] New Configuration Format
On Fri, 23 Sep 2016, Thomas Adam wrote: On Fri, Sep 23, 2016 at 12:42:14PM +0200, Lucio Chiappetti wrote: is <<< a perlism, or a typo for more customary << ? In shell, <<< is a here-string. I wasn't aware of the distinction between here-documents and here-strings (I had to check https://en.wikipedia.org/wiki/Here_document), I've always used only the former. Does this apply to ANY occurrences which in your new scheme will use the backslash like the old AddToFunc followed by lots of + I lines ? Yes. For me looks ok -- -------- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
Re: FVWM: [Draft] New Configuration Format
On Fri, 23 Sep 2016, Thomas Adam wrote: On Wed, Sep 21, 2016 at 09:26:08AM -0400, gi1242+f...@gmail.com wrote: I use FvwmPerl quite a bit ... Never used FvwmPerl nor perl, just a couple of FvwmScript, but I know here documents from shell scripting where I use them a lot. + I SendToModule perlwops eval <<< "EOF" blah EOF is <<< a perlism, or a typo for more customary << ? instead without having to worry about the backslashes. I think this is reasonable; I'll add it to the document. Thanks. Does this apply to ANY occurrences which in your new scheme will use the backslash like the old AddToFunc followed by lots of + I lines ? That would be rather nice, better than the backslashes. Thanks -- -------- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
Re: FVWM: [Draft] New Configuration Format
On Wed, 21 Sep 2016, Thomas Adam wrote: Secondly, take another look at the document, since it tells you how functions could be specified. "the document", if I'm reading the right one, is just a very short sketch (3-4 pages) with some examples ... compared to the much longer man pages I studied carefully long time ago (not necessarily understanding all, specially what I was not needing at the time, and surely not remembering it all :-) ) The point is that most users (I guess, that's valid for me anyhow) would only write a full configuration file once or twice in a lifetime, and edit it less than once per year. I am also possibly even more extremist than other users, as I won't update fvwm version until I update my OS, and won't update my OS until I change my hardware (but at that moment I'M DAMN SURE I WANT TO GO ON USING FVWM). So possibly the "change in config syntax" is just "another nuisance" of the sort I will be confronting when updating version (i.e. very seldom), and I suppose I am ready to consider it a "physiological" nuisance. I am looking at my .fvwmrc (refurbished in 2015, using fvwm 2.5.26 under openSUSE 11.3), the one referred to near the end of http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html, and I see I use the following sort of items - Desktop characteristics (name and size) - general characteristics inherite from something (I do understand ColormapFocus FollowsFocus or ImagePath and forgot why the rest) - Color sets (and FvwmBacker) - general Styles (Style *) - Styles for my modules and applications - some event handlers (DestroyModuleConfig and recreation) and associated functions - the big thing, the StartFunction which starsts applications and modules with some associated functions - my window menus with some associated functions - my root menu and some other menus (including some inherited stuff like for FvwmWinList - the definition of a button box and pager, of a task bar etc, inclusive of some usage of FvwmCommand, SendToModule, FvwmScript - the key and mouse bindings and some accelerators Now for some of those items (typically those fitting on one line) I understand the matter is a change of syntax of the arguments. What I consider a "physiological nuisance". It is less clear to me what I should do to update things which are multiline like AddToFunc +I ... or ModuleConfig or FVwmScripts or event handlers (the ones I understand less). But I guess I will confront with them in due time. Next hardware change if I wait long enough that "new fvwm" is ready by then, or even second next if I change earlier :-) -- -------- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
Re: FVWM: Is there a way to ignore other windows when placing one?
On Fri, 24 Jun 2016, lee wrote: Lucio Chiappetti <lu...@lambrate.inaf.it> writes: Is there a difference between having N desktops and having N pages which matters for practical use? well ... a desk has a name/title, a page has not ... and one uses GoToDesk and GotoPage ... do not think there are other practical differences in the way I use them ... ... or perhaps, one can think of a desk with no page as something of the size of your screen, stacked behind your screen along z. While pages can be thought of extensions off screen along x and y (right/left, top/bottom), could be useful in rare cases one wants to extend one window off screen (one can even scroll fraction of pages if I remember). I have the pager always covered up by some window. It just takes away so much screen space that it's only visible on empty, or almost empty, desks. So you actually see the root window :) Yes, I almost never use applications in full screen (perhaps an acrobat slide show), even the browser keeps to a sort of squarish shape leaving free an area on the right where the pager is. My screen is 1920x1080 and the browser is some sort of 1600x1024. So I usually don't see the root window or a pager window but the windows I'm actually using. What if you basically never saw the pager? Would you get confused? I guess so, but I keep the pager Sticky, StaysOnTop, so other windows never obscure it, I have such a family of widgets in the top right corner of my screen. From top I have: - a couple of buttons to call on and off a taskbar (see below) - calendar date and xdaliclock - buttons to change desk to 1, 2x1 1x2 and 2x2 pages - status indicator of main hosts on my LAN - button to call pager on/off - button to call procmeter on/off - pager - procmeter usually the pager obscures the procmeter button and the top of the procmeter (temperature, processes, load, paging) so I just see things like mail in inbox, ethernet traffic, CPU % below the pager. But I can withdraw them (do it rarely, sometimes I iconify a number of terminals and edits, and the icons (anchored bottom right) hide under the procmeter or pager). By habit I sometimes/often iconify stuff, but almost never shade stuff (shade is however a NICE fvwm feature, a shaded window is reduced to its title bar) In the rare case I forgot what is where (e.g. for the xclipboard, or because the window is hidden), I use MB2 bound to FvwmWinList, which lists all my windows by title, so I can switch there easily, That's like a last resort for me. The list can get so long that it's not easy to find what I'm looking for in it. My list is not terribly long. Usually I know where is what by habit, excepts for things like xclipboard, of which there can be a single instance. That's the thing I usually look via FvwmWinList. There can be windows hidden behind other (like a small popup from an alarm). For these I use a taskbar I can call on demand, which shows the windows in the current desktop by title. -- ---- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
Re: FVWM: Is there a way to ignore other windows when placing one?
On Mon, 20 Jun 2016, lee wrote: Interesting :) It seems very confusing, though. How do you remember what is where? I have a plain setup with just 6x6 pages and many times have to flip between them to find the particular page I want to go to. Well, after a long time working with 2 desktops each 2x2 pages, a while ago I junked the idea of separate desktops and pages. Now I have only 7 desktops each of 1 page (well, I have buttons to switch them to 2x1 1x2 and 2x2 but I'd never used them). I have the pager almost invariably exposed, and know which desk I am in because it is highlighted in the pager in a different colour. Three of the desks are named Mail Web and Net, the other are ABCD. I know in which desk I am also by a label in the bottom right corner, and by the background colour of the root window. Although I have root menu entries and keyboard accelerators to switch desktops, I almost invariably do it clicking on the pager, To know which application is in which desk, I usually rely on habits (mail is in Mail, browser is in Web, virtual machines if any are in Net etc.) and on the miniicons in the pager, In the rare case I forgot what is where (e.g. for the xclipboard, or because the window is hidden), I use MB2 bound to FvwmWinList, which lists all my windows by title, so I can switch there easily, I do use a lot the "sticky" mode of the window menu, to move a window across desktops (I make it sticky so it appears on all, switch to the final desktop, and unstick it), however I have also functions in the window menu to move the window to a specific desktop and I use them sometimes. You can get a flavour of the old and new approaches (and the configs) in http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/window.html -- -------- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
Re: FVWM: Survey: Are you affected by disappearing Firefox menus?
On Wed, 30 Mar 2016, Michael Großer wrote: I have a problem with newer versions of Firefox (and Thunderbird), so after several months (probably years) of suffering, I finally ... I am not sure if your question really concerns FVWM ... anyhow it is one and half year I abandoned Firefox in favour of Pale Moon (see my signature) which is how Firefox used to be, and more, and has a very responsing forum and supporting community ! -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not like Firefox >=29 ? Get Pale Moon ! http://www.palemoon.org
FVWM: interference firefox/acroread and fvwm on suse 11.3
I am noting a funny (and sort of annoying behaviour) when using firefox and acrobat reader in a fvmw session under suse 11.3 (newly installed). The topmost menu bar (File Edit ...) has white text onto black background, although I do not have any X resource in .Xdefaults which makes that. The same colour applies to all pull down menus, and for acrobat also to the bookmark window. The remaining menu bars and widget are normal (black on light grey). Actually the problem occurs both for a test user which imported all my old personal configuration (.Xdefaults, .mozilla/* etc. .fvmw) as well for a test user created afresh which has none of those files (and uses for fvwm the system-bundled fvwmrc ... actually it is using the one recently modified by Thomas Adams as a kind reply to a recent post of mine). The problem does not occur if firefox or acrobat are run from KDE. So this makes me think of some sort of interference, like firefox and acrobat are picking up those resources from somewhere in the fvwm configuration, or not finding something they expect. I would be glad to override this with X resources, but I'm not sure firefox and acrobat honour them in the usual way. I tried editres to get the widget tree and it returns nothing. Any idea of the reason, and a possible workaround within .fvwm2rc ? -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Italian Research at risk.La Ricerca italiana a rischio ! see http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/nobrain.html cfr.
FVWM: partial update Re: interference firefox/acroread and fvwm on suse 11.3
On Fri, 17 Dec 2010, Lucio Chiappetti wrote: I am noting a funny (and sort of annoying behaviour) when using firefox and acrobat reader in a fvmw session under suse 11.3 (newly installed). Apparently also openoffice is showing the same behaviour. The topmost menu bar (File Edit ...) has white text onto black background, The remaining menu bars and widget are normal (black on light grey). For firefox there is a partial workaround. I always hated themes but I found a sort of lightweight themes called personas. If one goes to the site where one can load them, one can also test them. They appear to replace the menu bar and other bars with some pictured background. My workaround was to choose a persona called simple fox (useless to say, no picture, just solid gray :-)). This makes me think that all those applications are NOT using a specific fg/bg colour for the top menu bar, but leaving it somehow transparent letting some w.m. default shining through. I suppose in KDE what shines through is determined by the default KDE theme ... how could I set a similar super-default in fvwm ? a plain *background: and *foreground: in .Xdefaults is no good. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Italian Research at risk.La Ricerca italiana a rischio ! see http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/nobrain.html cfr.
Re: FVWM: partial update Re: interference firefox/acroread and fvwm on suse 11.3
On Fri, 17 Dec 2010, Robert Heller wrote: The topmost menu bar (File Edit ...) has white text onto black background, The remaining menu bars and widget are normal For firefox there is a partial workaround. [persona simple fox] To make FireFox (and other GNome-flavored applications) 'play nice' with FVWM, I fire up a minimal bit of GNome: /usr/libexec/gnome-settings-daemon. By firing this up with a properly setup ~/.gconfmumble directory/files, I can get FireFox to be set up with the proper Gnome 'theme' (I created a minimual theme that suited). So you are confirming there are applications which are ill-behaved with fvwm ? I hope this info makes sense with Thomas Adam. I am REALLY reluctant to have to learn gnome just to bypass its features (if I understand correctly). Actually I'm not even sure it is installed (our institute-wide base pre-installation prepared by our sys man replied KDE when offered the choice KDE vs gnome) ... and under KDE those applications are well-behaved. Would you mind to share this minimal theme ? Or will Thomas Adam have a simpler workaround at .fvwmrc level if this diagnosis makes sense to him ? -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Italian Research at risk.La Ricerca italiana a rischio ! see http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/nobrain.html cfr.
Re: FVWM: partial update Re: interference firefox/acroread and fvwm on suse 11.3
On Fri, 17 Dec 2010, Thomas Adam wrote: Thomas Adam says it's still not an FVWM problem. :) Of course I did not intend to blame FVWM ! Actually I hoped it had a quick-and-more-or-less-dirty workaround ! (after all openoffice, acroread and firefox are not unusual applications ... I remember there were some adhoc style in the 9.2 .fvwmrc) gtk-chtheme to set your .gtkrc stuff up. Although anything more complicated will require you to edit it by hand anyway. Thats arabic to me. I do not think those stuff is even installed (I'll check on Monday), surely whereis does not find it. Since all the matter seems to force some sort of default bg/fg, I'd be glad on editing once forever a fixed file. I tried importing in my user home the .gnome* .gt* and .gconf* stuff which user root (which runs under KDE) has, but they have no effect. (after spending two hours discovering the mysql init.d script included in the distribution has a bug, which does exactly the opposite of what is written in the comments ... and which is sensible to do, and was written in my old notes, grr !) Nice weekend to everybody ! (possibly without snow) -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Italian Research at risk.La Ricerca italiana a rischio ! see http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/nobrain.html cfr.
FVWM: fvwm on suse 11.3
I've just switched on a new machine with suse 11.3. Apparently it has bundled fvwm 2.5.26. I'm getting some errors using my old .fvwm2rc (happily running on suse 9.2 with fvwm 2.5.10). E.g. no pager, no desktops etc. I suspect some paths have changed. Before I look into it directly, I created a fresh dummy user, and started a fvwm session from scratch. That comes out with some default theme, but does not create any .fvwm2rc (or config, as I realize it should now be called based on the man page). So I suspect it uses some system default. It would be a good idea for me to look at that to merge in useful stuff. But where is it based ? The man page mentions 2+5 possible locations, in ~, /usr/local/share and /etc. None of which exists. I have however found one in /etc/X11/fvwn2/system.fvwmrc (which is even signed Thomas Adams). Is it the correct one to look at ? Thanks -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Italian Research at risk.La Ricerca italiana a rischio ! see http://sax.iasf-milano.inaf.it/~lucio/WWW/Opinions/nobrain.html cfr.
Re: FVWM: How to find an error in a VERY HUGE config
On Sat, 21 Nov 2009, Michelle Konzack wrote: My config has over 2000 lines and I am searchin this error since several Just a silly suggestion. Why don't you use a sort-of- binary search ? Comment out one half (say the bottom half) of your config file and see if the error shows up. If yes, uncomment one half of the commented section (say the top half) and repeat If no, uncomment the entire bottom half, and comment out one half of the top half. And so on, until you located it. For 2048 lines you should locate it in at most 11 attempts. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) Citizens entrusted of public functions have the duty to accomplish them with discipline and honour [Art. 54 Constitution of the Italian Republic] For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
mailing list archives Re: FVWM: how to start iconified under mouse position
On Thu, 10 Sep 2009, Thomas Adam wrote: 2009/9/10 Lucio Chiappetti lu...@lambrate.inaf.it: ... mailing list archives, but it looks like all this thread has not been archived ! http://www.mail-archive.com/fvwm@fvwm.org/msg00877.html Thanks for the CORRECT pointer ... ... but why the mailing list page under fvwm.org points to http://www.fvwm.org/contact/, which in turn points to http://www.mail-archive.com/f...@lists.math.uh.edu/ ? I looked there !
Re: FVWM: how to start iconified under mouse position
On Mon, 7 Sep 2009, Thomas Adam wrote: 2009/9/7 Lucio Chiappetti lu...@lambrate.inaf.it: What I'd want to do is to redefine the F4 F5 F6 so that instead of issuing a Style command to a pre-existing Laser window, they create one, and use the rePointLaser function/event passing to it a different shift size (actually a couple of xsize,ysize). If F4, F5 and F6 have the predefined sizes already... Well ... the size of the icon is given by the png file associated to the style ... Key F4 PipeRead `echo SetEnv xsize 2 echo SetEnv ysize 4 start_application_name -options` I was baffled by the fact you used backquotes `` instead of ''. The former freeze everything. I am not fully there yet. What I wanted were two triplets of commands. Shift-F1/F2/F3 should CREATE a new pseudocursor of a given shape. Unshifted F1/F2/F3 should CHANGE the shape of the pseudocursor under the mouse. I thought to define a base style, three variants (LaserA/B/C) associated to the three icons, and a current/default style Style LaserBase StickyIcon, StickyAcrossDesks, Layer 9, !IconTitle, IconBox none, PositionPlacement UnderMouse Style LaserA Icon beam.png , UseStyle LaserBase Style LaserB Icon circle.png, UseStyle LaserBase Style LaserC Icon ellipse.png , UseStyle LaserBase Style Laser UseStyle LaserA The creation function will imply the initial style via the -name DestroyFunc PointLaser AddToFunc PointLaser + I Exec xterm -name Laser$0 -iconic I originally defined shift-F4/F5/F6 to invoke PointLaser B/C/A. It did work but ... see below. Now I use PointLaser (which means only style Laser) and pass the icon name as another environment variable PointerKey F4 A S PipeRead 'echo SetEnv xof 50 echo SetEnv yof 50 echo SetEnv ik circle echo PointLaser ' PointerKey F5 A S PipeRead 'echo SetEnv xof 100 echo SetEnv yof 50 echo SetEnv ik ellipse echo PointLaser ' PointerKey F6 A S PipeRead 'echo SetEnv xof 40 echo SetEnv yof 40 echo SetEnv ik beam echo PointLaser ' (it's a bit verbose, I hoped I could call something like MyF 50 50 circle and move all the PipeRead stuff in the function MyF, but I'm not too versed with PipeRead (euphemism, it's the first time I use it, though the use of transient environment variables is not unlike some techniques of shell programming I used). This way the event handler is like this DestroyFunc rePointLaser AddToFunc rePointLaser + I ThisWindow (Laser*) Move m-$[xof]p m-$[yof]p + I ThisWindow (Laser*) Style Laser Icon $[ik].png + I UnsetEnv xof + I UnsetEnv yof + I UnsetEnv ik DestroyModuleConfig Event-PL: * *Event-PL: Cmd Function *Event-PL: add_window rePointLaser The point is that the unshifted F4/F5/F6 change SHAPE to all existing pseudocursors (if more than one exist). Not surprising since now one common style Laser is used. Actually even the creation shift-F4/5/6 have the side effect that all pre-existing pseudocursors are changed to the last shape invoked. PointerKey (Laser*) F4 I N Style Laser Icon circle.png PointerKey (Laser*) F5 I N Style Laser Icon ellipse.png PointerKey (Laser*) F6 I N Style Laser Icon beam.png What I'd ideally need is a way to change the icon per window from the outside (having styles named Lasernnn with a sequential number ?) PointerKey (Laser*) F4 I N Style of the current window only Icon circle.png For the time being I'd live with what I have, since the real videoconf is coming soon, and I won't plan to have more than one pseudocursor at a time. But I'd like a smarter way to do this. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
Re: FVWM: how to start iconified under mouse position
On Tue, 8 Sep 2009, Thomas Adam wrote: + I ThisWindow (Laser*) Style Laser Icon $[ik].png No -- use WindowStyle here. + I ThisWindow (Laser*) WindowStyle Icon $[ik].png THANKS A LOT WindowStyle does exactly what I needed. Now it all works as wished. The solution was easy and fast enough to be tested and implemented soon. I will write a final report for the list in the next days AFTER my videoconf. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
FVWM: how to start iconified under mouse position
I am slowly progressing towards what I described in the thread Re: FVWM: again (semi-transparent) pseudo-cursors My purpose is to [make a presentation] in a videoconf [...], and being able to highlight parts of it moving the mouse (like a laser pointer in a local presentation). I can start my pseudo-pointer from the root menu as + LaserPointer Exec xterm -name Laser -iconic I can assign an icon (a png which is in my ImagePath ...) via the associated style, and change it via function keys (provided I point the current icon with the mouse). Currently I use a laser beam icon, a circle and an oval icon with the inside transparent. At creation they appear in the bottom right corner of the screen. Style Laser Icon beam.png , StickyIcon, StickyAcrossDesks, Layer 9, !IconTitle, PointerKey (Laser) F4 I N Style Laser Icon circle.png PointerKey (Laser) F5 I N Style Laser Icon ellipse.png PointerKey (Laser) F6 I N Style Laser Icon beam.png I can also switch off my pseudo-pointer with a function key PointerKey (Laser) F3 I N Close I can move it around with a function key or control-MB1 PointerKey (Laser) F1 I N Move Mouse (Laser) 1 I C Move (of course this has to be terminated by a mouse click) And I can park it in the bottom-right corner with a function key PointerKey (Laser) F2 I N Move -0 -0 *** now, what am I missing ? *** I'd like to find a way to create the icon straight where the mouse is currently pointed (to avoid having to retrieve it from the parking position). I thought I could do that e.g. with a shift-MB1 (currently I'm testing using a key on root window) provided I define a PROPER function Mouse 1 A S PointLaser Key A R A PointLaser I thought the function could be something like this but the part which should move the icon has no effect. The icon is created in the default bottom-right corner and not under the mouse pointer. DestroyFunc PointLaser AddToFunc PointLaser + I Exec xterm -name Laser -iconic + M Next (Laser) Move m+0 m+0 I tried adding this to the Laser style, to counteract something which could be in my configuration IconBox none, PositionPlacement UnderMouse but the effect is that the icon is created in some (reproducible) random place on the screen. I am not sure whether the Move command applies to the icon (or only when not iconified), nor that my syntax is correct. Nor I am sure of possible interferences with the rest of my .fvwmrc. I am reporting below some lines of .fvwmrc which might be relevant and perhaps require overriding. What should I do to have an application start iconified with the icon under the cursor ? Style * IconBox 1000x70-1-1 ColormapFocus FollowsMouse IgnoreModifiers L2 OpaqueMoveSize 100 Style * SloppyFocus Style * IconBox -20 0 -1 -1 , IconFill bottom right Style * SmartPlacement Style * NoPPosition Style * ManualPlacementIgnoresStartsOnPage, ActivePlacementIgnoresStartsOnPage -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
Re: FVWM: how to start iconified under mouse position
I made one step forward using FvwmEvent, but I'm now stuck about passing arguments to it. On Mon, 7 Sep 2009, Lucio Chiappetti wrote: *** now, what am I missing ? *** I'd like to find a way to create the icon straight where the mouse is currently pointed (to avoid having to retrieve it from the parking position). I thought I could do that e.g. with a shift-MB1 (currently I'm testing using a key on root window) provided I define a PROPER function Mouse 1 A S PointLaser DestroyFunc PointLaser AddToFunc PointLaser + I Exec xterm -name Laser -iconic I truncated this function here. I wrote the move part in another function DestroyFunc rePointLaser AddToFunc rePointLaser + I ThisWindow (Laser) Move m-40p m-40p which I bound (is this the right expression) to FvwmEvent in the staruto + I Module FvwmEvent Event-PL *Event-PL: Cmd Function *Event-PL: PassId *Event-PL: add_window rePointLaser This way my icon is positioned with its hotpoint at (40,40) from the mouse. For the beam icon (80x80) this means it's centered at the mouse. Now as you remember, I was using three different icons (beam, circle, and ellipse), which were swapped with F4 F5 F6. These icons have different sizes. What I'd want to do is to redefine the F4 F5 F6 so that instead of issuing a Style command to a pre-existing Laser window, they create one, and use the rePointLaser function/event passing to it a different shift size (actually a couple of xsize,ysize). But how to do that is unclear to me ... -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
Re: FVWM: again (semi-transparent) pseudo-cursors
On Fri, 4 Sep 2009, Thomas Adam wrote: 2009/9/4 Lucio Chiappetti lu...@lambrate.inaf.it: - associate to an application (which could be a do-nothing application e.g. an xterm with a special title never opened, always iconized) an icon which is partially transparent (say a large red ellipse with the interior transparent) No -- applications define for themselves the icon to use. This is sometimes changeable via ~/.Xdefaults by perhaps issuing a new PCF to use. Depends on the application though. Hmm ... a bit surprised by this reply. I have in my .fvwmrc a few Style commands which bind an xpm file to an application or instance, e.g. Style Pine Icon mail1.xpm where in this case Pine is an urxvt started as Test (Init) Exec urxvt -name Pine ... so I suppose I'm able to define a dummy application (I thought of some terminal window with a specific name which I will never open) and associate an icon to it. I was more concerned with : which sort of file should I use as icon in order for it to be partially transparent ? The only trasparent things I used (and I guess I'll have to look back in my notes) are transparent GIFs. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Wir muessen wissen / wir werden wissen Inscription on David Hilbert's grave http://tinyurl.com/3e5hfl
FVWM: [OT?] using part of the screen as full screen ?
This is not strictly a fvwm question, but maybe there is a fvwm solution to it, or I can be directed to a forum where this is on topic. I'd like to use only PART of my X server screen as it were the full screen. I'm aware of an utility like xrandr, but this changes temporarily the resolution in a fake way using still the entire screen. The background of my question is as follows. I have available a laptop (not the machine where I usually work) which I intend to connect to a videoconference device. Then I want to show an acroread PDF presentation and a firefox demo, ideally in full screen mode. On the laptop I run Suse Linux and use fvwm. The point is that some part of the PC screen does not appear on the videoconference TV set (and is NOT trasmitted outside as well !). Say in the most favourable hypothesis that the entire vertical span of the PC scren is replicated on the TV and transmitted, while some 200 pixels on the right fall off screen of the TV. I've spent one afternoon testing all possible sensible combinations of laptop screen sizes (Linux SAX2 and even Windows) with the settings of the videoconference device. Then I spent 4 more hours in a videoconference with the manufacturers of the device. The conclusions were that : - if the laptop is connected directly to the TV, there is no truncation (the XGA signal from PC is OK with the TV) - if a remote site sends it PC screen to the videoconference device it is not truncated (the remote videoconference device sends correct XGA) - there is some mungling which our videoconference device does to the XGA stream from the PC before dispatching it (to the TV and outside) - they seem to suggest withdrawing the device to factory for checks and reconfiguration. These times are not consistent with my needs (a demo scheduled in a couple of weeks). I have of course a possible workaround which is : - arrange a test with the real site where the demo shall be seen - manually resize the acrobat and firefox windows so that they occupy a fraction of the laptop screen ... and this fraction is exactly the part replicated to the videoconference ... but then there will still be needed adjustment to remove the window manager decoration and various service buttons which are usually absent in full screen mode. What I'd like to do is to wrap my X-session in one application which could act as an X server with a reduced screen size. Does such wrapper exist (ideally within fvwm or at least within Suse) ? Any hints will be much appreciated -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Wir muessen wissen / wir werden wissen Inscription on David Hilbert's grave http://tinyurl.com/3e5hfl
Re: FVWM: fvwm, xterm and unicode UTF-8
On Wed, 6 May 2009, Thomas Adam wrote: 2009/5/6 Lucio Chiappetti lu...@lambrate.inaf.it: I have no idea whether fvwm or whatever is the lead X process in the complex SuSe X login do execute .login. Nothing in FVWM does -- and no shells do by default. Sort this out by looking at what your shell is doing when it's invoking a login shell. A long time ago (soon after I moved to SuSE when I was choosing the window manager), I tried to trace all the steps of an X login, including the execution of sytemwide /etc/csh.cshrc /etc/csh.login and user .cshrc and .login. I still have some obscure notes about it. I seem to remember there are transient processes involved. This seems indicated also by the gap in the pid's in the following excerpt of a tree-like ps. PPID PID USER COMMAND 1 4971 root /opt/kde3/bin/kdm 4971 4989 root /usr/X11R6/bin/X -nolisten tcp . 4971 4990 root -:0 4990 6303 lucio /usr/bin/X11/fvwm 6303 469 lucioxterm 469 471 lucio -csh You should realize that an user of fvwm and csh like me is sort of an anomaly for SuSE who expects KDE and bash :-) I believe there is some transient bash involved in the X login. Anyhow the main question was whether an xterm started via Exec xterm from a .fvwm2rc menu should behave differently than an xterm started from a shell. What happened was that the former has correct unicode support, while the latter requires an explicit -u8. Now I've arranged everything, and I'm inclined to if ain't broken don't fix it. But I was just curious if the correct unicode support in the menu Exec xterm was an fvwm bonus, or just a coincidence of some local setup. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- Nature on government cuts to research http://snipurl.com/4erid Nature e i tagli del governo alla ricerca http://snipurl.com/4erko
Re: FVWM: fvwm, xterm and unicode UTF-8
On Tue, 5 May 2009, Bill Marcum wrote: A) local xterm started from menu via a plain Exec xterm this invokes a very simple resource file ~/XTerm (attached) the resulting shell has NO LC_CTYPE variable set the resulting xterm has UTF mode on though dimmed (control-MB3) UTF support works perfectly Is LANG set? What is the output of the locale command? LANG is not set in such xterm, and locale returns all POSIX. However everything works perfectly !! However I can see that my .login contains a setenv LC_CTYPE en_US.UTF-8 (this was a tentative I did a while ago). It should not matter since individual shells/ xterms should NOT execute .login (and in fact LC_CTYPE is not set). I have no idea whether fvwm or whatever is the lead X process in the complex SuSe X login do execute .login. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- Nature on government cuts to research http://snipurl.com/4erid Nature e i tagli del governo alla ricerca http://snipurl.com/4erko
FVWM: fvwm, xterm and unicode UTF-8
My apologies if this is (partially ?) off-topic for this list. I was trying to sort out UTF-8 support for the mail client alpine, using the test page http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-demo.txt (very nice). I have copied this to a local file, which I just cat. I was redirected also to the equally nice documentation page http://www.cl.cam.ac.uk/~mgk25/unicode.html#xterm While the answer for my case is clear (I run alpine in an rxvt which I start from .fvwm2rc and rxvt has NO UTF-8 support, so I'll have to sort that out otherwise), I am puzzled by the xterm behaviour. I use various xterms, both for local and remote use, and both started from a fvwm menu or from the shell, and I'm experiencing controversial behaviour. For all tests, environment: fvwm 2.5.10 under Suse 9.2 A) local xterm started from menu via a plain Exec xterm this invokes a very simple resource file ~/XTerm (attached) the resulting shell has NO LC_CTYPE variable set the resulting xterm has UTF mode on though dimmed (control-MB3) UTF support works perfectly (well, with default font it does not display Amharic text, who cares, but with Unicode best it does) B) remote xterm with ssh session started from menu via Exec xterm -class XTermLin -e ssh u...@host (actually I have several entries for different users and hosts, the host can be really remote or even my own local machine) the resulting shell has LC_CTYPE variable set for user root the resulting xterm has UTF mode on though dimmed (control-MB3) UTF support works perfectly (the only difference with the previous case is that the font names are different, Unicode best is just called font1) The resource file XTermLin is attached C) ssh xterm started from shell via an alias lin host user the alias lin is xterm -class XTermLin -T !:2...@!:1 -n !:2...@!:1 -e ssh !:2...@!:1 so this uses the same resource file as B. however this does not work : the UTF test file comes out as complete garbage LC_CTYPE may be set (for user root) or not set (for me) but this has no effect. I ssh'ed on the local machine. xterm comes out with UTF mode OFF (control-MB3), but even if enabled it does not work. Just comes out as a different type of garbage. The primary purpose of the XTermLin resource file (and a family of others) is nowadays to use a different background and foreground so I can immediately tell that gold-on-black windows are remote while black-on-beige are local. There is also some explicit font setting but these fonts should be UTF-8 compliant (iso10646-1), In fact I copied them from the system /usr/X11R6/lib/X11/app-defaults/XTerm These files are mostly historical (now all our systems are Linux, but before we had Suns and DECstations, and I had separate resource files with different backgrounds and some other adjustments. One of the historical files has a note before the font section one has to tell something more to make it a vt100 D) local xterm started manually from the shell So presumably uses resource file XTerm as (A) LC_CTYPEis not set xterm comes out with UTF mode OFF (control-MB3) behaves exactly as (C) i.e. does not work So apparently what happens is that when the xterm is started from a fvwm menu (A,B) UTF-8 support works. When it is started from the shell (C,D) it does NOT work even if one is using the SAME X resources. E) further test local xterm started manually from the shell as xterm -u8 this one comes out as UTF mode ON dimmed and works perfectly as (A) So it looks like that fvwm is somehow appending a switch -u8 IMPLICITLY ? .fvwm2rc contains some lines (inherited by the default Suse one) which make reference to StringEncoding=UTF-8:xft:sans-serif:encoding=iso10646-1:size=10; followed by some font specification, but these seem not relevant to applications like xterm. Can someone enlighten me ? -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- Nature on government cuts to research http://snipurl.com/4erid Nature e i tagli del governo alla ricerca http://snipurl.com/4erko!! resources commented as !? probably do not work !! respurces commented as !NO! definitely should not be used !! resources commented are reminder of how they can be set !! !! this is equivalent to -sp and sets the numeric keypad *SunKeyboard: true !! equivalent to -cr red : force cursor visible *cursorColor: red !! as -rightbar, move scrollbar to the *right* side :-) *RightScrollBar: true !! as -ls 200, increase scrolled lines *SaveLines: 200 !! useful resources -bg -fg -title ... ! *background: white !! !! stuff tested and not found useful !? *VT100*backarrowKey: true !? *SunFunctionKeys: false !? ScoFunctionKeys: false !? *HpFunctionKeys: false
Re: Start vs Init was Re: FVWM: question
On Thu, 19 Feb 2009, Thomas Adam wrote: 2009/2/19 Lucio Chiappetti lu...@lambrate.inaf.it: AddToFunc I StartFunction I Exec exec xterm do you really mean I StartFunction instead of InitFunction? With an I? I mean: AddToFunc StartFunction I Foo sorry, I was mislead since you typed an extra I *here ! AddToFunc I StartFunction I Exec exec xterm http://fvwm.lair.be/viewtopic.php?f=40t=1505 See the aforementioned link -- then you'll understand. Thanks, now it is perfectly clear (or thereabout). I have only one small doubt concerning modules. What I *had* was (my adaptation of the suse supplied configuration, commenting out lot of stuff and adding a few of my own) AddToFunc InitFunction + I Exec of one xterm + I Exec xsetroot with a particular colour -- + I Module for FvwmAuto and FvwmEvent -- + I Module FvwmEvent Event-FF ??? + I Exec of all other apps (clock, firefox, pine, rxvt ...) AddToFunc RestartFunction + I Exec xsetroot with a particular colour + I Module for FvwmAuto and FvwmEvent AddToFunc ExitFunction + I Exec xsetroot + I Exec xsetroot The items marked with -- were the ones repeated in RestartFunction. I do not know why ExitFunction called xsetroot (with NO arguments) twice. What I've done now (and tested with a restart, not a logout) is to get rid of RestartFunction, to get rid of the second xsetroot in ExitFunction, and to replace InitFunction with StartFunction as follows : AddToFunc StartFunction + I Test (Init) Exec of one xterm + I Exec xsetroot with a particular colour -- + I Module for FvwmAuto and FvwmEvent -- + I Module FvwmEvent Event-FF ??? + I Test (Init) Exec of all other apps (clock, firefox, pine, rxvt ...) So I do Test (Init) for all applications started only once in old Init. I left unchanged the xsetroot and the two Modules (--) previously called also in Restart. I am unsure what to do of the other Module (marked with ???) which was previously present only in Init. It is a module which resizes non-transient firefox windows to a fixed size selected by me (in contrast with the fact that firefox tends to use the latest size, and ignore x resources). Should I use Test (Init) Module FvwmEvent Event-FF ? CAN I do it (can one test a Module or only an Exec ?) I've done a ps and looks like that all three modules get new pid's after a restart, so I presume my current config is harmless. But I would appreciate a confirmation. -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- Nature on government cuts to research http://snipurl.com/4erid Nature e i tagli del governo alla ricerca http://snipurl.com/4erko
Re: FVWM: Alarm/calendar applications that work well in fvwm?
On Wed, 7 May 2008, Cameron Simpson wrote: On 06May2008 09:24, Lucio Chiappetti [EMAIL PROTECTED] wrote: it is possible to configure the list per-user to send daily MIME digests instead of individual mails ? To drift even further, I still don't understand why people prefer digests. I filter lists into separate folders on delivery Possibly at the end I would revert to something like that if I can't subscribe to this (or other technical list) by digest. Could you, as a digest preferrer, please explain what you like about them? I like the fact I receive a single digest per day, and I'd have to do nothing about it on my side (not that I would not know how to emulate what you suggest above by procmail, I do lot of other things by procmail http://sax.iasf-milano.inaf.it/~lucio/Procmail/) In general I have classified mailing lists in two categories, important and secondary (most technical lists are secondary). Normally I receive everything in my inbox (but a digest FORCES me to check it only daily and ENFORCES this rule) ... when I go on holiday I divert all important lists to a single separate folder (so that I can read it second-priority when I come back), and all secondary lists to /dev/null. Apart from that I normally look at the list of subjects at the beginning of the digest. In most cases I conclude there is nothing interesting today and delete the entire digest. In the case there is something interesting, I'll expand the digest in a temporary folder (just found a nice way to attach a formail invocation to a single key in pine) and look at that. NB strictly MIME digests with each message a separate RFC822 item. I find digests handy even on non-technical lists. We have a compulsory list for all the staff of my organization. In principle it should be used only for announcements (conferences, jobs, etc.), but it is often used also for trade union announcements, complaints, discussions, And every time the traffic increases somebody start complaining, So we are thinking to moderate it and create another discussion list. By default everybody would be subscribed to the discussion list, but could unsubscribe. To discourage people from unsubscribing, and to discourage real time discussions (or me too messages) we are thinking to make digests the default for the discussion list, so people would receive a daily digest unless they require otherwise. Anyhow, I repeat my question, is at all possible to receive digests from a majordomo list like fvwm ? -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not blame ME, I did NOT vote Berlusconi.
Re: FVWM: Alarm/calendar applications that work well in fvwm?
On Mon, 5 May 2008, Perry Hutchison wrote: BTW the FVWM list (and I think most other open-source help lists) prefers to keep discussions on the list, I agree with such policy Sorry, auto-typed the N response to Pine's Reply to all? ... Don't suppose you know a way to make it by default reply to the list when it's there, and to the sender only if it's not? I have no clue if it's possible to configure Pine that way, but I presume one can do it playing with pine roles and rules (I'm pretty sure I configured mine to use a specific alternate from address when replying to a specific moderated list to which I'm compulsorily subscribed using such alternate address). Since we are slightly OT about mailing lists, and since this FVWM list is recently giving rise to some more traffic than usual, does anybody know if it is possible to configure the list per-user to send daily MIME digests instead of individual mails ? I have now configured that way most of the mailman-based mailing lists to which I subscribe, but the FVMW list is a majordomo list, isn't it ? -- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html Do not blame ME, I did NOT vote Berlusconi.
Re: FVWM: acrobat/adobe reader, firefox and FVWM
1) about acrobat decorations Thanks to everybody for the replies, I also posted to an Adobe forum http://www.adobeforums.com/webx/.59b4f15e/0 and got an answer from Gaurav Jain that said the re-appearance of suppressed WM decorations when putting Adobe reader 7 in full screen was a known issue which was fixed in version 8. I've found a reasonable workaround for my case using AutoMove and making borders as thin as possible (unfortunately I have to do it at Style * because of the same non-honouring X11 requests issue, so I used 1 pixel) 2) about firefox and fvwm Still waiting for suggestion about how to force via fvwm (FPGraBFocus seems not enough ... should I learn FvwmEvent ?) that when I click on an URL link in acrobat on a desktop (and this correctly loads the URL in firefox in another desktop) I'm automatically switched to the other desktop. This is the behaviour I automagically have on Suse 9 / acrobat 5 / firefox 1.5 ... but not on Suse 10 / acrobat 7 / firefox 2 On Wed, 9 Apr 2008, Dominique Michel wrote: You are using the acrobat reader on linux when it suffer of a very bad rendering of the True Type fonts (another Adobe invention). This is not a linux issue, it does the same on windows. Seems not an issue for me. I do not see anything bad. Have you try software like xpdf and kpdf that doesn't suffer of this problem and that can do fullscreen viewing? I briefly tried xpdf yesterday to overcome the fullscreen decoration issue, but sincerely I found it was not quite suitable for my needs. Animated transitions were not working as expected, and the choice of accelerator keys was different. Also that won't be a general solution. It might solve my current program (displaying a presentation on a borrowed laptop), but not the general problem (taking only the PDF file with me and using whatever viewer I'll find on the computer available at destination, and/or publishing the PDF file on the web and then the user will use whatever viewer they have. The idea was to use PDF instead of powerpoint or openoffice because anybody will have a PDF viewer, and that will generally be some Adobe thing. -- --- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- () ascii ribbon campaign - against html mail /\ http://arc.pasp.de/
Re: FVWM: acrobat/adobe reader, firefox and FVWM
an addendum to my previous post On Wed, 9 Apr 2008, Lucio Chiappetti wrote: I added an entire line with: Style Adobe Reader StartsRaised, FPGrabFocus,NoTitle The window manager title bar correctly disappears when I start acrobat reader in a window, but as soon as I go full screen it appears again. !! It looks like a bug in acrobat reader 7 ... I tried also adding !Border, or replacing TitleAtBottom or TitleAtRight. When I initiate acroread they are honoured, but as soon as I go full screen the title bar appears on top, and then, even if I leave full screen, it remains there. Similarly, if I start acroread with a -geometry option, this is honoured initially, but is lost as soon as I go full screen (title bar appears anchored in +0+0). The only way I found to fix the thing is the following : - as soon as I am in full screen I use MB1 to get the window menu and invoke the Move function (or I could use standard Alt-F5) - with the arrow keys I move one glitch left, five glitches top (which displays -14 -25 to me, I have no idea why the movement is quantized:), then press MB1 - this freezes the window with the title bar almost entirely off screen (I still see some bits of the top and right border, I could get rid of them if I could nudge by 1 pixel at a time) Is there an fvmw command which causes a move to a predefined geometry ? If so I could associate it to the window menu, and/or an accelerator XXX, so I'd do ctl-L XXX at the start of my presentation (instead of plain ctl-L) -- --- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- () ascii ribbon campaign - against html mail /\ http://arc.pasp.de/
Re: FVWM: acrobat/adobe reader, firefox and FVWM
On Wed, 9 Apr 2008, Thomas Adam wrote: Why you're even touching ModulePath is unclear. It's set by FVWM. Leave it alone. I had a ModulePath in my original .fvwm. I did not write it. It was there in the suse 9.2 .fvwmrc (which I edited). The xsession error on the suse 10 laptop showed an error about the location of FvwmAuto etc., so I located them and corrected ModulePath. Is this an fvwm thing, an acrobat thing or both ? Oh, it will be almost certainly something in Acrobat Reader. Some xprop output would be nice here. This is beyond my understanding (I may give the impression of knowing more than I actually do :-)) What should I do exactly ? By the way, I just posted an addendum on the acrobat stuff. QUESTION: how can I force the automatical desktop switching ? I am not sure you can. This sounds like a quirk of Firefox to me. You are saying it was (is) a quirk of my OLD firefox, not a bug on the new version ? You could emulate it though if you wanted. Yes thanks, How do I do that ? assign FPGrabFocus to Gecko or what else ? -- --- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- () ascii ribbon campaign - against html mail /\ http://arc.pasp.de/
Re: FVWM: acrobat/adobe reader, firefox and FVWM
On Wed, 9 Apr 2008, Benjamin Karhan wrote: Key F1 A M AnimatedMove 100p 100p and ALT-F1 will move the given window to 100,100 i've never tried to use offscreen coordinates though... AnimatedMove itself works like a charm ... specifying the coordinates is a little tricky as it depends from the application. I started testing it with an xterm, using large negative values (slightly larger than screen size, so they'd offset the top-left corner of the xterm window from the bottom-right corner of the screen). To discover that acroread interprets them differently (its bottom-right corner from the bottom-right corner of the screen). So at the end I settled on Key F5 A SM AnimatedMove -0p -0p Now control-L Alt-Shift-F5 displaces my acrobat reader window so that the title bar and left margin are no longer visible. I still see the bottom and right margin, but never mind. ... still waiting for an emulation trick to switch to the firefox desktop when a click on an URL in acroread running on another desktop loads a web document into firefox. -- --- Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy) For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html --- () ascii ribbon campaign - against html mail /\ http://arc.pasp.de/
