[Solved] AwesomeWM interactive debugging?

Hi all,
I was wondering if anyone knows of a way to try out different Awesome settings interactively from within a session?
The standard way to test settings seems to be to modify rc.lua and then either restart Awesome, or run a new sessions in Xephyr.
Does anyone know if it's possible to open a Lua console that is linked to the current awesome session? It would be nice to be able to inspect things like "awful.tag.getnmaster()" from within the console.'
Any ideas?
Last edited by hughitt1 (2013-10-13 13:44:00)

Ah okay, I had tried using awesome-client before, but could not figure out how to access properties from the current session. Making naughty and awful both global variables, I was able to run:
awesome# naughty.notify({text=awful.tag.getnmaster()})
I would be nice if awesome-client had some more of the features in a typical lua console (e.g. print() and variable inspection), but this is a pretty helpful step.
Thanks for the suggestion!

Similar Messages

Remote JPDA interactive debugging of mod_plsql application with JDeveloper

We have an existing application running on Oracle DB v10.1.0.4 which we want to debug insitu in order to analyse its operation and functions. The application is best described as a 3-tier Oracle PL/SQL WebEnabled Application implemented using the specialised mod_plsql module of the Oracle HTTP Server [Apache 1.3].
We want to be able to interactively debug the pls/sql code whilst a user is
using the live application [from the web browser] using the JPDA remote
listening facilities of JDeveloper 10.1.2. How can this be done ?
After extensive research , I have found only two online resource that briefly
indicate this can be done: viz
1. http://www.oracle.com/technology/tech/pl_sql/pdf/Paper_30720_Ppt.pdf
- The debugging user sets a cookie via the browser UI specifying JDWP host and
port Causes mod_plsql to call Connect_Tcp before its normal calls and
disconnect after these.
2. http://web51-01.oracle.com/oowsf2004/1401.pdf
Web mod_plsql applications
Create a UI form to set a cookie in the browser
Pass the hostname and port to the owa_debug package
owa_debug transparently calls the server-side PL/SQL to connect with the
debugger listener for each subsequent page requested by that browser
This information is not enough for me to actually be able to do this. I need
more details/ example etc. I also have not been able to find ANY additional
info on the owa_debug package- which is critical to the whole solution. Any help appreciated.

These are the steps to set up the database and JDeveloper for debugging your mod_plsql Web applications.
--- Database setup ---
1. Load additional PL/SQL gateway debug packages:
cd $ORACLE_HOME/wwg/admin/owa
sqlplus "sys/... as sysdba"SQL> @pubowad.sql
SQL> @privowad.plb
SQL> @pubjdwp.sql
2. Grant debug connect session privilege to DAD database user:
sqlplus "sys/... as sysdba"SQL> grant debug connect session to <DAD-database-user>;
3. Create the Web HTML interface to initiate debugging in the DAD database user's schema:
cd $ORACLE_HOME/wwg/admin/owa
sqlplus DAD-database-user/...SQL> @owaddemo.sql
4. Recompile your application procedures with debug information:
sqlplus DAD-database-user/...SQL> alter session set plsql_debug=true;
SQL> create or replace procedure XXXX ...;
5. Sets DAD attributes for debugging:
In the DAD setting inside dads.conf, add
PlsqlOWADebugEnable On
PlsqlMaxRequestsPerSession 1
PlsqlExclusionList \#None\#
and restart your Oracle HTTP (Apache) listener.
--- JDeveloper setup ---
6. For remote debugging (e.g. debugging a PL/SQL Web application executed by mod_plsql), choose "remote debugging" as follows:
* Choose Tools -> Default Project Properties... menu.
* Select Profiles -> Development -> Debugger -> Remote tree node.
* Check Remote Debugging option.
* Select Listen for JDWP option.
* For Database Connection for Locating PL/SQL Source:, select the database connection just created.
7. Set a breakpoint in your application procedure:
* In Connections tree, expand Database node. Expand the new database connection node. Expand the database user node. Expand Procedures, Functions, or Packages node. Double-click the library unit you want to set a breakpoint in.
* In the program source of the library unit, single-click at the left-margin of the source line you want to break.
8. Start remote debugging:
* Start the debugger listener from Debug -> Debug Project menu. Specify the TCP/IP port number the debugger should listen for JDWP (debug) connection.
--- Invoke Web application in debug mode ---
9. In your browser, go to the URL http://<host>:<port>/<DAD-virtual-path>/owa_debug_demo.main_form. Check the "Enable JDWP Debugging" box, enter the "JWDP Debugger Host" as the host of your JDeveloper machine, enter the "JDWP Debugger Port" as the TCP/IP port that JDeveloper is set up to listen for JDWP (debug) connection. Then hit "Create Debug Session". This will create a cookie for you which will be used to track your "Debug Session" preferences.
10. In your browser, invoke your Web application's URL. You should your application session being connected to your JDeveloper for you to debug.

[SOLVED] Non-interactive SSH login and shell startup files

I have a problem getting git-annex connecting to my arch box because of
PATH not being correctly set. When diagnosing this problem I noticed that
~/.bashrc seems to be ignored for non-interactive SSH logins. More specifically,
ssh myhost env
Shows that when connecting to my other computer running gentoo, or to a
separate server (probably running some flavor of Debian, I am not sure),
environment variables defined in ~/.bashrc are present, but when connecting
to my arch box, they are not. All three computers have identical ~/.bashrc
files and ~/.bash_profile is set to read ~/.bashrc, and in all cases the file is
sourced for an interactive login (i.e., 'ssh myhost' followed by 'env'). I tried to
search through the various bash files in /etc, but didn't find anything related.
Is there some bash or SSH setting that controls this behavior?
Solution:
I looked into this a bit more and found BASH_ENV which can be used to point to
a file which is sourced for non-interactive shells. So I set
BASH_ENV=~/.bashrc
in /etc/environment, and now ~/.bashrc is sourced even for non-interactive
shells and thus my PATH is correctly set. What I still don't know is why
arch behaves differently in this respect, but I guess that will remain a mystery.
Last edited by Nuteater (2012-07-27 18:59:30)

try creating a .login file and put exec bash in there.
Not _super_ familiar with csh, but I *think* csh loads .login only on login shells (as apposed to always loading .cshrc).
Barring that, the following should work.
if (! $?prompt) goto cshrc_end
exec bash
cshrc_end:

[SOLVED]AwesomeWM loads to a grey screen

All that I have done.
# pacman -S awesome
Awesome automatically is in my sessions for GDM.
It loaded the first time, so I followed the wiki and cp'd the file in /etc/xdg/awesome/rc.lua ~/.config/awesome
That worked, I then changed the theme in /usr/share/awesome/themes/default/theme.lua to zenburn/theme.lua
Restarted AwesomeWM via
MODKEY+Ctrl+R
Restarted, and a grey screen showed. Now whenever I run Awesome it just loads to a grey screen. I can see the mouse and move it, but no panel shows up or wallpaper. I tried uninstalling via
pacman -R awesome
and trying again 3 times and that hasn't fixed it. The awesome version was 3.5.5 Kansas.
Last edited by BabbyUser (2014-11-26 20:17:43)

jasonwryan wrote:
BabbyUser wrote:
jasonwryan wrote:https://wiki.archlinux.org/index.php/Awesome#Themes
This doesn't help. No bar appears, or wallpaper, nor does the menu when I right click. It's as if it was frozen.
It doesn't help because you are not following the instructions...
Moving to NC.
How am I supposed to do anything when NOTHING loads. I know how to setup the wallpaper. Basically, when AwesomeWM loads, it FREEZES.

[Solved] AwesomeWM Java menus don't stay open

Hello,
I'm currently running awesome-git, and having trouble with Java Swing applications. The application I am attempting to use is called Freemind. When I attempt to open a menu, combobox, etc using a mouse click, the dialog immediately disappears. Using the keyboard shortcuts (e.g. alt + f) to open the menu, however, seems to work okay.
I found a lot of information about solving various Awesome/Java issues and tried out several different fixes including disguising the window manager name, but so far none have worked. I also tested Freemind in Gnome Shell to make sure it wasn't an application issue and there were no problems there.
System info:
Awesome 3.5 (2012-12-31, AUR: awesome-git)
Freemind 1.0.0 (beta 9, AUR: freemind-unstable )
java version "1.7.0_09"
OpenJDK Runtime Environment (IcedTea7 2.3.3) (ArchLinux build 7.u9_2.3.3-1-x86_64)
OpenJDK 64-Bit Server VM (build 23.2-b09, mixed mode)
Any suggestions?
Thanks,
Keith
Last edited by hughitt1 (2013-01-03 00:43:43)

Hi tdy,
Including it at the end of rc.lua works!
So for not explaining things better before (It's been a couple weeks since I tried fixing this). I believe the approaches I tried were:
1) the "old" fix (AWT_TOOLKIT=MToolkit freemind-unstable)
2) and also calling "wmname LG3D" before running freemind -- I checked to make sure the setting took effect (calling wmname without any arguments) and since it returned "LG3D", I assumed that calling it in rc.lua or .xinitrc would not have worked any better. I was wrong
Thanks for the help!
Keith

[SOLVED] Add GCC debug flags

I need to debug a program called tenace [1] which I maintain in AUR [2]. My question is how do I change the FLAGS to add debug (gdb) capabilities. The program is written in C.
This is the current PKGBUILD [3]
gdb reads:
Reading symbols from tenace...(no debugging symbols found)...done.
Please be a little verbose in your answer I don't know what flag to add not don't I know where I would add flags. It seems to me I want to change this line:
./configure --prefix=/usr LDFLAGS=-lm
but please correct me if I'm wrong.
[1] https://www.df7cb.de/bridge/tenace/
[2] https://aur.archlinux.org/packages/tenace/
[3] https://aur.archlinux.org/packages/te/tenace/PKGBUILD
Last edited by foppe (2014-04-01 22:52:32)

Thanks,
Adding
options=('!strip')
does what I want it to do. Running gdb gives a backtrace with the function where the error occurs. Those function-names were previously not available.
Marking as SOLVED

[SOLVED] wmii *interactive* status bar?

Hi to all!
I use to write infos in the /rbar of wmii, and everything works.
Now I would like to create buttons down there, like the tags' buttons in the bottom-left corner.
If you click on a tag, you get the corresponding screen. Analogously, I could put a status message for, say, my sound volume, and bind a program to it to increase or decrease the volume.
Is it possible?
Last edited by iosonofabio (2010-01-29 09:15:52)

iosonofabio wrote:
Actually, it is not so easy. I would like to generate different events depending on which button I clicked on, e.g.:
- change volume for button 1;
- change screen luminosity for butt 2.
So I need wmii to generate different events for different buttons, and not only a generic "RightBarClick 1 status" for the whole bar...
echo "Volume button" | wmiir create /rbar/volume
echo "Luminosity button" | wmiir create /rbar/luminosity
wmiir cat /event # Now click on then and observe the events.

[Partially solved] AwesomeWM - Hide borders on the fly

Hello there!
I'm looking for a way to hide 1px borders from selected ( or all ) windows on the fly, with the keystroke, just like I can hide and show statusbar with alt+b thanks to
awful.key({ modkey }, "b", function () mywibox[mouse.screen].visible = not mywibox[mouse.screen].visible end),
Is there similar function for borders, or maybe some workaround?
Some win32 games that must run in windowed mode could benefit from it very much, as well as cairo-dock that runs great, except that borders are drawn around its every instance.
Thanks in advance!
Last edited by 0x29a (2014-11-13 10:28:10)

Ha! Good thinking!
This worked (wmctrl save lifes!):
properties = { below = true }},
{ rule = { name = "cairo-dock" },
properties = { border_width = 0 } },
{ rule = { name = "cairo-dock-sub" },
properties = { border_width = 0 } },
Last edited by 0x29a (2014-11-13 10:27:40)

[SOLVED] Awesomewm - How do I start a program in a certain workspace?

For example, If I wanted to run 'xterm -e alsamixer' in workspace 2, how would I do that?
Last edited by itsbrad212 (2010-04-25 03:36:22)

Ok, I have the following in rc.lua:
{ rule = { class = "Swiftfox", instance = "swiftfox" }, properties = {tag = tags[1][1]}},
{ rule = { class = "Conky", instance = "conky"}, properties = {tag = tags[1][1]}},
{ rule = { class = "Alsamixer", instance = "xterm -e alsamixer" }, properties = {tag = tags[1][2]}}
EDIT: I'll have a look at this: http://bbs.archlinux.org/viewtopic.php?pid=662615
Changes:
{ rule = { class = "Swiftfox", instance = "Navigator" }, properties = {tag = tags[1][1]}},
{ rule = { class = "XTerm", instance = "xterm" }, properties = {tag = tags[1][2]}}
Firefox works, but how should I make a rule that only alsamixer (and not all xterms) go into workspace 2?
Last edited by itsbrad212 (2010-04-24 22:01:35)

[SOLVED] AwesomeWM Battery Meter

Hello all,
I've been rewriting my rc.lua on my netbook for Awesome, and my head is probably just drawing a blank, but I'm having a problem getting the battery meter to show something other than 0% Battery.
This is the code for my vicious widget:
http://pastebin.com/Bt60surG
and I do have the widget showing, it's just always red and says 0%.
I have the acpi package installed, and it works through terminal, and starts on boot, wasn't sure if it was necessary though.
like I said, probably overlooking something, but I have been hacking at it for a couple hours, searching around and what not, and I was like "yeah... forum time" Lol
So yeah.. Any ideas?
Last edited by themattbeballin (2011-11-12 22:49:44)

anrxc wrote:Default battery widget reads /sys/class/power_supply/<your battery ID>. Check that you have provided the correct ID.
If you did get it right check that the data in that directory is complete. Does "present" file contain 1? Does "charge_now"
contain a charge other than 0? If not you could get support for your battery from /proc with batproc widget or worst case
from batacpi widget that uses acpitool. But likely that is not the case, most likely you got the ID wrong.
Wow. That fixed it... When I was looking at it... ACPI was reading BAT0 but /proc was saying BAT1.
Worked. Thanks!

[SOLVED] awesomewm vicious.widgets.wifi returns nil

I cant seem to figure out why the returning args (all of them) are nil..
this is the widget code I have:
wifiwidget = wibox.widget.textbox()
vicious.register(wifiwidget, vicious.widgets.wifi,
function( widget, args)
return string.format("%.8s - %sdB", args[1], args[7])
end, 19, "wlp2s0")
although, i feel thats not the culprit because regardless of what i put there, every entry in args[] is nil.
I have iwconfig installed and it returns the correct information for wlp2s0
vicious wifi script is in /usr/share/lua/5.2/vicious/widgets
any thoughts or further info needed?
Last edited by pdizzle (2014-10-02 13:18:42)

resolved...
in the case of the wifi widget, the returned value is a table.
so to call the values i was looking for it should be formatted:
wifiwidget = wibox.widget.textbox()
vicious.register(wifiwidget, vicious.widgets.wifi,
function( widget, args)
return string.format("%.8s - %sdB", args["{ssid}"], args["{sign}"])
end, 19, "wlp2s0")

Forms 9i and Debug

Hi I here that there is an interactive debug feature in forms 9i, whereby you can see what the user is doing and debug it at runtime. I also hear that this does not work over firwalls.
Does anyone know
1)If the feature works, as it sounds useful and are there any restrictions
2)Why it does not work over firewalls (most support is undertaken by 3rd party companies over firewalls.) Could somekind of proxy server like Connection Manager (CMAN.ora) could be used to overcome this problem

Mark,
you cannot debug another user's session. You can run a form in debug mode using e.g. the following URL:
http://<server>:<port>/forms90/f90servlet?form=<myform>&debug=yes.
This will provide you the server name and a port number. Now with the according fmb file open in forms builder go to "Debug/Attach debug" and enter server name and port. Then you can set breakpoints and therewith debug the module.
The port is the reason why it doesn't work over firewalls. The FW admin would need to open that port, and for each debug session it will be a different port number.
I'm not sure about CMAN.
I only use this feature on my local 9ids and on a testing environment within our LAN.
Hope this helps
Gerald Krieger

Interactive debugger for Lightroom plugins

The Lua IDE "ZeroBrane" now supports interactive debugging of Lightroom plugins. This means you can set breakpoints, step through your code line by line, examine/watch your variables as you do so etc. It can be quite helpful when debugging a complex situation in a plugin, and I don't think it was previously available.
For more details see here,
http://notebook.kulchenko.com/zerobrane/debugging-lightroom-plugins-zerobrane-studio-ide
See the last screenshot for a view of the interactive debugger.
(Disclaimer: I am not affiliated with them but have worked a little with their software, and like it. Paul, the developer, is very responsive and helpful, in case you should still encounter the occasional quirk.)

I just tried it in Lightroom 5.6/Mac, and it worked well. (I don't have the latest version of ZBS installed, however.)
The only thing I see is that if you import it as local LrMobdebug, it should be "LrMobdebug.start()", not Lrmobdebug.start(), i .e. upper/lower case 'm'. (turns out this is wrong in the article too). However, LR should inform you of this when you start the plugin, with the error message "An internal error has occurred - attempt to index global Lrmobdebug (a nil value). So maybe you have caught this already.
The only real suggestion I have is to restart Lightroom, as that sometimes helps. Otherwise, I followed the steps in the article, i.e. script opened in ZBS, project directory set etc. I did not set a breakpoint as Lrmobdebug.start() already breaks itself.

990FXA-GD80 "FF" Debug Code

Debug code "FF" is showing on the motherboard on boot up - I have no display either. Any help would be greatly appreciated.
990FXA-GD80 AM3+ Military Class II
AMD Phenom II X6 1100T
4x4 DDR3 - Corsair 1333 mhz
WD Caviar Green HD 1 TB
EVGA GTX570HD Superclocked edition
Corsair H50 Hydro CPU Cooler
Thermaltake 750W PSU (TR2 RX 750W)
Win7 64 Bit
Thank you.

Do a search for "ff" in the main forum and you'll see the tremendous amount of people who are suffering from this.
https://forum-en.msi.com/index.php?topic=149035.0
https://forum-en.msi.com/index.php?topic=147290.0
http://www.tomshardware.com/forum/309538-31-solved-890fxa-gd70-debug-code-load-edit-bios
I discovered that I could use two 4 pin cables from my power supply to plug into the 8 pin power slot on MoBo as well as the 24 pin power. I'm a complete novice/newbie to the extreme so I essentially got lucky. It appears to be working so far, but I'm skeptical.
If you find out that your cause is the same as mine, make sure you plug both 4 pins in securely, its a bit of an odd fit.

SAPSQL_ARRAY_INSERT_DUPREC dump while saving PO in ME22N

Hi All,
SAPSQL_ARRAY_INSERT_DUPREC dump while updating PO in ME22N.
Error analysis
    An exception occurred. This exception is dealt with in more detail below
    . The exception, which is assigned to the class 'CX_SY_OPEN_SQL_DB', was
     neither
    caught nor passed along using a RAISING clause, in the procedure
     "/SAPSLL/OBJIDX_DB_UPDATE_R3" "(FUNCTION)"
    Since the caller of the procedure could not have expected this exception
     to occur, the running program was terminated.
    The reason for the exception is:
    If you use an ABAP/4 Open SQL array insert to insert a record in
    the database and that record already exists with the same key,
    this results in a termination.
I tried to find SNOTE for this, but cannot find for ECC5.0. Can anybody help.
Thanks.

This occurs because you try to insert a data with a key that already exists on the database.
I had once a similar problem with ME22 and I solved it by debugging and discovering which was the corrupt data. I think there was an error during PO savingand not all data was inserted, leaving some inconsistent data in tables. Later, when you try to save the purchase order, a data with same key is generated and tried to be inserted. If this is your case, you have two options: complete the data manually (inserting data in necessary tables) or delete the inconsistent data.
Regards.

[Solved] AwesomeWM interactive debugging?

Similar Messages

Maybe you are looking for