| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Comments:
<0> sudo runs the command as the other user.
<1> can i run postgres with sudo?
<0> If you configure sudoers properly.
<0> man sudoers
<2> man man
<3> Vote ***y http://rwk2.RaceWarKingdoms.com/cgi-bin/rwk.cgi?swimsuitStardemon
<3> http://rwk2.RaceWarKingdoms.com/cgi-bin/rwk.cgi?swimsuitStardemon
<4> what is this mean.... i found this on fstab "UMASK = 000 0 0"
<1> nandong, all user can read write
<5> hi
<4> ok,... so what is mean with first 0,... second 0,... and the third 0 .... ?
<5> how can I upgrade my tcp dump using tarball, I couldn't find any guides in google, I am using rh 71
<6> anybody speaks spanish?
<1> nandong, man umask
<4> ^^
<7> can somebody help me? Im on fedora core 5 and i cant seem to find drivers for this model gfx card "V9400-X/TD/128M" Asus NVida Chipset
<6> anybody may help me with ubuntu?
<1> Lingoman, www.nvidia.com
<1> download and run sh NVIDIA-linux-xxxx-pkgx.run
<7> tried it.. they say its only supposed by windows.. but im sure there would be a linux release maybe or something that would work with it.
<1> no
<6> I am a new user of linux may call me please?
<7> no?
<6> no what?
<6> no call me or no help me?
<1> i use NVIDIA
<6> Ups what?
<1> Distribution: Slackware 10.2.0 Kernel: 2.6.15.6-slicers X.Org: 6.8.2 KDE: 3.4.2 Qt: 3.3.4 XChat: 2.4.5 Perl: 5.8.7 VGA: nVidia Corporation NV5M64 [RIVA TNT2 Model 64/Model 64 Pro] (rev 15) @ 800x600/24 bpp/75 Herz
<1> Soundcard: Silicon Integrated Systems [SiS] Sound Controller (rev a0) CPU: Intel(R) Celeron(R) CPU 1.80GHz (1792.898 Mhz) CPU-Cache: 128 KB RAM: 238 MB / 247 MB Networkstats: Down: 15.1 Mb Up: 1.0 Mb
<1> i use NVIDIA-Linux-x86-1.0-7167-pkg1.run
<8> easy bash question... I have a variable with a list of comma seperated items, and I want to grab the $nth item from the list.
<1> knop, whats up?
<8> the question, of course, is how to do this. :)
<8> and malisa22i is on-joining from the looks of it
<9> randyg, cut -d':' -f<n>
<8> right now I just $() cut, but that makes it take a half second a line...
<10> lagi tangi turu
<9> cut -d':' -f<n> <<< $VARIABLE <- that if you have bash3
<9> if you dont
<9> echo "$VARIABLE" | cut -d':' -f<n>
<9> you could also use awk
<8> techdeck: well, comma seperated, so -d ','... that's what I have right now. it's too slow!
<9> oh ops
<9> whats slow about it?
<8> I know bash has built in string matching and substitution, but I can't find any way to make it simply split items based on another var...
<9> it cant do what you want
<9> not natively
<8> it fork()s and exec()s cut, which is very slow. :P
<9> well it maybe can but it wont be easy and will probably work faster without
<9> but if you want speed..
<9> use mawk instead
<6> I am a new user of linux may call me please?
<9> mawk -F',' '{print $4}'
<9> or whatever
<9> mawk should be much faster
<8> still would have the pain of forking and execing each time... and it's pretty damn hard to get faster than cut.
<9> if you can somehow use xargs instead of some loop you have in there it will be much faster
<9> but I don't really see what you're doing so I can only guess
<8> bar=$(echo $foo | cut -d ',' -f $field) is about the current line...
<9> bar=$(mawk -F',' '{print $'$field'}')
<8> takes a good second every time that line is run, which for a really long file...
<9> err
<9> bar=$(mawk -F',' '{print $'$field'}' <<< "$foo")
<8> won't help. the problem is the time $() takes to work, not the time of the program called.
<9> no it isnt
<9> $() runs instantly
<5> how can I upgrade my tcp dump using tarball, I couldn't find any guides in google, I am using rh 71
<8> $(somebinary) makes bash fork and exec that binary... which is very slow.
<9> its opening a subshell
<9> which is not slow at all
<9> I have no idea what are you talking about
<9> if its slow its not because of $()
<9> be sure of it
<9> show me the whole thing
<8> nor do I have any idea what you're talking about... forking and execing another program is SLOW.
<8> nah, the code is reallly ugly, not showing it to anyone. :P
<9> depends WHICH program
<9> mawk for instance is much faster than cut
<8> no, it doesn't depend which program. the actual exec*() system call is going to be slow, period.
<9> WTF
<9> no it isn't
<9> just try it and see..
<8> cut should be much faster than any awk variant for simply reading one character at a time until it gets to the right count of delimiters... but it doesn't matter. that's not why it's slow. :P
<9> !@#
<9> bull****
<9> try it
<8> sigh. ok, tell me why bash will be able to ask the kernel to start mawk running faster than it can tell the kernel to start cut running.
<9> it wont, they will start at the same time
<9> but mawk will finish first
<9> because it is faster than cut
<8> the speed difference between mawk and cut, if any, is trivial compared to the time it takes to exec the binary.
<9> why dont you try it?
<8> both mawk and cut should be able to do this in a few milliseconds tops. the extra second of delay is the time it takes to start either one of them.
<9> as you wish
<11> ladies, play nice.
<4> refer to awka.sourceforge.net/specifics.html mawk or awka is faster than cut .....
<4> ^^
<0> How do I force a page to refresh every 5 seconds in HTML?
<8> the difference between the two is trivial compared to the time it takes to exec the binary in the first place. :P
<4> :P
<9> randyg, c=0; for i in ${VAR//,/ } ; do [[ "$c" = "$whatever" ]] && echo $i && c=0 || let c++ ; done
<9> thats the bash way
<9> if you think this is better
<9> :p
<11> why not time it and see
<4> yup.... use stopwatch somehow ...... ^^
<11> uh
<11> use 'time'
<8> this script is running on a user-mode-linux kernel, which seems to have pathetically horrible times for execing programs... it doesn't matter if cut or mawk is used, simply starting it makes it slow. :P
<8> stopwatch... i.e. time? :P
<9> so is this solution better?
<9> it should be even slower
<9> and change c=0 to let c++
<9> actually..
<8> sec, still reading that. :P
<9> c=0; for i in ${VAR//,/ } ; do [[ "$c" = "$whatever" ]] && echo $i ;let c++ ; done
<9> like that
<0> Nevermind my question. It was easy to find the answer.
<4> <0> Nevermind my question. It was easy to find the answer. : ???
<0> I asked a question a few minutes ago, then did a quick search. I found the answer before anyone acknowledged or answered.
<0> So I said nevermind.
<4> ops
<8> techdeck: wootness! that is about 50 times faster than before.
<4> i havent read it ......
<8> now the output scrolls faster than I can read it, rather than one line per second.
<4> :P
<9> randyg, god knows what is going on your box
<9> :p
<8> techdeck: as I said, it's a user-mode-linux kernel, and seems to have really horrible exec times... it takes a bloody second to start any program!
<8> well, 1/3rd a second or so, and my program does 3 of those per line output.
<9> then use a normal kernel :P
<8> then pay for me to have a coloed real server instead of a virtual server. :P
<4> :D
<9> you can have a virtual server with a normal kernel
<9> I have one :p
<9> xen is excellent
<4> like vmware isnt ?
<8> $14/month for 2gb disk, 20gb transfer, and THREE ip addresses... oh, and no restrictions on what kinds of services you're allowed to run, legal or otherwise.
<9> vmware is ok, but xen gave me better performance
<8> UML should, in theory, be much faster...
<9> randyg, that ****s
<9> :p
<9> and yeah I heard UML is very fast too
<8> getting three ips makes life much nicer, and it's a lot cheaper than most placegive ips for...
<9> randyg, hmm I just noticed something
<9> if you have anything with a space within the ,, it wont catch it
<9> so wait let me give you a fix
<8> something like it needs c=1 instead of c=0 to work? :P
Return to
#linux
or
Go to some related
logs:
#netcafe
#india
Ranjan dalnet
#india
#chat-world
#allnitecafe
#allnitecafe
#india
Bibbidi-Bobbidi-Boo mp3
#worldchat
|
|