| |
| |
| |
|
Page: 1 2
Comments:
<0> isn't it great when the channel topic answers your question? =P
<1> Yes.
<1> Speaking of that...
<1> pr3d4t0r: We could have refered ric24 to http://home.vrweb.de/~juergen.kahrs/gawk/XML/ if he is silly enough to use GNU awk. :)
<1> Unless, of course, you did.
<0> Ok, question. Suppose I have 2000 text files, each approx 10mb in size. I want to delete the first line of all of them. I don't want it to take all day. What do I do? (If I kill the first line in a text editor and save the file, it takes much less time than if I awk '{blah}' file> newfile ; mv newfile file ... How does the text editor do it so quickly?
<1> awk is scanning every line.
<1> I have no idea what you are doing with your text editor.
<1> if(FNR != 1) {print}
<1> Though I doubt that is the best solution.
<0> ..it has to do with actually writing all of the file back to disk or not.. The text editor somehow avoids that. I just found out that the sed -i command line argument does, too.. it's 'edit in place'... I dunno how it works, but it's fast, it seems.
<2> Hej.
<2> newmanbe: Thanks for the heads up.
<2> newmanbe: I prefer to avoid gawk's bells and whistles because they're incompatible with POSIX awk.
<1> And The One True awk.
<2> newmanbe: HeH.
<2> newmanbe: I'll go with POSIX (with apologies to A. W. and K.).
<2> newmanbe: As for Sebboh's question... couldn't tail help somehow?
<1> Using Plan 9's tail, yes.
<1> I don't know about your POSIX tail. :)
<0> ?? I can't imagine any use of tail that wouldn't involve re-writing the whole file to disk. But for that matter, I don't know how text editors avoid re-writing a whole file to disk, either.. But surely they must be able to do that: my ram usage doesn't spike when I save a small change to a large file...
<1> Which doesn't mean anything.
<1> Its very easy to write only part of a file.
<0> Ok. That's what I was wondering.
<0> Schools? ;)
<0> Well of course it's *possible* to modify only the beginning of a file.. you'd just leave the unchanged blocks in place on the disk and point the last block in your changes back to them. But, I don't think that my text editor (mcedit for example) addresses files in that way.
<3> I have a bunch of filenames with \n in them. How can I rename them all in a nice automated fashion? filenames are foo\nbar\n[...]baz.txt; i want to just get rid of everything up to and including the last \n
<3> rename(1) freaks
<4> Guerin: how does rename freak, paste
<3> xmb: thanks, i did manage to fix it using a for loop
<4> gg
<3> rename said something along the lines of `unable to stat filename containing newline'
<4> gay
<3> yeah, i thought so
<3> i wonder if we should blame perl
<4> bad coding style maybe
<4> problem with quotes aha
<4> =p
<2> Hej.
<5> Evening.
<2> goldfish!
<6> hi all
<6> who wanna answer my question
<6> a wanna have a script
<6> i need a script which does some work for me
<6> it's so easy
<6> when a exec iwlist scan
<6> it shows me the wireless network list
<5> pr3d4t0r: Hey hey.
<6> so i wanna choose the ones which has encription key off and the one which has the best signal
<6> then it automatically connect that wireless network
<5> Stop haxing wireless networks.
<6> ???
<5> hehe
<5> I'm joking.
<6> rkirmizi@laptux ~ $ iwlist scan
<6> lo Interface doesn't support scanning.
<6> eth0 Interface doesn't support scanning.
<5> unlucky.
<6> eth1 Scan completed :
<6> Cell 01 - Address: 00:12:BF:53:F7:67
<5> yeah
<5> oh no...
<6> ESSID:"KORDON"
<6> Protocol:IEEE 802.11bg
<5> You didn't paste :/
<5> BAH!
<6> Mode:Master
<6> Channel:6
<6> Encryption key:on
<6> Bit Rates:1 Mb/s; 2 Mb/s; 5.5 Mb/s; 6 Mb/s; 9 Mb/s
<6> 11 Mb/s; 12 Mb/s; 18 Mb/s; 22 Mb/s; 24 Mb/s
<6> 36 Mb/s; 48 Mb/s; 54 Mb/s
<6> Quality=53/100 Signal level=-69 dBm
<6> Extra: Last beacon: 549ms ago
<6> rkirmizi@laptux ~ $
<6> sorry about that
<6> do u use pastebin or sth. like that
<6> is it allowed to paste in this channel?
<2> rkirmizi: Please use the pastebin.
<5> Use of a pastebin is encouraged.
<2> rkirmizi: What you did is VERY RUDE.
<5> But, you know for next time :-)
<2> rkirmizi: There is a pastebin in the /topic, or you can use mine http://eugeneciurana.com/pastebin - it has awk syntax highlighting.
<6> has anyone have an idea about the problem ?
<6> can we do this work with awk and bash implementation
<2> rkirmizi: Sure.
<6> So
<2> rkirmizi: Put that in a pastebin so that I can look at it.
<6> who wanna help me?
<6> ok
<2> Hrm... Turkey.
<2> goldfish: Do you want to run with this one?
<6> this is iwlist eth1 scan result
<6> http://pastebin.com/870489
<5> pr3d4t0r: already working on it.
<4> lalla
<4> rkirmizi: i have wlan scripts if you want
<4> xmb.ath.cx/code/wlan/
<4> see the M script for wlan access
<4> err, MM script
<4> M is short for something else
<6> which one
<6> there are a lot
<4> MM
<6> isn't it so long
<4> its just defining of my wardriven aps in bulgaria
<4> its small and tight
<6> a read that awk scripts are usually just a few rows
<2> goldfish: As we said in Mexico: God makes them, they pair themselves together.
<4> its a shell script
<4> nice predator
<4> =)
<6> thz
<6> i ll try this
<6> see u later
<6> xmb: thx alot
<4> np
<5> awww
<5> he left.
<4> MM big, not mm small
<5> :-(
<5> *boggle*
<5> Hm, I wonder why it thinks 11 is bigger than 100
<5> strtonum to the rescue. Silly me.
<5> pr3d4t0r: Can you specify files inside awk scripts? Like, /foo/{stuff;} file ? or can files only be p***ed on the command line
<5> Well, that question is to anyone :-)
<2> goldfish: Yes.
<2> goldfish: And yes.
<2> goldfish: Look at getline, close, etc. for opening your own files. You may use redirection to files.
<2> goldfish: And, say you have an awk script called script.awk
<2> goldfish: script.awk < filename.txt or script.awk filename.txt have the same effect.
<5> Ah ok. Thanks. I have lots of awk commands in my xinitrc for displaying stuff on my toolbar, just wondering if it's feasable to put it all into one awk script
<5> well, not lots, a few.
<7> "goldfish" at 212.2.174.130 pasted "iwlist parser" (36 lines, 663B) at http://sial.org/pbot/22710
<5> if anyone would like to have a look, and tell me if there's any nicer way to break up the output, and reconstuct it.
<5> xmb: !!!!!!!
<4> what
<4> goldfish
<4> the parser?
<4> cool you tried but probably too much code
<4> id do in maybe 5 lines or 10
<4> lets see
<4> looks good =)
<4> its aight
<5> yeah, using split() is rather ugly
<4> goldfish: you dont need strtonum
<4> awk makes that automagically
Return to
#awk
or
Go to some related
logs:
eradioportal ubuntu
qemu active-sync usb_add
#osdev
hfstools
#css
php tmp_name is empty
#perl
#mysql
jedit ubunto
usb-storage.ko slow
|
|