« first day (1248 days earlier)      last day (3700 days later) » 

2:00 PM
@Kiwy Urgh. Don't like nano, never did.
But that's how it should be; choose whatever tool works best for you and get the job done
 
@JennyD I think nano can be define as a emacs for the dumb
 
@Kiwy Am I mixing it up with pico?
I might be
Because pico was just horrible.
 
@JennyD don't know never used pico but that's possible yes
 
@Kiwy It was the default editor in... whatever the IMAP client I used to use back in the dark ages was called.
 
nano and pico are basically the same, and both terrible :)
 
2:02 PM
@casey Thanks, that's what I thought :-)
 
pico is what those of us who have been using unix/linux since the 90's at least remember, and nano is what the newer folks know
 
@casey @JennyD the only good text editor I know nowadays is really notepad++ but well not available on any linux/unix
 
! emacs and vim are the best, for different reasons but I know of nothing that compares
 
@Kiwy On my Mac I'm trying to get used to Textmate, but I still get tangled up with it not having the right finger macros
 
I have also heard good thing about notepad++ but since it's windows only I've never tried it. It looks like I might be able to convince sublime to be as good as my emacs but I've never taken the time.
 
2:04 PM
@casey Agreed. Vim for anything shortish, emacs for everything else.
3
 
@JennyD exactly
 
@JennyD ha, I'm actually running a userscript that enables some emacs bindings when editing SE posts :)
 
you both have issues :D vim is usable yes, but I really can't find anything fantastic about it
 
@Kiwy Vi is everywhere. Nano/pico isn't.
 
@terdon my biggest complaint about sublime is that it isnt free. well that and it doesnt syntax highlight fortran :)
 
2:06 PM
@JennyD it's not because something ios available in most place that it make it fantastic :D see Microsoft Windows
 
@casey I'm sure it does for normal languages that normal people use in a normal day. :P
 
@Kiwy Let me rephrase, then. For a sysadmin, being able to use vi is essential because it's one of the tools you can trust to be available on any unix system.
 
@kiwy the ease of composing very powerful commands in vim is what makes it fantastic.
@terdon what are insinuating? :)
 
slm
@terdon that's what we use in our corelab on windows systems
 
And for quick editing, its yanking/copying/searching/replacing commands are very handy
 
2:08 PM
Yeah, a basic familiarity with vi is essential. What if you're on an embedded system for example?
@casey who me?
 
:)
 
@slm and you like? It does look quite powerful.
 
@terdon this is awesome sause with a slice of lime
 
@Kiwy what is, sublime?
 
slm
@terdon yeah it's easy to install, you just drag around the executable it works well enough, I don't develop in it daily, but it has tabs and all that so it's tolerable
We might need to reopen this one
0
Q: xmodmap: dead key after switching two keys

muffelMy keyboard switches the (German) keys [^] and [<], so I want to switch the two keys. I identified the keycodes using xev and created a file ~/.xmodmap containing keycode 94 = asciicircum degree keycode 49 = less greater When executing xmodmap ~/.xmodmap hardware key 94 now correctly outputs ...

 
2:11 PM
@JennyD and that's why I'm able to use the basics, but I'm still really dissapointed when I tend to replace regexp because I never ever know what character I should escape or not
 
@Kiwy Then you need to learn regexps better :-)
 
slm
wish @drav was around, he knows xmodmap better then most
 
@JennyD I do master regexp pretty well but not on linux unix system and more on modern system where the escaped character are logical
 
Yea, my xmodmap-fu is limited to swapping caps and control and turning right-win into compose
 
slm
That's what the op is doing but it isn't working
 
2:13 PM
@Kiwy I don't find it illogical on unix... but then I've pretty much only used them there.
 
@Kiwy "Master" regex but not on *nix? ummmm....
 
slm
research a bit more before reopening
 
Escaping is pretty logical. You just escape any characters that have special meanings in regular expressions
@slm yeah, I'm trying to figure out the problem
 
@terdon and@JennyD the most illogical thing I can came up is the use of parenthesis and group, if I want to use group I use () if I want to match a parenthesis then I will escaped and hae \(\) the things is on Java and other regexp engine I use it work like this but it work the other way around on nix
 
@Kiwy ...No. No, it doesn't.
 
2:17 PM
@JennyD well screw the regexp on Nix* I hate them, I have no problem to write regexp but when it comes to escaping caracteres it driving me nuts
 
@Kiwy Well, if you actually believe what you wrote above, I can understand that you are confused.
 
@Kiwy no no, that's just basic regex, who uses those? Java uses a version of PCREs and so can grep with -P and perl of course
 
@JennyD each time I try it and the worst example is using sed and the horrible s/.../.../g synthaxes, I never manage to make a regexp replacement work the first time specially if I need group
 
@Kiwy you're using the wrong tools.
 
@Kiwy You really need to learn about regexps. You have not mastered them.
 
2:21 PM
@terdon why it is the wrong tool ?
it's pretty nice to make replacement in a big file
 
@Kiwy if you want to capture groups, use, for example: perl -pe 's/(foo)/$1 bar/' or sed -r 's/(foo)/\1 bar/' see? No escaping, no nothing!
 
@slm Not sure whats going on with that guys xmodmap, as far as I can tell, the commands hes using should work
 
Also, in vim you can choose whether to escape parentheses when using them for grouping or when matching against them. Again - learn the tool...
 
@terdon I try this $$$ so much time that I now use notepad++ instead each time I need something like this I end up looking how to substitue group using sed...
@JennyD Yes I certainlly do not master the gnu tools and specially VIM which one of the most complicated thing on earth but I never find a god doc about vim, (yes I know man is a good start)
 
Next time you're on vi and need regexps, type :help patterns
But if all you're doing is replacing stringa with stringb, then I'd use sed instead of firing up an editor.
 
2:28 PM
@JennyD Well editing replacing occurences, stuff like this :D
 
@Kiwy which is why I emphasized all :-)
 
vim isn't that complicated for basic use. What throws people off is the modal editing.
 
@JennyD I noticed ;-)
@casey what do you call modal editing ?
 
@Kiwy modal editing is what vi does. It has modes, normal mode, insert mode, visual mode, replace mode, etc
and each mode interprets your keystrokes differently
the "normal" editing you are used to is vi's insert mode
 
@casey OK, didn't know it was called like this
 
2:31 PM
but it doesnt start there
and you cant do anything but insert text (and move with arrows) in insert mode
if you know 'i' to get into insert mode, '<esc>' to get into normal mode,
 
The thing is that the regexes you know are Perl's. Many POSIX tools implement them as well
 
':wq' in normal mode to write and quit
you can get by with vi
 
@casey to me one of the biggest problem with vim is the fact that sometime over an ssh session it will not accept arrow and on another machine yes and then you find a very old vim using jklm
 
@Kiwy thats why some people will tell you best practice is to use hjkl instead of the arrows all the time
hjkl works on all vi, arrows not so
 
@terdon and that's the only good thing perl invent, the regexp :D
 
2:35 PM
oooh boy, you're on thin ice now
 
...
 
I have never seen any language that is half as good as perl at manipulating text
 
I was just going to say that @terdon
 
All you kids with your python obsessions...
 
@casey and if you're not used to using hjkl, go play nethack for a while until you've got it
@terdon Agreed
 
2:37 PM
@JennyD absolutely!
 
@terdon @casey perl might be very good at manipulating string that's true but the synthax of this language is hell
 
@Kiwy the syntax can be quote obfuscated, but you don't have to write perl that way
 
@Kiwy Compared to what? Serious question - what you feel is good or bad will be affected by what you first started with
 
@JennyD well python is quiet nice as it doesn't let you too much choice in formting things
 
You can write perl to be almost like English: perl -ne 'next if /foo/; next unless /bar/' for example
grrrr python, I hate the whitespace means something rule.
 
2:39 PM
@Kiwy Which is why I dislike python. Whitespace should be a formatting choice, not an information carrier.
 
matching a string is a pain, printing is a pain, you need a bloody IDE to be able to code it comfortably
 
And that's why I asked .
 
@terdon and that's exactly the problem, you can write perl whatever you want to do in dozen of different way the exact same thing
 
@Kiwy s/problem/great thing/
 
yea, if you are just matching text and acting on the matches, perl beats python any day
 
2:39 PM
@Kiwy exactly! That's not a problem, that's an asset! It frees you to write in the way you find most clear and comfortable
 
I use python for visualizations and not much else
 
I found that my perl got better and more readable once I had also worked a while with functional languages.
 
It is better for object oriented code, that's true. Perl's syntax there is annoying
 
@terdon and two sysadmin will wirte the same thing in two differents ways and when you will get the source 2 years after, you not be able to understand them because you do not write them
 
@terdon Oh yes, it's very noticeable that the OO was tacked on as an afterthought
 
2:41 PM
I do not critizise the functionnality but the way the langague is written
 
@Kiwy so? Just because someone can write bad code does not mean it's a bad language!
 
@Kiwy that is a complaint about programming in general, not any one language
 
@casey nod
 
You're just used to Python whose whole philosophy is: Do it the Right® way
 
If a language doesn't allow you to implement your solution in more than one way, then you're not working with a real language.
 
2:42 PM
That's silly, the philosophy should always be :do it YOUR way
 
@terdon and if you don't like it, then do it that way anyways and stop whining
 
Well the chance of having an understandable code in python (and I'm not a huge fan of it) is way more important than having a understandable perl code
 
Linux, open source == sinatra!
@Kiwy sorry, but you really have no idea what you're talking about
 
@Kiwy Obviously it's that way for you, since you know python better than perl. For me it's the other way around. You can't simply disregard the fact that knowledge of the language will always be required to understand what's written in it.
 
I have looked at Perl code before knowing perl and understood something of it. I find Python opaque, you need to know all sorts of things like the fact that . means "child of" for example
You're also forgetting that a LOT of python's syntax actually comes from Perl
The hello world is identical for example
In order to do just about anything in python you need a cryptic line : import sys. What the hell does that do? How is it clear?
 
2:45 PM
@JennyD fact is I have a very small knoledge in both, and I have a big knoledge in java and c and bad code can be written in any language for sure, but still python with this space mean something makes code more readable and way nicer to read years after, you can have great code in perl, but a language with too much freedom is in my opinion not to be used with other personns
 
@Kiwy You sound more and more like a windows person...
 
@terdon you should say that you can write a perl version of helloworld that look the same as the python one :P
 
and you throw virtualenv into the mix, the python PATH for finding modules, many places modules can be located on disk and it really can get quite confusing
 
@JennyD that hurt...
 
@Kiwy Or you could look up the history of both languages, showing that terdon is right.
 
2:47 PM
@Kiwy umm, considering that Perl must be about 20 years older, that's like saying your dad looks like you
 
@Kiwy Wanting to limit people's options on how to do things, so that there is only one possible way - not really the unix philosophy, is it?
 
What the hell does "too much" freedom even mean?
 
@JennyD its the python PEP8 philosiphy
 
@casey I think I'll stick with perl, thanks :-)
 
:)
 
2:49 PM
@JennyD the nux philosophy is about sharing, and it's always easier to share when you share some basic languge python makes code more uniform and though more sharable in my opinion and perl is in my opinion a language still very powerful but with so much cababilities that it make the codes a bit obscure
 
Huh, I was wrong, python is much older than I thought: 1991 (though still younger than Perl: 1987)
@Kiwy our point is that that is the fault of the coder, not the language. It's the same as speaking, I can say something in a clear or obfuscated way in English, that's not the language's problem but mine.
 
@Kiwy I think you're confusing Unix philosophy with FOSS philosophy. There is more to unix than that.
 
@terdon python just has poor adoption rates. Python 3.x was originally released in 2008 and most people it seems still use 2.7
 
huh
 
in fact, what little python code I maintain could only in the last year or so be updated to 3.x because the modules it depends on didnt support it
 
2:53 PM
Perfect: As Alex Martelli put it: "To describe something as clever is not considered a compliment in the Python culture."
See? Major difference in philosophy, perl hackers like clever
 
Speaking of crappy code, anybody else remember the perl scripts from Matt's script archive from the 90's? The web site is actually still around, and still eyebleedingly horrible
 
@JennyD nope. wassat?
 
I can't convince you andyou can't convince me, perl is powerful but being able to express in ten different way the same code is not a feature in my opinion but more a bug, I don't say pyhton is perfect but at least it tries to correct that.
 
@terdon For starters it had a formmail.pl that was loved by spammers all over the word...
It's horrible. Really horrible.
 
wow nice website oldfashion way :D
 
2:55 PM
It i want to remember bad perl code, I just go visit /. :)
 
I cleaned crap from that site off of our web servers far too often back when I worked at an ISP
 
@Kiwy python code can be expressed in multiple ways too. I can use at least 3 methods off the top of my head to process command line arguments for example
the whole matplotlib interface lets you do things matlab style or a more "pythonic" style
and plenty of other "I can do this more than one way" opportunities
e.g. I can use loops, inefficient styles code adapted from C, list comprehensions, etc to solve many problems
 
@casey I like list comprehensions. They are elegant. But I usually don't use them in perl since most of my coworkers aren't very good at perl so I need to write in a way that they can understand and maintain.
 
@JennyD agreed
 
3:11 PM
anyone see the question here? unix.stackexchange.com/review/reopen/40705
 
@casey the Common Lisp version is even more flexible. LOOP and so forth.
@casey congrats on 3k. when did that happen?
 
couple days ago. ive been slow lately
 
@Braiam what about it? More to the point, what question?
 
check it again
looks reasonable after you round some corners looking at Stephane answer
 
@FaheemMitha I never got the hang of Lisp.
 
3:18 PM
@JennyD It's easy. Probably easier than many other languages. With their weird and complex syntax rules.
@casey Ok.
C++ is particularly insane, for example.
 
@FaheemMitha I know enough to edit my .emacs and .gnus, otherwise I'm still at the Hello, World stage
 
@JennyD That's emacs lisp. Quite different.
 
@FaheemMitha True. I have looked at some common lisp, but remained at the HW stage their. I know a little more about emacs lisp, but not much.
 
@JennyD Ok. Common Lisp may be overkill for many things. I'm interested in it for scientific algorithmic work, which can get quite complex.
Though, really, it is quite a simple language with an undeservedly fearsome reputation, imo.
 
@FaheemMitha And which is far beyond me, which is part of the reason why I'm still a sysadmin instead of a developer
@FaheemMitha I have never been in a situation where it's been the natural language for me to do something in, so I've never spent the time
 
3:22 PM
Anyone know how to read a symlink? Not its target but the link itself, I'm wondering what the format is.
 
@JennyD Well, people say that. I don't find it so at all. The one problem I've had in my limited use of it, is that the syntax is so uniform it is difficult to keep track of where one is in the code. There are few sign posts like in more coventional languages.
 
@FaheemMitha No, I meant that scientific algorithmic work is far beyond me.
@terdon I've got that one by my desk :-)
 
@JennyD It's got some overhead, granted. But to some extent it is a matter of getting used to it. I guess most people never do. Actually, probably text parsing is one thing it is fairly natural for.
@JennyD Have you tried it? It is not a uniformly difficult area.
 
@FaheemMitha exactly - I've always had something to do in some language that felt more immediately attractive ...
 
3:25 PM
@Braiam I got it in the review queue, did not understand why it showed up there
 
@FaheemMitha No, I took only a very basic bit of university CS. I'm likely to be heading in that direction with my current employer, though.
 
@JennyD Right. That's probably a common experience.
@JennyD Ok.
 
@Anthon views
 
Anyway, people consistently say that it scales quite smoothly for difficult stuff, unlike most languages, where one eventually runs into some sort of wall. Given my limited experience, I can believe it.
Actually, python is in many ways quite similar to CL. It has much of the same feel, and even some of the same syntax.
If you lopped off the more powerful bits of CL and switched to infix syntax, you'd get something that would not be very different from Python.
 
pythons big shortfall is the global interpreter lock, and that removes a lot of utility potential for me
 
3:30 PM
@Braiam I never noticed the answer, sometimes I hate the way the review queue presents things (or maybe I should pay more attention to other things htan the question itself).
 
@casey Yes, the GIL is a major weakness.
 
@casey the GIL is a "feature" of CPython, if that is a problem use e.g. stackless.
 
Anyone know if there are any filesystems that allow hard linking to directories?
 
@terdon I think so, but I've always seen it come with big flashing neon warning signs.
 
@JennyD OSX's time machine allows them apparently
> To prevent endless recursion, most modern operating systems do not allow hard links on directories. In addition, hard links on directories would lead to inconsistency on parent directory entries. A notable exception to this is Mac OS X v10.5 (Leopard) and newer, which use hard links on directories for the Time Machine backup mechanism only. Symbolic links and NTFS junction points are generally used instead for this purpose.
 
3:36 PM
I have a faint memory of there being some other obscure FS that allows it too, but I can't recall which or if I have just imagined it.
 
Must be
Huh, I was going to answer that find files with newlines questions and suggest find -regex but I can't get it to match newlines!
 
@terdon I think you're supposed to use mount --bind instead
 
@JennyD I am not attempting to do this, just idle curiosity because I got annoyed that this (bad) answer is being upvoted.
 
@terdon It is a crappy answer. I don't have enough rep there to downvote it, though.
 
@terdon I take offense to the use of "folders" in that answer
None of my filesystems have folders!
 
3:41 PM
:)
So, anyone know how I can read a softlink? I mean the actual contents of the file that contain the target path?
 
@terdon readlink?
 
@JennyD that will print the target, I am guessing that symlinks are actually stored somewhere with content like "name=foo, target=bar" or will I have to read the filesystem to get to that?
 
@terdon, that's indeed a very interesting quesiton
 
I'm guessing it should be in the directory
 
@terdon It's a file containing a string with the target location. man 3p readlink will tell you more.
 
3:47 PM
@JennyD you mean man readlink? I get nothing for man 3 readlink
 
@terdon, both hardlinks and simlinks exist on NTFS to be posix compliance and a nice software allow easy creation of them
and shortcuts are really different I think
 
> Depending on the type of file an inode describes, the 60 bytes of storage in inode.i_block can be used in different ways. In general, regular files and directories will use it for file block indexing information, and special files will use it for special purposes.
 
@terdon There's the unix command readlink, and then there's the C function readlink contained in unistd.h . Where you find that man page would depend on your system, I suppose.
 
@terdon 3p, not 3
 
@MichaelMrozek tried both, both gave no manual entry for readlink in section
 
3:49 PM
@MichaelMrozek doesn't work either
 
@MichaelMrozek heh, thanks, simpler :)
 
@JennyD damn so I have to write a C program to get at this :(
 
4:02 PM
Here's a trivia question. Locate all mercurial repositories in a directory.
 
@terdon Unfortunately, yes. Or just use readlink, since all that inode contains is the path to the actual file, i.e. exactly what you will get out of a c program using that function.
 
huh, asked already. big surprise.
 
@JennyD ah, so no special formatting? Just simple text, cool then.
Anyone know if fdisk will print this error when run on a GPT disk?
 
16 mins ago, by Jenny D
@terdon It's a file containing a string with the target location. man 3p readlink will tell you more.
 
> Disk /dev/sdb doesn't contain a valid partition table
@JennyD fair enough, I thought it might have a more complex format.
 
4:08 PM
Home time. See you!
 
@JennyD waves
 
@terdon:
#include <unistd.h>
#include <stdio.h>

int main(int argc, char** argv)
{
char buffer[1024];
int datalen;

if (argc != 2)
{
printf("Usage: %s LINK\n", argv[0]);
return -1;
}

datalen = readlink(argv[1], buffer, 1024);
if (datalen == -1)
{
printf("Error accessing %s\n", argv[0]);
return -1;
}
buffer[datalen] = 0;

printf("Contents of the symlink %s: %s\n", argv[0], buffer);
return 0;
}
 
@Braiam the question as asked is a bit silly, but Stephane's answer is useful. I voted to reopen
the question could use a bit of editing
 
casey@convect code % ./read-a-link /home/casey/Research
Contents of the symlink ./read-a-link: /data/casey/Research
 
@casey thanks! :)
 
4:13 PM
@terdon depends. Some GPT disks also have a partition table (which doesn't cover the whole disk if it's >2TB)
 
np, and I just noticed a bug in the last printf, it should be argv[1]
so it gives you the name of the link and its target
rather than the name of the program
 
GUID Partition Table (GPT) is a standard for the layout of the partition table on a physical hard disk, using globally unique identifiers (GUID). Although it forms a part of the Unified Extensible Firmware Interface (UEFI) standard (Unified EFI Forum proposed replacement for the PC BIOS), it is also used on some BIOS systems because of the limitations of master boot record (MBR) partition tables, which use 32 bits for storing logical block addresses (LBA) and size information. , most current operating systems support GPT. Some, including OS X and Microsoft Windows, only support booting f...
/protective MBR
which is a convention, not a necessity
 
@terdon no problem
 
@Gilles I tried to fix it somehow adapting a fitting question to the answers
 
@Gilles thanks, I was wondering about an AU post whether that specific output would be returned if you run fdisk -l on a GPT disk
 
4:30 PM
Can anyone explain why this fails? touch "$(printf "a\nb")"; find . -regex '.\n.'
 
4:46 PM
@terdon You need the ./ at the beginning, but it still doesn't work with \n. The only way I could get to work is find . -regex './.'$'\n''.'
 
@MichaelMrozek good point about ./, yes. And I can't get it to work in any way either. I tried all of find's -regextype choices. Strange, I'm guessing it must read line by line
 
Well, it did work the one way I said, using $'\n' outside of the string. Kind of messy though
 
Yeah, sorry, I misread that as -name $'\n'
Strange though. That should not be needed in a regex
 
5:00 PM
Oh well, posted a Q.
 
5:20 PM
good afternoon
I expect I'm going to have to explain that comment I just left...
@terdon \n has no special meaning, even in Perl regular expressions. It does, however, have special meaning in interpolated perl strings, of which qr// is one type. Search for \n in man perlre... — derobert 1 min ago
 
@terdon there are many different regex syntaxes
5
Q: Why do some regex commands have opposite intepretations of '\' with various characters?

Cory KleinTake, for example, this command: find . -regex ".*\.\(cpp\|h\)" This will find all the .h and .cpp files in your directory. The period character '.' in regular expressions usually means "any character". To get it to match only an actual period, you must escape it using the backslash characte...

5
Q: What does \? mean in a regular expression?

user5997The following command is used to search for a 7-digit phone number: grep "[[:digit:]]\{3\}[ -]\?[[:digit:]]\{4\}" file What does \? stand for?

 
@Gilles I know, but I thought ERE had \n
 
@terdon No. Some tools with ERE also support \n and the like, but it's not part of ERE syntax.
 
That's what I didn't know, OK.
 
@terdon BTW: If you did ... if m/\n/ in Perl, it'll match. If you do ... if m'\n', it should not.
 
I haven't tested this, so I risk being made a fool of, but its what the docs say :-P
 
awk documentation mentioning \n et al as extras to ERE syntax: pubs.opengroup.org/onlinepubs/009695399/utilities/…
 
@Gilles thaht was my next question. Find claims to implement awk regex, and awk matches newlines:
terdon@oregano foo $ printf "f1l1\nhas newline:f2l1#f1l2 does not:f2l2#" | awk -F: 'BEGIN{RS="#"}; ($1~/\n/){print $1}'
f1l1
has newline
So, shouldn't find -regextype posix-awk .\n.` work?
@derobert thanks! Sneaky Perl!
 
@terdon BTW, if you look in perlop, a bunch of quotelike operators work like that.
 
@terdon I added these citations to my answer
@terdon I think it would have been a good idea, but it's probably too late now for backward compatibility
 
5:39 PM
@Gilles so find does not actually implement posix-awk?
 
Simple solution, find2perl :-P
 
@derobert cool! I did not know that one
 
It doesn't seem to support the -regex option, but it'll do the rest, and its trivial to modify the perl program it spits out
 
Ah, but does awk as opposed to gawk match newlines?
Does anyone have access to POSIX awk? Not gawk?
Both mawk and gawk regexes seem to match \n
 
@terdon Arguably yes, but it's complicated: the /.../ syntax of awk mixes string literal escapes and regex escapes (and likewise in perl)
 
5:44 PM
Oh, also, I bet you could (at least for interactive use) do that find by actually inserting a <NL> in your pattern. E.g., by typing ^V^J
 
@derobert I guess, I am just wondering about the details of the -regex option. I don't know if POSIX-awk matches \n, only that gawk and mawk do.
@Gilles so you're saying that \n is handled behind the scenes by the interpreter and not by the regex handler?
 
@terdon I haven't looked at the source to each utility to see what they're doing, but in general, yes, that's what I'd expect
 
@Gilles is busybox awk == POSIX awk?
 
@terdon I don't know, I think it tries to conform (if you enable enough features at compile time), but of course it could be buggy, or it could deviate from the standard in order to keep the code small
 
@Gilles I just tried and busybox awk also matches newlines, that's why I'm asking. But if this is a feature of the awk interpreter and not of it's regex language, that would be expected.
@derobert @Gilles thank you both, it's much clearer now.
 
5:55 PM
@terdon it's a feature of the awk language, and one that's included in the POSIX spec
 
/\n/ matches a newline. How it's done internally is implementation-dependent.
 
And as such, separate from its regex language. Gotcha.
 
@Braiam meh, speed isn't the main problem with svn
 
yeh, the problem with svn is svn
2
 

« first day (1248 days earlier)      last day (3700 days later) »