jlevy / the-art-of-command-line Goto Github PK

I think having a section explaining the basic usage of man pages would be good.

Stuff like usage, topics, searching, etc

Learn at least one text-based editor well. Ideally Vim (vi), as there's really no competition for random editing in a terminal (even if you use Emacs, a big IDE, or a modern hipster editor most of the time).

Specifically...

as there's really no competition for random editing in a terminal (even if you use Emacs

This simply isn't true, it should be redacted.

I suggest something a little more balanced.

Learn at least one text-based editor well. Ideally Vi/Vim or Emacs, so you can immediately use your editor from the a terminal. Be aware the Vi/Vim will almost always be part of any standard *nix distribution.

The tee command is quite useful if you want to watch the output and write it to a logfile at the same time.

http://linux.die.net/man/1/uptime

I propose removing fortune, ddate, and sl commands due to their limited usefulness, and having nothing to do with the ideology of The Art of Command Line.

Discussed at #82

Probably uname -a. Perhaps near lsb_release.

In "Processing files and data":

Know basic awk and sed for simple data munging. For example, summing all numbers in the third column of a text file: awk '{ x += $3 } END { print x }'. This is probably 3X faster and 3X shorter than equivalent Python.

In "One-liners":

• Summing all numbers in the third column of a text file (this is probably 3X faster and 3X less code than equivalent Python):

      awk '{ x += $3 } END { print x }' myfile

There seems to be no mention of scp or rsync, which I use on a fairly regular basis. Would it make sense to add it in here?

When typesetting mixed English-Chinese text, it's suggested to add space between English and Chinese words. For example, 使用 Github 来做 instead of 使用Github来做.

Reference: http://www.zhihu.com/question/19587406

While not the focus of the guide, there are a handful of macos commands that should be added in their own section:

• open and open -a to open desktop apps
• pbcopy and pbpaste to copy and paste values into other apps
• mdls to list metadata

Others?

consider adding https://github.com/soxofaan/duviz

(disclaimer: I'm the author of that tool)

su: http://linux.die.net/man/1/su

sudo: http://linux.die.net/man/8/sudo

At the top of every day use there's an example of several keyboard shortcuts for deleting words and erasing the line. That's in the emacs cli mode, which is the default, but you can also use the vi mode. This allows you to use the vi macros for navigation, which can be much more fluid when the user is accustomed to vi or vim. Just an idea!

First i don't want to advertise my blog here.

I'm creating issue just because i feel my notes helps some peoples.
If you still thinking i'm advertise here feel free to close/delete this issue.

Before few days, I tried everyone to explain some basics linux like cp, mv and from where to get help, how to use vim etc.

I thought to create my blog and publish all my notes so i've to just give them url when they want some help.

Newbie Guide - Linux Basics

Linux Ideas and History
Linux Usage Basics
Running Commands and Getting Help
Browsing the Filesystem
Users Groups and Permissions
Bash & Shell Scripting
Standard IO & Pipes
Text Processing Tools
VIM - An Advance Text Editor
Basic System Configuration Tools
Investigating and Managing Processes
Configuring the Bash Shell
Finding and Processing Files
Network Clients
Advanced Topics in Users Groups and Permissions
The Linux Filesystem
Package Management
System Logs and SELinux
System Initialization
Deadly Linux Commands

All above topics are covered at https://miteshshah.github.io/linux/basics/

https://github.com/Xfennec/cv

And merge/combine with seq, probably in main section, not obscure section.

http://linux.die.net/man/1/apg

Piping is used in many of the examples, but I don't think it's ever explained.

glances is a very nifty tool that incorporates many other tools like htop, network status, sensor status, etc. into one simple and powerful utility. I'd think it'd be a good addition to this guide!

I'm not sure that it's such a great idea to set StrictHostKeyChecking=no to someone's ssh config without them knowing what that means?

Perhaps something like limiting it to only local subnets?

Machine hogged down with disk I/O, which process is responsible?
Often it won't show up high in top because it will be mainly waiting for I/O and therefore not consuming many CPU cycles.

iotop shows which processes are using disk I/O the most.

As the guide is so Bash-centric I suggest mentioning the builtin "help" command, as it can be the only source of extra info for some commands (other builtins)

For example, disown is mentioned so I tried to find out more about it

$ man disown
No manual entry for disown
$ disown --help
bash: disown: --: invalid option
disown: usage: disown [-h] [-ar] [jobspec ...]
$ disown -h
bash: disown: current: no such job
$ help disown
disown: disown [-h] [-ar] [jobspec ...]
    Remove jobs from current shell.

    Removes each JOBSPEC argument from the table of active jobs.  Without
    any JOBSPECs, the shell uses its notion of the current job.

    Options:
      -a    remove all jobs if JOBSPEC is not supplied
      -h    mark each JOBSPEC so that SIGHUP is not sent to the job if the
        shell receives a SIGHUP
      -r    remove only running jobs

    Exit Status:
    Returns success unless an invalid option or JOBSPEC is given.

You mention du -sk, as a human I prefer du -sh.

👍 Oh, and I love your write-up! ...still finding things to learn.

"man ascii" is mentioned, for western europe "man latin1" is also very helpful.

What it does (excerpt from man comm):

NAME
       comm - compare two sorted files line by line

SYNOPSIS
       comm [OPTION]... FILE1 FILE2

DESCRIPTION
       Compare sorted files FILE1 and FILE2 line by line.

       With  no  options,  produce three-column output.  Column one contains lines unique to FILE1, column two
       contains lines unique to FILE2, and column three contains lines common to both files.

       -1     suppress column 1 (lines unique to FILE1)

       -2     suppress column 2 (lines unique to FILE2)

       -3     suppress column 3 (lines that appear in both files)

I'm willing to write a piece on that if it is needed. Let me know if required.
Aliases, keybindings, custom Prompt Statement can be mentioned briefly.

I'm willing to do this if it is needed 😃

In section Basics:

- Learn about file glob expansion with * (and perhaps ? and {...})
+ Learn about file glob expansion with * (and perhaps ? and [...])

{} is for grouping commands.
[] is for file glob expansion.

There are a few of these. We should consider which is the best recommendation, perhaps listing all the main ones.

https://github.com/clvv/fasd
https://github.com/rupa/z
https://github.com/joelthelion/autojump

Any others?

C-a and C-w are emacs commands, and there are even more! One can use C-e to move to the end of the line and C-b and C-f to move one character. But one of the most useful is C-p to move up one line to the previous command (and analougue to that there is C-n to move down a line). To me it is more convenient than using the up-arrow key.

file, type, whereis, whatis and apropos were very useful to me while exploring the linux system.

ls | wc -l is useful for counting files in a directory.

Great stuff you've got so far, I will link people to this document when they want to learn the shell!

Hey,
Do not you mind I'll make russian translation of this document?
Will you accept my pull-request with README-ru.md file?

Please try to distinguish between normal, unixy key bindings (like ^U), and editor-specific key bindings, like ^K, ^A, and ^E. There is little point to learning emacs key bindings for shellish things, when vi knowledge is another heavy prerequisite for modern administration.

Converts text files between DOS and Unix formats.
http://manpages.ubuntu.com/manpages/raring/man1/fromdos.1.html

We have some examples of search/replace, but they are too simple.

• Should show a capturing group, e.g. sed with \1 in it and/or perl or repren alternatives.
• Maybe even one with an arbitrary transform on the group (like upper- or lower-casing).
• Another with multiline search replace, e.g. "foo,bar,baz" -> "foo\nbar\nbaz"

pv, the pipe viewer
http://www.ivarch.com/programs/pv.shtml

In Basics section.

tail -f (or even better, less +F),
tail -f -F (or even better, tail -f -F),

Better than tail -f is also tail -f -F which gives a perfect way to deal with long tail -f regardless of logrotate mechanisms. It avoid to be surprised by rotations logs, ie, seeing no more log lines while there are ones, and allow long time tails. A must, once tasted.

Command tail follows several files permanently. Command less +F does not, one has to go to next file. Not better than less in that case.

see http://linux.die.net/man/1/bash REDIRECTION section.

The general format for appending output is:
[n]>>word

The format for appending standard output and standard error is:
&>>word

This is semantically equivalent to
>>word 2>&1

Here Documents, Here Strings ...

https://github.com/junegunn/fzf

You didn't recommend the best ;-) terminal multiplexer: tmux!

See discussion at https://news.ycombinator.com/item?id=9722587

The problem with (cd /some/other/dir; other-command) is that cd may fail and other-command that may be potentially destructive will still get executed in the original directory. With && when the first command fails, the second is not executed.

Consider most important aspects of aliases to include. See also #60.

Add a link to http://explainshell.com/ early on (abstract, or Basics), and possibly link every command example to its explanation (ls -lart ~).

The current suggestion for replacing text in a file is to use perl, e.g.:

perl -pi.bak -e 's/old-string/new-string/g' my-files-*.txt

When I need this, I use sed -i, e.g.:

sed -i.bak 's/old-string/new-string/g' my-files-*.txt

Is there a reason users should prefer perl? Could that be documented (at least in this issue, if not in the document itself)?

ctrl-u deletes from the cursor position to the beginning of the line. (I think that's a useful distinction versus "the whole line".)

Nice write-up.

diff is great to not only compare files but also entire directories.

It could help to also mention patch and xdelta.