Common Options

source of randon data

/dev/random and /dev/urandom can be used as the random number generator --random-source=FILE

Traversing Symbolic links

chown, chgrp, and chmod all accept -R or the recursive option, which means do this command for all directories underneath what this command specifies


ls ~/Documents/


chown joshua ~/documents/

find

find and report all php files in the specified dir current directory

#+BEGIN_SRC sh :dir ~/programming/waypoint/nutripledge/ find . -name '.*php' -print #+END_SRC

#+RESULTS: | ./calendar1/__MACOSX/supercali-1.0.8/._modules.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._week.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._year.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._upcoming.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._grid.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._weeklist.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._day_week_functions.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._quarter.php | | ./calendar1/__MACOSX/supercali-1.0.8/modules/._day.php | | ./calendar1/__MACOSX/supercali-1.0.8/._edit_users.php | | ./calendar1/__MACOSX/supercali-1.0.8/._delete_event.php | | ./calendar1/__MACOSX/supercali-1.0.8/._admin_actions.php | | ./calendar1/__MACOSX/supercali-1.0.8/._edit_links.php | | ./calendar1/__MACOSX/supercali-1.0.8/._bad_password.php | | ./calendar1/__MACOSX/supercali-1.0.8/._login.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._notify.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._session_start.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._nav.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._start.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._footer.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._top_nav.php | | ./calendar1/__MACOSX/supercali-1.0.8/includes/._hidden_fields.php | | ./calendar1/__MACOSX/supercali-1.0.8/._edit_categories.php | | ./calendar1/__MACOSX/supercali-1.0.8/._upload_events.php | | ./calendar1/__MACOSX/supercali-1.0.8/._edit_groups.php | | ./calendar1/__MACOSX/supercali-1.0.8/._edit_event.php | | ./calendar1/__MACOSX/supercali-1.0.8/._user_profile.php | | ./calendar1/__MACOSX/supercali-1.0.8/lang/._en_us.php |

To find all of the specified files in all of the subdirectories starting with dir

#+BEGIN_SRC sh :dir ~/programming/waypoint/henriott-group/ find . -name \*.php -print #+END_SRC

To find all of the selected files (including sub-directories) and then delete them

#+BEGIN_SRC sh :dir ~/programming/waypoint/nutripledge/ find . -name \*.pdf -exec /bin/rm {} \; #+END_SRC

#+RESULTS:

finding multiple file names


find . -name '*png' -o -name '*svg' -o -name '*jpg'

./donate/foss_factory/logo.png

./hurd/status/hurd-fvwm-screenshot-2009-11-12.png

./hurd/status/hurd-iceweasel-screenshot-2012-03-21.png

./logo/boxes-redrawn.png

./logo/boxes-redrawn.svg

./microkernel/OS-structure.svg

./open_issues/c42ea6c9-f574-4b33-8440-f846b05856de.png

./unsorted/FunnyHurd/HurdCarDeal.jpg

./unsorted/FunnyHurd/HurdMagician.jpg

./unsorted/GrantBowHurdPage/diagram.png

./unsorted/JoachimNilssonHurdPage/patch_kit.jpg

./unsorted/SeenHurd/lmf10_1999.jpg

./unsorted/TestWebMenu/lmf09_1999.jpg

./unsorted/WebHome/hurd_sm_mf.png

./user/jkoenig/gsoc2011_classes.png

grep/egrep/fgrep

grep searching a list of file/s for a pattern and outputs the matching lines. Grep only supports a handful of regular expressions. If you want to use ALL of the regexp expression goodies, you need to use egrep. fgrep is fast grep. It interprets the string as a literal string and not a regexp.

There is no way to match newline characters with grep. info:grep#Introduction #+BEGIN_SRC grep OPTIONS PATTERN [INPUT_FILE_NAMES | - ] #+END_SRC

The "-" for grep means standard input.

ie:

#+BEGIN_SRC sh :dir ~/ grep -rc th | grep -m 5 Sass - #+END_SRC

#+RESULTS: | .gem/ruby/2.3.0/doc/sass-3.4.21/ri/Sass/Tree/MediaNode/value-i.ri:1 | | .gem/ruby/2.3.0/doc/sass-3.4.21/ri/Sass/Tree/MediaNode/resolved_value-i.ri:1 | | .gem/ruby/2.3.0/doc/sass-3.4.21/ri/Sass/Tree/MediaNode/new-c.ri:1 | | .gem/ruby/2.3.0/doc/sass-3.4.21/ri/Sass/Tree/MediaNode/name-i.ri:1 | | .gem/ruby/2.3.0/doc/sass-3.4.21/ri/Sass/Tree/MediaNode/query-i.ri:1 |

If you want to search for a regexp in the current directory and all sub-directories

#+BEGIN_SRC grep -lr "" #+END_SRC

command options

-e PATTERN

lets you specify a pattern by regexp, and you can specify multiple patters!


  grep -e be -e lint unofficial-packages.txt

bemenu-git r223.c3abc43-1
csslint 0.10.0-1

So I can see "be"menu and css"lint", but you could do the same thing with


egrep "be|lint" unofficial-packages.txt

bemenu-git r223.c3abc43-1
csslint 0.10.0-1

Note that the e flag should be given last on the flag list. If it is given first, then the rest of your flags will be thought of as the argument. ie:


  grep -ei -

-r search recursively, ignore symlinks

--exclude-dir=GLOB

--exclude-file=GLOB

--include=GLOB

-i ignore case

-R searches recursively following all symlinks exclude any dirs that match the GLOB The GLOB can use "?,*[...]" to match. I don't believe that you can use a regexp expression though exclude any files that match the pattern. The GLOB can use "?,*[...]" to match. I don't believe that you can use a regexp expression though include any files that match the pattern. The GLOB can use "?,*[...]" to match. I don't believe that you can use a regexp expression though

This lets you say that you want to ignore case in your grep search


grep -i soups cooking.org

* Soups
** Do not push cut garlic into soups. EVER.

You can see that it changed my pattern into [Ss]oups

-w word regexp


grep -i soup cooking.org

* Soups
** Do not push cut garlic into soups. EVER.
   You will ruin the soup.  It will be way too garlicky!

Suppose that we do not only the word "soup" and not to match "soups." or "Soups". We could do this:


grep -i "\bsoup\b" cooking.org

   You will ruin the soup.  It will be way too garlicky!

or to save some time, we could do this:


grep -iw soup cooking.org

   You will ruin the soup.  It will be way too garlicky!

-x line regexp

-c count matching lines

This just says that the regexp should match the whole line. So it's like saying "^" PATTERN "$" output the total count the matching lines in the file


  grep -c the hurd.org


  grep -c the pacman.org practical-gnus.org

-l print files that match the pattern

-L print files that do not match the pattern

-m stop reading a file after Max matches have been found

pacman.org:4
practical-gnus.org:65


grep -nm 6 -e the pacman.org practical-gnus.org

pacman.org:15:*** Failed to commit transaction package: path/to/file exists in the filesystem pacman.org:21: The problem is usually trivial to solve. A safe way is to first check if another package owns the file (pacman -Qo /path/to/file). If the file is owned by another package, file a bug report. If the file is not owned by another package, rename the file which 'exists in filesystem' and re-issue the update command. If all goes well, the file may then be removed. pacman.org:25: Every installed package provides a /var/lib/pacman/local/$package-$version/files file that contains metadata about this package. If this file gets corrupted, is empty or goes missing, it results in file exists in filesystem errors when trying to update the package. Such an error usually concerns only one package. Instead of manually renaming and later removing all the files that belong to the package in question, you may exceptionally run pacman -S --force $package to force pacman to overwrite these files. pacman.org:30:¹ The Official name is Arch Linux, but I prefer to call it Arch GNU/Linux, to respect the hard word that GNU put into creating the GNU/Linux operating system. For more information, you can check out: What's in a Name? and GNU/Linux FAQ. practical-gnus.org:18:Copyright: This work is licensed under the [[http://creativecommons.org/licenses/by-nc-nd/3.0][Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License]]. practical-gnus.org:23:People think it is hard because they get lost in its features. It would be much easier if they learned the essential 5% features and ignore the remaining 95%. practical-gnus.org:32:- Usable when offline. Check the "offline" part in the section "Advanced tips" practical-gnus.org:33:- powerful combined with other plugins like yasnippet practical-gnus.org:37:- "C" means "Ctrl" and "M" means "Alt". For example, "M-x" means pressing "Alt" and "X" together. practical-gnus.org:41:- "Group buffer" means the list of folders.

-A (after)

-B (before)

-C (context)

-H with filename

-h no filename

egrep

permissions

print NUMber of lines after the match print NUM of lines before the match print NUM of lines before and after. egrep is extended regexp In Unix everything is a file. When you write a text document called "text.text", that is a Unix file. When you create a directory (Windows users call this a folder) called "Music", you also are creating a Unix file. Obviously, "text.text" and "Music" behave very differently. One has your favorite songs, and the other is a text document, BUT the linux kernel assumes that they are both Unix files. All unix files have read, write, and execute permissions specific to each unix file. This typiccally means that the ordinary user joshua cannot read, write, or execute anything in the /root/ folder.


read     = 4
write    = 2
execute  = 1
owner    = read | write = 6
group    = read         = 4
other    = read         = 4
The basic syntax of the command to set the mode is

chmod 644 [file name]
In C, that would be

#include <sys/stat.h>

chmod("[file name]", 0644);
 #+END_SRC

#+RESULTS:
#+BEGIN_LaTeX
read     = 4
write    = 2
execute  = 1
owner    = read | write = 6
group    = read         = 4
other    = read         = 4
The basic syntax of the command to set the mode is

chmod 644 [file name]
In C, that would be

#include <sys/stat.h>

chmod("[file name]", 0644);

chmod

chmod [-R ] [file | directory]

Change a file's modifications. This specifies if what user, group, and if everyone else can read, write, or execute the file.

Permissions can modify the permissions of the user's, the group's and all others.

chown

chown [-R] [file | directory]

Change the file's owner. The file's owner is typically root or an ordinary user.

Examples include

chgrp

chgrp [ -R ] [file | directory]

Change the group that the file or directory belongs to.

basic usage of the find command in GNU

#+BEGIN_SRC sh find . -name "pattern" -print

find . -name "this-file.*" -print

find . -name "this-file.*txt" -print

touch -amc filename updates the files timestamps

join

find . -name "this-file\.txt" -print #+END_SRC Access Modify Change

join [OPTIONS] [FILE1 | - ] [ FILE2 | - ]

A way of combining two files based on a common column.

available options

Either file1 or file2 can be "-", which is standard input. So that is pretty cool.

-a FILE-NUMBER

print a line for each unpairable line.

-i ignore case differances

-1 FIELD_NUMBER

Join on the nth column. For example is the following line =this is an example line=

"an" is on the 3rd field.

-2 FIELD_NUMBER

Join on the nth column. For example is the following line =this is an example line=

"example" is on the 4th field.

JOIN in action


join -1 1 file1.txt file2.txt

1 12 18 this is 2 11 17 an interesting 3 10 16 experiment isn't 4 9 15 it? 5 e e 2nd field match 6 7 13 7 6 12 8 5 11 9 f f 2nd field match 10 3 9 11 2 8 12 1 7


join -1 2 file1.txt file2.txt

xargs

lets you get stdin and feed it to other programs that otherwise would not like it, allegedly


ls *org | xargs -i echo the{}et

theaccrisoft.orget theapache.orget thearch.orget theauto-complete.orget thebash.orget thebbdb.orget thecore-utils.orget thedired.orget thedisk.de.orget theelisp.orget theemacs.orget theevil-dvorak.orget theevil.orget thefirefox.orget thegawk.orget thegimp.orget thegit.orget thegnuplot.orget thegnus.orget thegpg.orget thehurd.orget thehurd-rpc.orget thehurd-talk.orget theinetutils.orget theiptables.orget thejavascript.orget thejson.orget theledger.orget thelinux-kernel-coding-style.orget themach-paging-interface.orget themagit.orget themaking-money-with-free-software.orget theman.orget themysql.orget thenetworks-networking-internet-ip4-tcp-udp.orget theorg-babel-code-block-args.orget theorg-babel-intro.orget theorg-babel-sql.orget theorg.orget thepacman.orget thephp.orget thepixi.orget thepractical-gnus.orget thesed.orget thesql.orget thesystemd.orget thethundertalks.orget thetoward-a-new-strategy-of-os-design.orget thetranslator-primer.orget thetranslators.orget thewhm.orget thewordpress.orget


ls *org | xargs -n 3

4yearsOfLife.org 4yearsOfLife.org.org a-certain-matchbox-member.org
a-girl-hitting-a-man-is-hypocritical.org approaching-and-meeting-people.org cooking.org
cover-letter.org example-latex.org gnu-assign-copyright-hurd-email-assign.org
how-to-succeed-in-the-army.org intro-to-gnu-linux.org letter-to-dad.org
letterToNicole.org letter-to-vincent.org libre-software-response.org
professor-snape-rachel-sweet-bread.org references.org romance-story.org
soihub-pages.org university-place.org why-proprietary-software-is-wrong.org

show available disk space

sort

options

-f ignore case

-g general numeric sort

-h human numeric sort

sort [OPTION] [FILE] This is good for sorting

-M month sort

-n numeric sort

-r reverse sort

-R random sort

-t SEPARATOR

How I got my local apache server working again:

50M -10G 1.5K use SEPARATOR as the field separator. By default that separator is a space. (info:coreutils#sort invocation)

For some reason, my ~/ folder had permissions drwxrws---

To fix it I did

#+BEGIN_SRC bash :exports both cd /home/; chmod g-s joshua; ls -lh | grep joshua

#+END_SRC