Summary Up to Shells

Unix Structure – Kernel + Shell

Linux, GNU Project

File structure - inodes

Standard codes:

intr	^c	Stops a program
erase	^h	backspaces
werase	^w	erases last word typed
kill	^u	kills the current input line
quit	^\	stops the program and saves core in a file
stop	^s	pause screen output
start	^q	resumes screen output
eof	^d	no more data
suspend	^z	temporarily stops (i.e., suspends) a program
resume	^y	resumes running a program

Unix commands

· Navigation: cd (. Is this folder, .. is up a folder, ~ is home)

· info: date, who, who am I, whoami, man, pwd

· file/ dir control: mv, cp, rm, chmod, ln, rmdir, mkdir, chown, chgrp, tar

· display: cat, more, pr, echo, od, du, whereis, which

· file info: find, ls, wc

· filters: egrep, sort, tail, diff, sed

· processes: ps, fg, kill, wait, (use & at end of command to send back)

Unix Metacharacters

· * - match anything

· ? – match any one character

· \ - escape to convert metacharacter to plain character

Redirect

Ø > - create file output

· >> append file output

· < file input

· 2> - error output redirect

· | pipe the output into the next command

· tee – write the current input to a file

·

Use vi, ed

Ø Input mode via a, o, i

Ø Command mode via esc

o Change text:

§ x – delete a charcter

§ r – replace a character

§ R – replace a word

§ J – join

§ ~ - upper/lower

§ d – delete (and put in buffer)

§ p – put what is in buffer #

§ = - indent according to c program rules in a c program (follow by G to indent to end)

o Fill Buffer :

§ y – yank

§ c – cut

o Movement:

§ G – end

§ gg – top

§ line – move that many lines forward

§ h or j or k or l – move cursor

§ 0 –beginning of line

§ $ - end of line

§ / - find the string

o : commands: Structure is line start, line end, command (other structure exists too)

§ s – substitute one line

§ %s – substitute many lines

§ w – write (if filename write to that one)

§ q – quit

§ ! – don’t give me warnings

§ p - print

Regular Expressions

Atoms

Ø Single characters

Ø Dots – any character except new line

Ø Classes [ ] – any one inside or [^ ] – not any one inside

Ø Anchors – where in text - ^ begin line, $ end line, < begin word, > end word

Ø Back references - \# matched text from a subset ()

Operators

Ø Sequence

Ø Alternation | this or that

Ø Repetition: * (0 or more); + (1 or more); ? (0 or 1); {start,end}

Ø Group: () subset for backreference; repeat group sequence

Ø Save

Use with Grep

Ø Searches for the regular expression in a file or from a piped output

Ø Shows only matched lines

Ø Options: -v reverse (not found match); -n line numbers; -i ignore case; -c count lines

Use with SED

Ø Edits a stream with essentially vi commands

Ø -i to edit file contents

Ø -r to use extended regular expression syntax

Ø -f execute the sed commands in the script, piping each output into the next line

Ø -e sequence of commands with –e before each

Ø -n suppress all output (so that you can specify printing with p in sed command)

Ø { ; ; } set of commands to run when address matched (a little like –e)

Ø Commands

o Most commands that are available in vi

o x – swap pattern space and hold buffer

o = line number

o c – change a cline

o a – append after cursor

o i - insert

o d – delete

o p – print

o s – substitute

o y- transform (character set mapping)

Filters :

sort – sort all input in order by some key :

· -f fold upper and lower case together

· -n numeric order instead of text order

· -k# column number to sort by

· Example: sort –nk2 file1 file2

· Remember it will sort as many files as you give so: sort filea fileb filec puts all 3 files together in a sorted list

uniq - deal with duplicate lines :

· -c tells how many times a line is repeated

· -d print only lines that are duplicates of the lines above them

· -u prints only lines that are not the duplicate of the line above them

· Example: sort file1 file2 | uniq -d

· Remember to sort when you want duplicates anywhere within the list considered

comm – compare files listing files in columns next to each other

if given 2 sorted lists, it will create 3 columns: (Use –column number to suppress a particular column)

· 1 : only in the first file given

· 2 : only in the second file given

· 3 : those in both files

· Example: comm -3 <(sort -n csc270file) <(sort -n csc271file)

tr – translates characters from one given set to another set

· tr a-z A-Z file : makes uppercase

· tr ab 12 file : changes all a’s to 1 and b’s to 2

· tr –c ab 1 file : changes everything but a and b to 1

· tr –s ab 12 file : changes all a, aa or a…a to 1 and b,bb or b..b to 2

Finding files:

· Find a file in a a subdirectory:

o du -a | grep yourfile

o find . -name *yourfile*

· Only list files in current dir

o find . -maxdepth 1 -type f

o ls -p | grep -v / | column

§ p writes a slash after all filenames

· Only list subdirs in current dir

o ls -ld */