UNIX 命令

cut	截取的方式有三種，一是按照字符位置，二是按照字節位置，三是使用一個分隔符將一行分割成多個field，並提取指定的fields。 cut命令有5個參數，其中-c,-b,-f分別表示"character", "byte"以及"field"截取方式。當採用field模式截取時，需要用"-d"參數指定一個分隔符，分割符只能爲單個字符。另外還有一個"- s"，suppress，表示如果行中沒有給出的分割符則不輸出該行 echo Long, long ago | cut -f 2,3 -d " " echo hello, world | cut -c 8-12
paste	paste -d -s -file1 file2     -d 指定不同於空格或tab鍵的域分隔符。例如用@分隔域，使用- d @。     -s 將每個文件合併成行而不是按行粘貼, 多行轉成一行。 paste file1 file2 paste –d : file1 file2 paste -s pas1 pas2 （把第一個文件的多行轉成一行作爲第一行顯示，把第二個文件的多行轉成一行作爲第二行顯示） paste –s file1 (把文件中的多行轉成一行) ls /etc | paste –d " "  （把一行的內容按指定分隔符放到一列顯示）
cal	cal 2004 cal 04 2004
who	who whoami who am i
date	date +%Y%m%d date +%Y%m%d.%H%M%S
clear
tty	Display the name of terminal one is using
uname	uname -a （顯示unix版本號）
more	Display each file pagewise. On pressing enter key, the 2nd file gets displayed.   more file1 file2
stty	change and display terminal settings
bc	open arbitrary precision calculator
printf
type	display description and location of a command type cat type vi
which	just display the location of the command. which cat which vi
echo	echo * | wc  all files and directories in current directory. echo */ | wc   all directories in current directory. echo *.* | wc   all files which have extensions in current directory --above commands don’t count hidden files.
ls	-a, -l, -i, -r, -t, -s, -u(display and sort by atime instead of mtime)
cp	-r -i -f -a -U
mv	move the contents of file1 to file2. The file1 is deleted.
pwd	pwd -P
cd	cd .. cd test cd   (change to user’s directory) cd ~ cd ~tmapbatch cd -
mkdir	mkdir –p /test/test1/test2 mkdir –m 711 test
rmdir	-r -i -f
rm	Removal of a file needs write permission in its directory. -r -f -i
ln	(hardware link, Symbolic link & Soft Link) hardware link : ln cprog1 cprog1.bat               have same i-node numbers symbolic link : ln -s cprog2 /tmp/cprog2.bat  have diff i-node numbers   soft link can be used to link file on a different file system soft link can be used to link directories.
touch	可以修改atime, mtime touch –a file1  (update access time by current datetime) touch –m file1 (update modification time by current datetime) touch –t [[cc]yy]mmddhhmm[.ss] file1 (update both atime and mtime by specified datetime)
basename	basename /home/dpmbatch/andytest/1.txt basename $0 .ksh (ignore the suffix)
dirname	dirname /home/dpmbatch/andytest/1.txt
input 0 output 1 error 2	ls 1>infy   is as same as  ls > infy ls abcd 2>errors cat errors
tr	主要用於刪除文件中控制字符或進行字符轉換 tr -s -c -d ["string1"] ["string2"] input_file     -c    用string1的字符集的補集替換此string1，要求字符集爲ASCII     -d    刪除string中所有輸入字符     -s    刪除所有重複出現字符序列，只保留一個，即將重複出現字符串壓縮爲一個字符串   tr -s "[a-z]" < oops.txt  or cat oops.txt | tr -s "[a-z]"   tr  -s “ ” “:” < oops.txt   \n -- \012 \t -- \011 \r -- \015   刪除空行 tr -s "[\012]" < testing > testing1 && mv testing1 testing   小寫到大寫 cat testing | tr "[a-z]" "[A-Z]" tr "[a-z]" "[A-Z]" < testing   刪除指定字符 刪除不是字母的內容，步驟爲先用換行符替代全部非字母的內容，然後刪除重複的換行符。其中，替換用-c，刪除重複用-s tr -cs "[a-z][A-Z]" "[\012*]" < diary.txt   匹配多個字符 tr "[0*4]" "*" < hdisk.txt   把4個0替換成4個*
*, ?, [], [!]	ls [!ab]*
regular expressions	[^a-z] ab\{2,4\}
cmp	it displays the byte and the line number of the first difference. -l display for each difference -s return 0 if same, 1 if different, 2 if file is inaccessible.
sort	-d dictionary order sort -n arithmetic/numeric order -r reverse order -k n Sorts on field n -b ignore leading blanks -t set the delimiter   sort –t ‘:’ –k 3 file1
comm	compare 2 sorted files. return three-columnar output first column contains unique to the first file second column contains unique to the second file third column common to both files.   -1 does not print column 1 -2 does not print column 2 -3 does not print column 3
head	head -5 file1
tail	tail -5 file1 tail -f file1 參數-f使tail不停地去讀最新的內容，這樣有實時監視的效果
cat	-A -n
nl	nl file1
tac
more
less
od
wc	-l -c -w
diff	reports the difference between files. -b ignore trailing whitespaces and compare other strings of space characters.   diff file1 file2
uniq	to remove adjacent repeated lines -u  display only the lines not repeated in the file named input -d  display only the lines that are repeated in the file input -c  precede each line displayed by the number of times it occurs. +n ignore the first n characters including whitespace -m ignore the first m fields and any blanks.   uniq –u file1
	tr -cs 'A-Za-z' '[\n*]' < test2 | tr 'A-Z' 'a-z' | sort | uniq -c | sort -nr | head -20
grep	-c report only the number of matching lines -l list only the names of files containing pattern. -v display all lines except those containing pattern.
umask
chown	chmod user1 file1
chmod	chmod ugo+rwx file1
chgrp
default permission	file 666 – 022 = 644   directory 777-022 = 655   if read permission is off, you cannot use ls; if execute permission is off, you cannot cd into direcotry
vi	/pattern ?pattern :s/ptn1/ptn2  (first occurrence) :s/ptn1/ptn2/g (all occurrence in current line) :m, n s/ptn1/ptn2/g (all occurrence from line m to n) :., $ s/ptn1/ptn2/g (all occurrence from current line to end of file)
sed	sed '/^$/d' hello sed '/^ *$/d' hello sed '/^[ ]*[\t]*$/d' hello or sed '/^[ ]*[              ]*$/d' hello sed -n '1,2p' hello sed -n '3,$p' hello
sendmail	SUBJ="Workflow ${WORKFLOW_NAME} failed!"         TO=""           (         echo To : "${TO}"         echo Subject : "${SUBJ}"           echo "${1}\n${2}"         echo "Workflow ${WORKFLOW_NAME} failed!"                 )|/usr/lib/sendmail ${TO}