vi-global-replacements

Regular Expressions

expression	description
. (dot)	(dot) Any single character except newline
* (splat)	zero or more occurances of any character
[...]	Any single character specified in the set (refer to sets below)
[^...]	Any single character NOT specified in the set
\<	Anchor - Matches beginning of word
\>	Anchor - Matches end of word
^	Anchor - beginning of the line
$	Anchor - end of line
$...$	Grouping - usually used to group conditions
\n	Contents of nth grouping
\	Escapes the meaning of the next character
\$	Escapes the $ character
\\	Escapes the \ character

Examples of Sets - [...]

example	description
[A-Z]	SET contains capital letters A to Z.
[a-z]	SET contains lowercase letters a to z.
[0-9]	SET contains digits 0 to 9 (All numerals).
[./=+]	SET contains . (dot), / (slash), = (euqal), and + (plus).
[-A-F]	SET contains the dash, and capital letters A to F. (dashes must be specified first)
[0-9 A-Z]	SET contains all digits, a space, and capital letters A to Z.
[A-Z][a-zA-Z]	First character position SET contains capital letters A to Z. Second character position SET contains all letters.
[a-z]{m}	Look for m occurances from the SET of lowercase letters a to z.
[a-z]{m,n}	Look for at least m occurances, but no more than n occurances, from the SET of lowercase letters a to z.

Regular Expression Examples

example	description
/Hello/	Matches if the line contains the value Hello
/^TEST$/	Matches if the line contains TEST by itself
/^[a-zA-Z]/	Matches if the line starts with any letter
/^[a-z].*/	Matches if the first character of the line is a-z and there is at least one more of any character following it
/2134$/	Matches if line ends with 2134
/$21\|35$/	Matches if the line contains 21 or 35 Note the use of ( ) with the pipe symbol to specify the 'or' condition
/[0-9]*/	Matches if there are zero or more numbers in the line
/^[^#]/	Matches if the first character is not a # in the line
Notes:
Regular expressions are case sensitive
Regular expressions are to be used where pattern is specified

Ranges

Ranges may precede most "colon" commands and cause them to be executed on a line or lines. For example :3,7d would delete lines 3-7. Ranges are commonly combined with the :s command to perform a replacement on several lines, as with :.,$s/pattern/string/g to make a replacement from the current line to the end of the file.

range	description
:n,m	Range = Lines n-m
:50,100s	substitute lines 50 to 100
:.	Range = Current line (colon and dot)
:$	Range = Last line
:'c	Range = Marker c
:'a,'b	Range = Between Marker a and Marker b
:%	Range = All lines in file
:g/pattern/	Range = All lines that contain pattern

address

control what line(s) in file to pattern match

address	description
:s	substitute (only current line)
:1,$s	substitute lines 1 to end of file
:%s	substitute all lines of file (same as ":1,$s")
:g	global (all lines of file, with search pattern)
address match	pattern to match within search line(s)
command	WHAT to do on matched line(s) (usually s)
flag(s)	control how many time PER line command is executed require confirmation before each replace ( c )

Regular Expression Patterns

______________________________________________________________________________

 :s/pattern/new-string/  replace first occurance of pattern on current line
______________________________________________________________________________

 :s/pattern/new-string/g replace every occurance of pattern on current line
______________________________________________________________________________

 :g/pattern/s//new-string/g replace every occurance of pattern with new-string
______________________________________________________________________________

 :g/pattern/s/old-string/new-string/g    find each pattern occurance, then
  ^         ^                       ^    replace old-string with new-string,
  |         |                       |    for every occurance on every line
 global    command                global
 range                            flag
______________________________________________________________________________

>> Global search and replace  -->  :1,$ s/old/new/g
                                     ^  ^  ^   ^  ^
In english, this means:              |  |  |   |  |
                                     |  |  |   |  |
            From 1 to $ (end of file)   |  |   |  |
                                        |  |   |  |
            substitute -----------------/  |   |  |
                                           |   |  |
            occurrences of "old" ----------/   |  |
                                               |  |
            with occurrences of "new" ---------/  |
                                                  |
            globally (i.e., all instances of "old")

______________________________________________________________________________

Global Replacement Examples

command_line	command description
:s/old/new/c	after confirmation ([y] then <ENTER>) current_line replace FIRST old match with new
:s/old/new/	on current_line replace FIRST old match with new
:s/old/new/g	on current_line replace ALL old matches with new
:%s/old/new/	on EVERY line replace FIRST old matches with new
:10,20s/old/new/	lines 10-20 replace FIRST old match with new
:11,$s/old/new/	lines 11-EOF replace FIRST old match with new
:g/find/s/old/new/ :g/find/ s/old/new/	EVERY line containing find replace FIRST old match with new
:g/find/s//new/	EVERY line containing find replace FIRST find match with new ( s// = use find as old pattern )
:%s/\$//g or :g/\$/s///g	EVERY line replace ALL "$" matches with NULL
:%s/^M//g or :g/^M/s//g	EVERY line replace ALL "^M" matches with NULL
:.,$s/$/append/	current_line-EOF replace end_of_line with append
:.,$s/^/insert/	current_line-EOF replace start_of_line with insert
:g/^[ ]<tr><td>/s/$/<\/td><\/tr>/g	EVERY line replace ALL old matches with new ^[ ]<tr><td> = from start of line, zero or more spaces, the string "<tr><td>", then zero or more characters. s/$/<\/td><\/tr>/ = substitute end of line ($) with "</td></tr>"
Notes:
EOF	End Of File
NULL	in effect, delete
:v/./d or :g/^$/d	Delete all empty lines. (Add match for blanks if required.)
\$	$ escaped via \
\/	/ escaped via \
^M	get ^M (carriage return) via <CTRL-V><CTRL-M>