Differences

This shows you the differences between two versions of the page.

--- linux:sed [2021/12/11 12:06] – created utedass
+++ linux:sed [2022/09/12 00:30] (current) – external edit 127.0.0.1
@@ Line 3: / Line 3: @@
 ===== Using as regex to get stuff =====
+By default, everything that is not deleted (by a delete command) is printed to stdout. To inhibit this, pass '-n' flag.
+Forward dash ''/'' is usually a separator character. It is used to separate different commands and their arguments.
+Truth to be told, almost any character can be used as a separator. The first character after the command seems to be used as a separator. Thus '' 's/some text/other text/p' '' could be written as '''sKsome textKother textKp' ''.
+sed takes a string as argument, which can contain commands, that in turn contain matching strings.
+The string that is passed to sed must contain a valid command. A valid command is fucked up. It can be either the first character, or the trailing character.
+For example the command to print a line that matches something would be
 <code bash>
-sed -n ''
+echo testfil.txt | sed -n '/something/p'
+</code>
+While the command to substitute comes before a string. To substitute "dude" with "shit":
+<code bash>
+echo testfil.txt | sed 's/dude/shit/'
+</code>
+But you could also combine them, and print only the lines that matches the substitutions. Thus having commands both first and last in the string:
+<code bash>
+echo testfil.txt | sed -n 's/dude/shit is fucked up/p'
+</code>
+===== Examples =====
+<code bash>
+sed '/actual pattern/command'
+# ex
+sed '/Matches this string/p' # command p is printed. But since -d flag is not provided, so will everything else too
+sed -n '/Matches this string/p' # Only prints rows containing "Matches this string"
 </code>