How to delete commands in history matching a given string?

Question

I need to delete all commands in my history matching a string. I've tried:

$ history | grep searchstring | cut -d" " -f2 | history -d
-bash: history: -d: option requires an argument

$ history | grep searchstring | cut -d" " -f2 | xargs history -d
xargs: history: No such file or directory

$ temparg() { while read i; do "$@" "$i"; done }
$ history | grep searchstring | cut -d" " -f2 | temparg history -d
(no error, but nothing is deleted)

What is the right way to do this?

Answer 1

The history command just operates on your history file, $HISTFILE (typically ~/.history or ~/.bash_history). It'll be much easier if you just remove the lines from that file, which can be done many ways. grep is one way, but you have to be careful not to overwrite the file while still reading it:

$ grep -v searchstring "$HISTFILE" > /tmp/history
$ mv /tmp/history "$HISTFILE"

Another way is with sed:

$ sed -i '/searchstring/d' "$HISTFILE"

Answer 2

Michael Mrozek's answer will work if you don't care about removing commands from the current session. If you do, you should write to the history file before doing the operations in his post by doing history -a.

Also, after you have removed the entries that you want from your history file, you can reload it by issuing history -c, then history -r.

Answer 3

For those looking for a one-liner:

while history -d $(history | grep 'SEARCH_STRING_GOES_HERE'| head -n 1 | awk {'print $1'}) ; do :; history -w; done

So, if you wanted to eg. delete multiple lines with a password in it, simply replace "SEARCH_STRING_GOES_HERE" with the password. This will search your entire history for that search string and delete it.

2 things to be aware of

• grep uses regex unless you supply -F as an argument
• The command will show 1 error, once there are no more matches. Ignore it.

Answer 4

Based on Michael and Chris' answers, I came up with the following. Add it to your ~/.bashrc file and then load it with either . ~/.bashrc or source ~/.bashrc.

:<<COMMENT
    Deletes all lines from the history that match a search string, with a
    prompt. The history file is then reloaded into memory.
Examples
    hxf &quot;rm -rf&quot;
    hxf ^source

See:
- https://unix.stackexchange.com/questions/57924/how-to-delete-commands-in-history-matching-a-given-string

COMMENT
#The unalias prevents odd errors when calling". ~/.bashrc" (May result in
#"not found" errors. That's okay).
unalias hxf
hxf() {
    read -r -p "About to delete all items from history that match &quot;$1&quot;. Are you sure? [y/N] " response
    response=${response,,}    # tolower
    if [[ $response =~ ^(yes|y)$ ]]
    then
        #Delete all matched items from the file, and duplicate it to a temp
        #location.
        echo -e "grep -v &quot;$1&quot; &quot;$HISTFILE&quot; > /tmp/history"
        grep -v "$1" "$HISTFILE" > /tmp/history
    #Clear all items in the current sessions history (in memory). This
    #empties out $HISTFILE.
    echo &quot;history -c&quot;
    history -c

    #Overwrite the actual history file with the temp one.
    echo -e &quot;mv /tmp/history \&quot;$HISTFILE\&quot;&quot;
    mv /tmp/history &quot;$HISTFILE&quot;

    #Now reload it.
    echo -e &quot;history -r \&quot;$HISTFILE\&quot;&quot;
    history -r &quot;$HISTFILE&quot;     #Alternative: exec bash
else
    echo &quot;Cancelled.&quot;
fi

}

Answer 5

cat "$HISTFILE" | grep -v "commandToDelete" >> "$HISTFILE" && exit

This worked for me. history -c && history -a didn't reload the history file properly for me. So instead I just exit my session right after rewriting the history file so the one in memory doesn't overwrite it.

Answer 6

The following commands remove any history rows with ThisIs\*#$(#ATest123

for i in $(history | grep 'ThisIs\*#$(#ATest123' | awk '{print$1}' | sort -nr); do history -d $i;done

• For loop through the command around $()
• Show history
• filter (grep) out only rows with ThisIs\*#$(#ATest123
• Use awk to only show the first column (the numbers)
• Reverse order of the numbers and sort in reverse r. Make sure to sort as a number n
• Use the list of numbers from the previous command to delete each entry history -d $i

If working with a variable do the following

remove_password='ThisIs\*#$(#ATest123'
for i in $(history | grep '"$remove_password"' | awk '{print$1}' | sort -nr); do history -d $i;done

Answer 7

Based on the idea from How can I remove duplicates in my .bash_history, preserving order?, I have following in may .bashrc,

cleanHistory() {
    cleanHistoryPattern="^Sp .+|^mf .+"
    history -w
    history -c
    # remove commands match pattern, still can use them in current session
    cat $HISTFILE | grep -v -P "$cleanHistoryPattern"|sponge $HISTFILE
    # remove duplications
    tac $HISTFILE | awk '!x[$0]++' | tac | sponge $HISTFILE
history -r

}
trap cleanHistory EXIT