Linux-CLI-Useful

Quick storage of infrequent but super useful Linux commands:

General

Disk Management

du -sh * | sort -h

• du (disk usage)
  • S(ummary), H(uman readable)
• sort
  • H(uman readable)

Rename

Mass renaming files, comes up a lot. Used with perl regex syntax:
rename -n 's/Lab(\d)_Solutions/L0$1_Sols/' Lab*_Solutions.pdf

• Note that regex isn’t escaped like in nvim

Stow

Creates symlinks. Super useful for keeping config files in one place, but spreading them around
~/dotfiles $ stow nvim -t ~/.config/nvim

• From dotfiles, stowing all files in directory nvim
• -t flag denotes target, and not just placing everything in root
• Target is the existing nvim dir in config

Clipboard

Since we have xclip installed, we can use:

• alias xc="xclip -selection clipboard"
  So things like ls | xc work.

PDF

Pdftotext

Can extract text from pdfs pretty quick. Rad!
pdftotext "$file" - | grep "pattern"

• • flag tells it to not create a file.txt output, and instead send to buffer
    $ for file in *.pdf; do echo -e "\n *** $file ***\n"; pdftotext $file - | rg -i lock; done

pdfinfo

Can be used to extract pdf metadata

• Especially for seeing page numbers
  pdfinfo "$file" | grep Pages
• Pulls out only the line that describes number of pages

Quick text processing

Counting Lines:

wc -l

• Process line count. Useful for finding number of matches at the end of a regex loop

Translate or delete chars:

tr (1:1 character mapping)

• options for things like --delete, --squeeze-repeats, --truncate-set1 (or -d, -s, -t)
• Has multiple defined sets, like newline (\n), or letters ([:alpha:])
• Examples:
  • tr '[:lower:]' '[:upper:]' - Caps
  • tr -d ' ' - Deletes spaces
  • tr -s '\n' - Squeezing consec newlines

Stream Editor:

sed

• Operates with basic regex: sed 's/regex/replacement/g
  • Can use perl-like regex with -E flag
  • Can work in place with sed -i {regex} file.txt
• Can select lines: sed '1,5d' file - Deleted lines 1-5
• Capture groups (dabbing kirby): \(\) and \1 for bringing back

Filter/Calculate/Rearrange tabular data

awk

• Print processes with >5% CPU usage
  • ps aux | awk '$3 > 5.0 { print $2, $11 }'
  • Field 3 is CPU, Field 2 is PID, and Field 11 is the command.
• Calculate size of all log files
  • ls -l *.log | awk '{ s += $5 } END { print "Total bytes:", s }'
  • ls -l puts file size in column 5.
• Rearrange names from First Last to Last, First
  • echo "Linus Torvalds" | awk '{ print $2 ", " $1 }'
$0 is the whole lines, $1, $2, ... columns
    $NF, number of fields or the last column

Present columnar data

cat data.csv | column -t -s ','

• T(able), S(eparator) which here is comma. By default is whitespace