custom_envs

Just a repo in which to save some useful and fancy UNIX/Bash functions and environment customisation

- Felipe M. Almeida

Customizing prompt

# Customizing PROMPT
SIMPLEPROMPT='\w> '
FANCYPROMP='\[\e[1;31;40m\]┏━┅◉ \[\e[4;34m\]\w\[\e[0;40m\] ┅┅◈ \d ⌚ \A \[\e[1;38m\]┅┅◈ \u@\H (\!)\[\e[0m\]\n\[\e[1;31m\]┗━►\[\e[0m\] '
PS1="$FANCYPROMP"

#this will change the title to your full current working directory
PROMPT_COMMAND='echo -ne "\033]0;${PWD##*/}\007"'
# End of: Prompt customization }}} #

Useful functions

Query/check the latest release tag of a Git repo.

This function enables that you easily check which is the latest release tag available in a github repository. Usage: get_latest_release {username/repository}.

get_latest_release() {
  curl --silent "https://api.github.com/repos/$1/releases/latest" | # Get latest release from GitHub api
    grep '"tag_name":' |                                            # Get tag line
    sed -E 's/.*"([^"]+)".*/\1/'                                    # Pluck JSON value
}

Force release update in Github.

This function enables that you easily update a published Github release to "watch" or to use/save files from the latest commit. Usage: force_git_release_update_to_commit {relase tag}.

force_git_release_update_to_commit() {
    #do things with parameters like $1 such as
    git tag -f -m "Forcing minor update of docs" -a $1
    git push
    git push -f --tags
}

Download files from onedrive

This function enables that you easily download files stored in onedrive in the command line using the "sharing" url link. Usage: onelink 'https://1drv.ms/t/s!ApRqRchcfy5ngoUeuYHA65us-fYO-Q?e=LbC7AJ' {output file name}.

onelink() {

  if [ ! -n "$2" ]; then
    name='outputfile'
  else
    name="$2"
  fi

  link=$(echo -n "$1"|base64|sed "s/=$//;s/\//\_/g;s/\+/\-/g;s/^/https:\/\/api\.onedrive\.com\/v1\.0\/shares\/u\!/;s/$/\/root\/content/" | head -n 1);
  link=\'${link}\'
  bash -c "wget -O ${name} ${link}" ;

}

Check disk usage/mounted

checkdisk() { ps aux | grep fsck; }

SSH config

How to config a ssh config, with jumps? So you can connect to servers, using another as "entrance" without using passwords? Super handy to use with VS Code.

Host main
	HostName 192.168.33.10      # can be the ip (192.168.33.10) or the address (example.com) or machine name (machine)
	User user                   # the user name to login
	Port 3457                   # port for connection
	IdentityFile ~/.ssh/id_rsa  # path to the public key created with ssh-keygen

# to access this one we must be in "main"
Host secondary
	HostName second
        User user
        Port 22
	IdentityFile ~/.ssh/id_rsa_second    # this key must be from your local machine to the second server, not from the "main" to the "second"
	ProxyJump main                       # this is the key -- it tells ssh to access this machine using the "main" as entrance
	ServerAliveInterval 240              # configuration of time for server to survive
	ServerAliveCountMax 2                # max number of alive servers

Keys must be created with ssh-keygen
- Can be manually placed inside ~/.ssh/authorized_keys inside the "server" machine
- Or can be added to the "server" machine with ssh-copy-id

sam 2 bam pipe

Just to save a manner on how to pipe a sam and return a sorted bam. Useful to create aliases.

samtools view -Shu - | samtools sort - -

bam 2 paf

Convert a bam to PAF using paftools.js from minimap2

samtools view -h <input.bam> | paftools.js sam2paf -

paf to bed (with sequences)

Create a BED file with 4 columns (chr, start, end, seq) that will be useful for replacing DNA strings in fasta using the function for that in my python scripts. It depends on bedtools and seqkit

It creates using the coordinates from the reference sequence in the PAF file.

paf2bedseq() {
  # order: PAF FASTA SELECTION (REF / QUERY)
  # for instance, if selects REF
  # it will display the coordinates of the REF
  # and its relative sequences in QUERY

  # create bed
  if [[ $3 == "REF" ]]
  then
    SELECTED_COLS="6,8,9"
    SEQUENCE_COLS="1,3,4"
  else
    SELECTED_COLS="1,3,4"
    SEQUENCE_COLS="6,8,9"
  fi

  # read paf file
  while read line; do

    # coords the user desires to have a sequence
    echo $line | cut -f "${SEQUENCE_COLS}" > SEQUENCE.bed

    # coords to replace because it must has the coordinates of one FASTA
    # with the relative sequence from the other "FASTA"
    SELECTED_COORDS=$(echo $line | cut -f "${SELECTED_COLS}")

    # get fasta from that selected col
    bedtools getfasta -fi $2 -fo SEQUENCE.fasta -bed SEQUENCE.bed

    # create tabular file with sequence only the sequence file
    DESIRED_SEQUENCE=$(seqkit fx2tab SEQUENCE.fasta | cut -f 4)

    # clean env
    rm SEQUENCE.bed SEQUENCE.fasta

    # echo
    echo -e "$SELECTED_COORDS\t$DESIRED_SEQUENCE"

  done < $1

}

fmalmeida / custom_envs