Why not samtools faidx sample.fa and then cut -f1-2 sample.fa.fai? The fasta index file will give you sequence name and length in the first two columns, and you probably already have this file indexed.

Using SeqKit's fx2tab:

seqkit fx2tab --length --name --header-line  foo.fasta

This prints sequence name and length with empty columns for sequences and qualities.

You can make your own code for this,

Copy and paste following code in fasta_len.py file:

import sys
from Bio import SeqIO

FastaFile = open(sys.argv[1], 'rU')

for rec in SeqIO.parse(FastaFile, 'fasta'):
    name = rec.id
    seq = rec.seq
    seqLen = len(rec)
    print name, seqLen

FastaFile.close()

Make this file executable by:

chmod u+x fasta_len.py

and run as,

./fasta_len.py file.fasta

Note: you should have installed biopython

Assuming you already have EMBOSS installed you could use 'infoseq' to get the format you desire. For example:

$ infoseq -auto -nocolumns -delimiter ',' -only -noheading -name -length sequences.fa
2AAA_HUMAN,589
AAPK1_PIG,385
AAPK1_MOUSE,559
AAPK1_PONAB,554
AAPK1_HUMAN,559
AAPK1_RAT,559
...

Depending on the identifier format used in your source fasta file you may need to choose between the appropriate -name, -accession, -gi or -seqversion option for the identifier column, or use the "-sformat pearson" option to maintain the original source identifier. For example to return the original identifiers when they are in db|acc|id format:

$ infoseq -auto -nocolumns -delimiter ',' -only -noheading -name -length -sformat pearson sequences.fa
sp|P30153|2AAA_HUMAN,589
sp|Q09136|AAPK1_PIG,385
sp|Q5EG47|AAPK1_MOUSE,559
sp|Q5RDH5|AAPK1_PONAB,554
sp|Q13131|AAPK1_HUMAN,559
sp|P54645|AAPK1_RAT,559
...

Granted that this is a ridiculous way to do this, I thought it might be useful to demonstrate the algorithmic steps and how they are implemented using Perl. You can think about how to modify and improve various steps, and write your own scripts.

Make sure that you have Perl installed on your computer.

C:\Users\bongbang\Desktop\perl -v

Copy and paste the following code into a text file and save it as calculateLength.pl

#!/usr/bin/perl
use warnings;
use strict;

# Point to where your input fasta file is located
my $fasta_file = "/Users/bongbang/Desktop/my_sequences.fasta";

# Declare that you want to read this file, and print the error message
# in case that the input fasta file does not exist at the specified path
open(FASTA, "<", $fasta_file) or die("Probably wrong path: $fasta_file\n");

# We will linearize the sequences into this hash
my %singleLineSequences;

# Initialize a variable to store sequence ids
my $sequence_id;

# Read the fasta file line by line
while(<FASTA>){
    my $line = $_; chomp($line);

    # if this is a new sequence with the header at the beginnning
    # Extract sequence id using regular expression (\S+) and
    # store it into the variable $sequence_id

    if ($line =~ m/^>(\S+)/){

        $sequence_id = $1; # e.g., YKR054C            

        # Reserve an entry in your hash
        # $sequence_id = YKR054C
        # $singleLineSequences{'YKR054C'} = ""
        # No sequence has been added yet, hence the empty value
        $singleLineSequences{$sequence_id} = ""; 

        }

    # if the line is not a header but part of a sequence, 
    # append this part to the corresponding sequence id in
    # your hash entry

    else {

        # paste the current line to the end of the sequence
        $singleLineSequences{$sequence_id} = $singleLineSequences{$sequence_id} . $line;

        }
   }

# Now that your hash contains single line sequences, you can simply
# loop over each sequence in your hash, determine the sequence length,
# and print it out

foreach my $sequence_entry (keys %singleLineSequences){

    # grab a hash entry and store it into a variable
    my $currentSequence = $singleLineSequences{$sequence_entry};

    # determine length of the sequence
    my $lengthSequence = length($currentSequence);

    # print the result: id,length
    print $sequence_entry . "," . $lengthSequence . "\n";
}

Go into the directory including your script, and execute it as follows:

C:\Users\bongbang\Desktop\perl calculateLength.pl

Hi all,

check this: https://github.com/alejandrogzi/chromsize

Usage

Usage: chromsize --fasta <FASTA> --output <OUTPUT> [-t <THREADS>]

Arguments:
    -f, --fasta <FASTA>: FASTA file
    -o, --output <OUTPUT>: path to chrom.sizes

Options:
    -t, --threads <THREADS>: number of threads [default: your max ncpus]
    --help: print help
    --version: print version

build in rust, the simplest and fastest way to get them. Works as a binary, library and is ported to python. Here is the benchmark with a lot of tools: