Merge pull request #20876 from nsoranzo/sort1_drop_Python_script

"""

Sorts tabular data on one or more columns. All comments of the file are collected

and placed at the beginning of the sorted output file.

"""

# 03/05/2013 guerler

import argparse

import subprocess

import sys

def stop_err(msg):

    sys.exit(msg)

def main():

    # define options

    parser = argparse.ArgumentParser(description=__doc__)

    parser.add_argument("-i", "--input", type=argparse.FileType("r"), help="Tabular file to be sorted")

    parser.add_argument("-o", "--output", type=argparse.FileType("w"), help="Sorted output file")

    parser.add_argument("-k", "--key", action="append", help="Key (see manual for bash/sort)")

    parser.add_argument("-H", "--header_lines", type=int, help="Number of header lines to ignore")

    # parse

    args = parser.parse_args()

    try:

        # retrieve options

        input_fh = args.input

        output_fh = args.output

        header_lines = args.header_lines

        key_args = []

        for k in args.key:

            key_args.extend(["-k", k])

        # sed header

        if header_lines > 0:

            sed_header = ["sed", "-n", f"1,{header_lines:d}p"]

            subprocess.check_call(sed_header, stdin=input_fh, stdout=output_fh)

            input_fh.seek(0)

        # grep comments

        grep_comments = ["grep", "^#"]

        exit_code = subprocess.call(grep_comments, stdin=input_fh, stdout=output_fh)

        input_fh.seek(0)

        if exit_code not in [0, 1]:

            stop_err("Searching for comment lines failed")

        # grep and sort columns

        if header_lines > 0:

            sed_cmd = ["sed", f"1,{header_lines:d}d"]

            sed_header_restore = subprocess.Popen(sed_cmd, stdin=input_fh, stdout=subprocess.PIPE)

            pipe_stdin = sed_header_restore.stdout

        else:

            pipe_stdin = input_fh

        grep = subprocess.Popen(["grep", "^[^#]"], stdin=pipe_stdin, stdout=subprocess.PIPE)

        sort = subprocess.Popen(["sort", "-f", "-t", "\t"] + key_args, stdin=grep.stdout, stdout=output_fh)

        # wait for commands to complete

        sort.communicate()

        assert sort.returncode == 0, f"sort pipeline exited with non-zero exit code: {sort.returncode:d}"

    except Exception as ex:

        stop_err("Error running sorter.py\n" + str(ex))

    # exit

    sys.exit(0)

if __name__ == "__main__":

    main()

<tool id="sort1" name="Sort" version="1.2.0">

    <description>data in ascending or descending order</description>

    <requirements>

        <requirement type="package" version="3.8">python</requirement>

        <requirement type="package" version="2.14">grep</requirement>

        <requirement type="package" version="4.4">sed</requirement>

        <requirement type="package" version="8.31">coreutils</requirement>

    </requirements>

    <command detect_errors="exit_code">

python '$__tool_directory__/sorter.py'

--input='$input'

--output='$out_file1'

    <command detect_errors="exit_code"><![CDATA[

## Sorts tabular data on one or more columns. All comments of the file are

## collected and placed at the beginning of the sorted output file.

#if $header_lines > 0:

    sed -n 1,${header_lines}p '$input' > '$out_file1' &&

#end if

grep '^#' '$input' >> '$out_file1';

#if $header_lines > 0:

    sed 1,${header_lines}d '$input' |

#else:

    < '$input'

#end if

grep '^[^#]' | sort -f -t \$'\t'

#if (str($style) == 'num'):

    #set $style = 'n'

#set $order = '' if (str($order) == 'ASC') else 'r'

--key=${column},${column}${style}${order}

-k ${column},${column}${style}${order}

#for $col in $column_set:

    #set $other_column = str($col.other_column)

    #end if

    #set $other_order = '' if (str($col.other_order) == "ASC") else 'r'

    --key=${other_column},${other_column}${other_style}${other_order}

    -k ${other_column},${other_column}${other_style}${other_order}

#end for

--header $header_lines

    </command>

>> '$out_file1'

    ]]></command>

    <inputs>

        <param format="tabular" name="input" type="data" label="Sort Dataset" />

        <param name="column" label="on column" type="data_column" data_ref="input" accept_default="true"/>