deletes files in the rrna_free_reads and trimmed_reads to save space

delete_intermediate_RNASeq.py

#!/usr/bin/env python
# coding: utf-8
from pathlib import Path
import os
from itertools import chain
from datetime import datetime
__author__: tbrittoborges
__date__: 05/02/2020


# RNA-Seq workflow
expected_lines = {
        0: 'SHELL := /bin/bash\n',
        1: '\n', 
        2: 'all: TRIMMING RRNA MAPPING\n'
    }
     
def check_makefile(file_path):
    'Check makefile for RNA-Seq type with its three first lines'     
    with open(file_path) as fin:
        for i, line in enumerate(fin):
            if i == 3:
                return True
            elif line != expected_lines[i]:
                return False

def calculate_dir_size(dir_path):
    'Computes directory size in Gb'
    return sum(x.stat().st_size for x in dir_path.rglob('*')) / 1024e6

    
def delete_and_annotate(path):
    # scan recursively and deletes files 
    for f in Path(path).rglob('*'):
        os.unlink(f)
    # documents the deletion time
    with open(Path(path) / 'DELETED') as fout:
        now = datetime.now()
        dt_string = now.strftime("%d/%m/%Y %H:%M:%S")
        fout.write(f'Content removed on {dt_stringdt_string}')       

prj = Path('/beegfs/prj/')

with open('/scratch/tbrittoborges/report_on_dir_size.txt', 'w') as fou:
    for f in prj.rglob('workflow/Makefile'):
        print('=====',
        'path: ', f,
        'makefile check? ', check_makefile(f),
        'accessible? ', os.access(f.parent, os.W_OK),
        'rrna_free size (Gb)', round(calculate_dir_size(f.parent / 'rrna_free_reads/'), 2),
        'trimmed_reads size (Gb)', round(calculate_dir_size(f.parent / 'trimmed_reads/'), 2),
        '=====', sep='\n', file=fou, flush=True)
        
        if os.access(f.parent, os.W_OK) & check_makefile(f):
            delete_and_annotate(f.parent / 'rrna_free_reads/')
            delete_and_annotate(f.parent / 'trimmed_reads/')