playnig with os.makedirs() & pairtree

pairtree.py

"""
Skimming <https://realpython.com/working-with-files-in-python/>
for some forgotten reason, I came across the section "Creating Multiple Directories",
which showed how the command:

... os.makedirs( '2018/10/05' )

...will create a hierarchy of directories like:

.
|
└── 2018/
    └── 10/
        └── 05/

So cool! Which then got me thinking of how that could be useful for creating pairtree paths.
Here's some weekend play for that.

(Also used pathlib for the first time, hence the explicit reminder var names and asserts.)
"""

import hashlib, logging, os, pathlib, pprint, shutil, sys
from pathlib import Path


assert sys.version_info >= (3, 0)  # python3


## set up logging -------------------------------
logging.basicConfig(
    level='DEBUG',
    format='[%(asctime)s] %(levelname)s [%(module)s-%(funcName)s()::%(lineno)d] %(message)s',
    datefmt='%d/%b/%Y %H:%M:%S' )
log = logging.getLogger(__name__)
log.info( '\n\nstarting log\n============' )


## set up directory references ------------------
source_files_dir_obj = Path( '../source_files' ).resolve()
assert type(source_files_dir_obj) in [ pathlib.PosixPath, pathlib.WindowsPath ]

destination_files_dir_obj = Path( '../output_directory' ).resolve()
assert type(destination_files_dir_obj) in [ pathlib.PosixPath, pathlib.WindowsPath ]


## get source files -----------------------------
source_path_objs = list( source_files_dir_obj.iterdir() )
assert type(source_path_objs[0]) in [ pathlib.PosixPath, pathlib.WindowsPath ]
log.debug( f'source_path_objs, ``{source_path_objs}`` ' )

source_file_objs = []
for path_obj in source_path_objs:
    if path_obj.is_file():
        name = path_obj.name
        assert type(name) == str
        if name[0:1] != '.':  # macOS `.DS_Store`
            source_file_objs.append( path_obj.resolve() )
log.debug( f'source_file_objs, ``{source_file_objs}`` ' )


## iterate through source files ----------------
for path_obj in source_file_objs:
    path_str = str( path_obj )
    with open( path_str, 'rb' ) as f:  # 'rb' necessary if some of the files are images

        ## get hash (of contents) ---------------
        contents = f.read()
        assert type(contents) == bytes, type(contents)
        hxdgst = hashlib.md5( contents ).hexdigest()
        assert type(hxdgst) == str, type(hxdgst)
        log.debug( f'hxdgst, ``{hxdgst}``')

        ## get first and second pairs -----------
        ( first_pair, second_pair ) = ( hxdgst[0:2], hxdgst[2:4] )
        log.debug( f'first_pair, ``{first_pair}``; second_pair, ``{second_pair}``' )

        ## make pair directories ----------------
        new_dir_path = f'{str(destination_files_dir_obj)}/{first_pair}/{second_pair}'
        log.debug( f'new_dir_path, ``{new_dir_path}``')
        os.makedirs( new_dir_path, mode=0o770, exist_ok=True )  # mode: user/group read/write; does NOT seem to be respected

        ## copy file ----------------------------
        filename = path_obj.name
        full_source_path = f'{str(source_files_dir_obj)}/{filename}'
        log.debug( f'full_source_path, ``{full_source_path}``')
        full_destination_path = f'{new_dir_path}/{filename}'
        log.debug( f'full_destination_path, ``{full_destination_path}``')
        shutil.copy2( full_source_path, full_destination_path )


log.debug( 'done!')