d3zyre/copy-all-files-from-folder

can do various things related to copying/moving/trashing/deleting and counting files

• 👋 Hi, I’m @D3Zyre (Cedric Boucher)
• I'm @425525 on discord! (display name: 42)
• 🌱 I’m currently in my third year of Computer Software Engineering Coop at the University of Alberta
• 📫 How to reach me: [email protected]

Total time spent programming since September 2022: (I started programming long before this, though)

(time spent programming in software other than VS code or while not logged in, i.e. MATLAB, is not tracked)

Updated August 28 2023

copy-all-files-from-folder's Issues

keep track of which types of error occur and return that, instead of just a general count

Create time tests / benchmarks for Filelist

I might not do this. The speed of certain methods of Filelist are fully compute / implementation dependent, and those could definitely be tested for speed, but others are very much I/O dependent and probably can't easily be improved. This could be useful for determining good constants for multithreading group sizes, or testing whether multithreaded methods are actually faster than their singlethreaded counterparts, on the same hardware.

Make the script read in the config file on its own, without needing to pipe it (as the first user input option)

error handling for if a (input file) goes missing before it gets processed?

Create thorough unit tests for Filelist class

The unit tests that are currently in place only ensure that nothing has errors or unhandled exceptions, but they don't verify that Filelist is actually operating correctly.

allow keeping folder structure in output

use regex instead of "file_starts" and "file_extensions"

Copy (and probably also move) overwrites files in destination if they already exist and are the same size?

add an option to print progress bar for each function

get (and print) an ETA along with the progressbar (based on files/sec or MB/s for the appropriate move type)

MB/s for copying or cutting between drives, files/s for cutting on the same drive, trashing, or permanently deleting

change command line interface to arguments instead of prompts

Separate functions/classes into separate files

would make it easier to navigate the code

add way to find duplicate files (check size and then checksum or something)

in duplicate finder, replace keys/s progress bar with something more linear and useful

finish multithreading

make duplicate file finder more useful; integrate with other functions, that take in a file list and spit a limited one back out

get duplicate files new output format

returns a tuple of all the files that are duplicated between path1 and path2,
as a tuple of tuples of the filepaths of all matches.
for example: ( (match1, match1, match1), (match2, match2) )

print a progress bar for returned threads when waiting for threads

convert all type checks to isinstance checks

this is better style and improves compatibility

add an additional argument to set on top of operation D for permanent deletion

Ask user to verify on certain actions, like permanent deletion? unsure

duplicate file finder should not return two copies of the same tuple but with reversed order

combine features of different functions into single larger functions to reduce code duplication and increase usability

moving files should delete old folders afterwards, if they were emptied

Fix progress bar units in move_files()

dumb idea: parallelize

duplicate finder using an excessive amount of RAM

can do checksum (or just raw comparing bytes?) when file size matches to double check that it's the same file

Create thorough unit tests for the main functionality of "Copy-All-Files-From-Folder"

making sure that files can actually be copied, moved, trashed and deleted correctly, etc

make the filepath list an object that contains that as well as the sizes and only obtains information (through getters) when needed if not already obtained

add way to find small/empty files and/or large files

Allow user to choose what happens when destination filename already exists

create GUI

if output folder does not exist, should create it instead of erroring

Add file size / date / checksum checking capability for files that have the same filenames

as opposed to the current behavior of just rename the file. maybe we don't always want that, and some files should just be deleted from source or ignored

restructure code so that outside functions return the files to process for copying/moving/deleting

I want to add the size comparison to what you can do

break filepaths list into chunks in slow function to multithread operation

maybe number of chunks could be related to number of cores, or simply divide by X paths at a time

get file sizes in advance instead of each time they need to be used?