- hyper - to download the books and serve the books if need be (concurrent)
- ncurses/conrod - for selecting the books
- toml or something for storing user selection and correct checksums
twox-hash - for checking integrity of files and generation checksums- clap - to specify premade manifests, package format, etc.
- zip and tar - for creating distributable packages
- clippy - good code
- Parse original free-programming-books .md to get list of pdf links and titles
- Use that to create a UI - maybe a list of choices that are either selected of not
- struct for each entry
- Title
- Link
- selected: bool
- Begin download process by checking the
downloadeddirectory - get list of wanted filenames
- get list of files in
downloadeddirectory - compare the two - if is not downloaded then download
- Spawn hyper client(s) to download the files
- Single or multiple process?
- Have error list to allow for retry, much like vim-plug
- Package downloaded files into .tar.xz and .zip files
- All files in
downloadeddirectory or only the ones in the manifest? - Maximum compression ratio (If specifiable)
- OPTIONAL (should be specifiable in args) - Serve books
- Spawn hyper server
- On connect spawn new thread
- UI of some sort to pick categories?
- Have an update function in UI -> spawn git process and then reparse
- Store choices in some files?
- Check if file exists before downloading - check hashes
- Maybe serve the books? - Useful for classes