Until libtorrent provided an "old_name" field in `file_renamed_alert` we relied
on a workaround with storing old file names to remove empty leftover folders.
PR #15356.
This also suppress the compiler warning:
src/base/bittorrent/torrentimpl.cpp:228:36: warning: comparison of integer expressions of different signedness: ‘int’ and ‘const size_t’ {aka ‘const long unsigned int’} [-Wsign-compare]
- warning C4267: 'initializing': conversion from 'size_t' to 'int', possible loss of data
Caused by mismatch between size_type of std and Qt containers. It is safe to cast to int as all of those containers hold low number of objects.
If torrent failed to write, it stops downloading and goes to
"upload mode" instead of errored state so it just keeps seeding.
Now qBittorrent indicates this state as "errored" and allows
the user to manually bring the torrent out of this state.
Implement ResumeDataStorage class as base for all "resume data" storages.
Implement BencodeResumeDataStorage class and put all existing logic of
"resume data" loading/saving there.
Define "torrent ID" concept, which is either a SHA1 hash for torrents of version 1,
or a SHA256 hash (truncated to SHA1 hash length) for torrents of version 2.
Add support for native libtorrent2 info hashes.
Originally, it was just a wrapper for libtorrent::torrent_handle class, so it mimicked its name.
It was then transformed into a more complex aggregate, but the name was retained (just by inertia).
Unlike libtorrent::torrent_handle class in whose name "handle" means the pattern used,
it does not matter for qBittorrent classes and just eats up space in the source code.
Move files/folders renaming functions to core classes.
Query file/folder for renaming by its current path.
Add ability to rename content folders from WebAPI/WebUI.
Vladimir Golovnev