tr_peerUpdateProgress() is guessing at the progress size, but when we don't have the data size yet that guess isn't very good. Add boundary
clamps on the peer.progress field.
Previously we made sure to include stdbool.h (via transmission.h) before utp.h, since the latter used 'bool' without defining it. The new snapshot defines it unconditionally in non-C++ code, so now we need to include it first.
Once we've scored all n candidates, we sort them by score so that we can pick out the k best candidates. If n is large, sorting them can be expensive. If we use the Selection Algorithm, we select in O(n) without having to sort.
The bug was that I fixed#4332's off-by-one improperly in tr_cpMissingBlocksInPiece(). The piece's last block has to be calculated separately because its byte size may be different than the other blocks, The mistake in r12515 was that the last block could wind up being counted twice.
Add the new torrent to the session's torrent list by prepending it instead of appending it. That way we don't have to walk the list in order to add it. tr_session.torrentList is an unordered list, so there's no real difference between prepending and appending.
Instead of recalculating interesting pieces for each peer we loop through, calculate them just once into a bitfield and then reuse that bitfield inside the loop.
getPeerCandiates() used to read through all the torrents to determine the global peer connections, and then again to determine the global peer candidates. Now this is done in one loop instead of two.
This commit also changes tr_recentHistory from being a general-purpose tool to being a little more hardcoded for the only purpose it's used, in tr_peerMgr. If its files (history.[ch]) don't find any other "customers" in libtransmission, eventually it should be demoted to being a private helper class inside of peer-mgr.c and have the history.[ch] files removed from the build.
For example, if we're both seeds, or if the peer is not connectible, don't bother caching it for the next session. If it's still alive, we'll find it up through DHT or tracker announces next time around. As with r12253, this commit's intention is to reduce the number of unproductive peer_atoms that we waste time trying to connect to.
This is partially to address #4145 "Downloads stuck at 100%" by refactoring the bitset, bitfield, and tr_completion; however, the ripple effect is larger than usual so things may get worse in the short term before getting better.
livings124: to fix the mac build, remove bitset.[ch] from xcode
1. make some accidentally-public functions/variables private
2. comment out tr_generateAllowedSet() since we're not using it
3. argument list indentation
Jordan Lee