Most status alerts should have priorities above normal level. Good reason for this is resume data verification case. For a large torrent(having lots of pieces) whole alert queue will be overflowed
with a `piece_finished_alert` right after resume data has been verified. Thus alerts like `torrent_checked_alert`, `torrent_finished`, `state_changed_alert` will not go the alerts queue.
* Introduce alert_priority enumeration
* Bump state_changed_alert's priority
* Set dht_direct_response_alert priority to `critical`
Calling blocks_flushed can cause the piece entry to be freed so its
callers need to be aware of that and avoid dereferencing the pointer if
the entry is freed.
If `default_storage::has_any_file` fails during fastresume data check, `piece_manager::check_no_fastresume` will skip storage initialization. In such case, any storage operation that require part file will cause an application crash.
Versions of boost as recent as 1.63 do not create a stub alias for python3
which causes build failures if no version of python3 is configured.
Duplicate that behavior so that building works even on older versions of
boost.
* Fix test_remove_torrent fails when torrent is downloaded too quick
* Fix typo when printing session alerts
* Fix test_lsd and test_upnp on multi-homed systems
* Fix test_tracker 'connect_alert' predicate
Evaluate torrent total blocks count only once. Equalize synchronization
of 'disk.disk_blocks_in_use' check used to break cycle with TEST_CHECK
statement. This makes tests that refer to web seeds execute much faster.
Numerous issues cause test fails on Windows(and Linux?). Most of them
found in print_alerts procedure.
Make print_alerts predicate check short-circuit, i.e. exit early on
first match. Fixes cases when appropriate variable may be re-assigned
multiple times with incorrect values.
Make stop_process wait until actual process successfully terminates.
This change is applicable to Windows only and helps to avoid tests
cleanup errors.
Make stop_proxy procedure do what it suppose to do. Previously it was
doing nothing with intention that all proxies are stopped after test
completion. Under Windows there are problems with proxy that has it's
socket connection abandoned. Thus failing the rest of test cases.
* Fix stop_all_proxies map using incorrect key to erase elements
* Fix wait_for_alert case with negative 'num'
* Make print_alerts predicate const& to avoid extra copy
* Make listen_alert and downloading_alert predicate procs reentrant
d-komarov