* Reformat CMake code
* Bump minimum CMake version to 3.12
* Add target sources separately via `target_source()`
* Make `tr_win32_app_info()` add target sources on its own
* Don't use `include_directories()`
* Don't use `add_definitions()`
* Limit use of `add_compile_options()`
* Move VDKQueue target declaration to a subdirectory
* Add `tr_disable_source_files_compile()` helper
* Add `tr_target_glib_resources()` helper
* Add `tr_gettext_msgfmt()` helper
* Enable AUTOUIC for Qt client
* Enable AUTORCC for Qt client
* Remove AUTO{MOC,RCC,UIC} source group overrides
* Add `tr_target_idl_files()` helper
* Move source group setup to `tr_qt_add_translation()`
* Add `tr_target_xib_files()` helper
* Prefer `target_sources()` to intermediate variables
* Use explicit visibility versions of `target_*()` commands
* Prefer genexes to conditions in `target_*()` commands
* Add `tr_allow_compile_if()` helper
* Leave only top-level `project()`, remove the rest
* Minor fixups
* Fixup Mac QL plugin install
* Fixup IDE target folders and source groups
* Compile images into an assets catalog
* Optionally sign resulting Mac bundle
* Move 3rd-party/utility/test targets to their own folders
* Enable ARC via compile options unconditionally
* Install base and localized resources into correct directory
This seems like a Xcode quirk: if subdirectory containing the resource file
matches the destination directory name (and probably if there's more than one
resource file being installed there), the resulting folder structure then
contains an extra subdirectory, i.e. Resources/dirname/dirname/resname.
* Group source files for better structure in IDEs (Xcode, VS)
* Remove Xcode project
Using Xcode is still possible via "Xcode" CMake generator.
* Use ad-hoc signature if one is not specified
* Always use ad-hoc signature on link, leave proper signing to CPack
* Install localized resources into correct directory (QL plugin)
* Don't include CTest (we don't use additional targets it provides)
* Revert "Remove Xcode project"
This reverts commit dc069f654f.
* Fix testing against reference crypto implementation
Build would fail on CI if using cyassl/wolfssl or polarssl/mbedtls as a
backend and testing it against reference openssl implementation due to
missing include directories, but only if bundled (non-system) b64 is
used.
* Iniitalize size arguments to `wc_DhGenerateKeyPair`
Despite it not being mentioned in the documentation [1] and v4.6 release
notes [2], the two size parameters became [3] not purely "out", but
"in/out", to avoid potential buffer overflows.
[1] https://www.wolfssl.com/documentation/manuals/wolfssl/group__Diffie-Hellman.html
[2] https://github.com/wolfSSL/wolfssl/releases/tag/v4.6.0-stable
[3] 4364700c01
* feat: add tr_strbuf class for building tmp strings
Based on fmt::basic_memory_buf, this is a growable string buffer that
has an initial size that's large enough to build most filenames or URLs
without needing heap allocations.
Adds a couple of extra helpers such as a `c_str()` method to make dealing
with old zero-terminated string APIs easier.
* refactor: add tr_torrent_metainfo class
Can be used for parsing bencoded .torrent data without instantiating
a tr_torrent. This will be used in all the places where client code
needs to test a .torrent file for validity / to add a preview window
before adding the torrent.
* refactor: encapsulate request tracking in a class
Introduces a new class to peer-mgr, `ClientRequests`, which tracks what
active requests we've got pending: which blocks, when the requests were
sent, and who they were sent to.
This shouldn't change peer-mgr behavior. Its goal is to carve out some
of peer-mgr's data structures and encapsulte them behind an API that's
simpler to understand.
* refactor: move ActiveRequests to its own file
* perf: avoid duplicate call to tr_cpMissingBlocksInPiece
* Add in-kernel copying support for Linux (sendfile64(2), copy_file_range(2)), FreeBSD 13 (copy_file_range(2)), MacOS (copyfile(2)), and Windows (CopyFileExA).
* Fix macro name USE_COPY_FILE_RANGE.
* Minor bugfixes for userspace fallback.
* Fix linux sendfile64 bugs.
* Remove some overzealous asserts.
* Allow transmission-test-copy to take an optional argument for an external reference file.
* Fix return value error of tr_sys_path_copy.
* Use COPYFILE_ALL for Macs without COPYFILE_CLONE.
* Add in-kernel file copying for several platforms.
Numerous operating systems now have support for copying files directly in the
kernel, which is generally more efficient than copying in a userspace read(2)/
write(2) loop. (This becomes particularly relevant for 4th gen PCI-E storage,
which approaches the latency of DRAM.) For Linux I use sendfile64(2), and, for
later kernels, copy_file_range(2). FreeBSD 13 will also support
copy_file_range(2). MacOS has copyfile(2), and Windows has CopyFileExA.
Operating systems lacking such a syscall continue to use the existing
read(2)/write(2) loop.
* Appease uncrustify.
* Appease uncrustify.
* copy-test: generate random content at run time.
* copy-test: Stylistic changes and more check()s.
* copy-test: files_are_identical should follow test idioms
* tr_sys_path_copy: numerous tweaks as requested by review.
* s/old file/source file; s/new file/destination file.
* tr_sys_path_copy: handle win32 wide characters in paths.
* Uncrustify.
* test-copy: Use non-string create_file_with_contents.
* tr_sys_path_copy: numerous fixes.
Per review: generate test file content at runtime; tidy use of check();
fix style; re-measure file sizes in the copy; define a macro when the
system does not provide it; use Unicode APIs on Windows; and fix
documentation.
* Updated as per comments.
* Rebase kernel-copy changes onto 3.0 with gtest.
* Undo irrelevant comment change.
* Fix syntax error.
* Use tr_malloc() instead of tr_valloc().
* Use EXPECT instead of TR_ASSERT in gtest.
* Add error handling.
* Acceptable coding style has changed again.
Now it's camelCase. Also use nullptr instead of NULL, etc.
* Fix east/west const.
Co-authored-by: Mike Gelfand <mikedld@users.noreply.github.com>
Mike Gelfand