This way all the qualifiers (`const`, `volatile`, `mutable`) are grouped
together, e.g. `T const* const x` vs. `const T* const x`. Also helps reading
types right-to-left, e.g. "constant pointer to constant T" vs. "constant
pointer to T which is constant".
There're places where manual intervention is still required as uncrustify
is not ideal (unfortunately), but at least one may rely on it to do the
right thing most of the time (e.g. when sending in a patch).
The style itself is quite different from what we had before but making it
uniform across all the codebase is the key. I also hope that it'll make the
code more readable (YMMV) and less sensitive to further changes.
In systemd v209, released over two years ago, the various libsystemd-*
libraries (libsystemd-journal.so, libsystemd-login.so, libsystem-daemon.so,
libsystemd-id128.so) were merged into a single libsystemd.so library to
reduce code duplication and avoid cyclic dependencies.
Implement BSD/Darwin (kqueue) and Windows (ReadDirectoryChanges) mechanisms
for receiving directory change notifications. Use events instead of polling
for changes. Retry file parsing up to 3 times before giving up.
Huge thanks to missionsix for preparing first two versions of the patch.
Defer validity checks until path gets to the remote side, where they
actually make sense. Add simple checks for download directory path to
ensure it's not relative, since one cannot know what current working
directory of the remote process is.
Seems like there could be a defect in uClibc making errno not
thread-local. Don't rely on errno value but check function return value
instead which is a better failure indicator.
Return errors from `tr_loadFile` and `tr_variantFromFile` via tr_error.
Fix `tr_sessionLoadSettings` to not fail on Windows if settings.json
does not exist.
Factor out demonization implementation to platform-specific files.
Implement daemonization on Windows using service API. Improve *NIX
implementation by handling signals asynchronously to prevent potential
issues of running complex code inside the handler.
Some crypto libraries (like CyaSSL, MatrixSSL and CommonCrypto) either
don't have or expose this functionality at all, expose only part of it,
or (like OpenSSL) have heavyweight API for it. Also, for the task as
easy as BASE64 encoding and decoding it's much better to use small and
simple specialized library.
This should not affect non-Win32 platforms in any way.
As for Win32 (both MinGW and MSVC), this should hopefully allow for
unpatched compilation. Correct functioning is not yet guaranteed though.
On a way to factoring out OpenSSL support to a standalone file to ease
addition of other crypto libraries support in the future, move helpers
providing BASE64 encoding and decoding to crypto-utils.{c,h}. OpenSSL-
related functionality is moved to crypto-utils-openssl.c.
Add new functions to be implemented by crypto backends:
* tr_base64_encode_impl - encode from binary to BASE64,
* tr_base64_decode_impl - decode from BASE64 to binary.
Change `tr_base64_encode` and `tr_base64_decode` functions to expect
non-negative input data length which is considered real and never adjusted.
To process null-terminated strings (which was achieved before by passing 0
or -1 as input data length), add new `tr_base64_encode_str` and
`tr_base64_decode_str` functions which do not accept input data length as
an argument but calculate it on their own.
Mike Gelfand