Capture whence information in all Serval logging API entry points.
Introduce the LogLevel enum type to avoid exposing internal C API
details in the Swift API.
Rename Swift API methods and types to use camelCase naming style, to
distinguish it from the C API, which uses under_score naming style.
Add servald_features.o to the libservaldaemon.a static library (was only
in libservaldaemon.so before), and create the "servald_features.h"
public header so that an iOS app's Swift code can simply refer to the
servald_features() function in order to link in all the Serval DNA
daemon features.
The iOS build now produces two frameworks instead of one:
- 'serval_dna' is the C API
- 'ServalDNA' is the Swift API (daemon)
The top-level modulemap file now produces a C API module called
'serval_dna' (renamed the library to libserval_dna.a) so that the Swift
API code can be identical for iOS and non-iOS targets.
On iOS, the log outputter uses Apple's Unified Logging System, but on
Mac OS, the logger uses the older Syslog API so that it will compile on
platforms older than macOS 10.13 High Sierra.
Fixes a bug that appeared in XCode 9.3, that wrapping clang and swiftc
with the xcrun(1) command no longer sets the #include root directory, so
compiling for iOS failed because it was using Mac OS header files.
To be consistent with documentation, the response must contain the
details of the locked identity. Added relevant assertions to the
lock/unlock test case.
Change the default 'rhizome.datastore_path' to "rhizome", ie, a sub-
directory of the instance directory.
Add a new erename() operation to "os.h".
Clean up the code around the fix 39f582cca (two commits ago) for a bug
introduced in 2014: the 'rhizome.db' file was always being created in
the instance directory, regardless of the setting of the
'rhizome.datastore_path' config option.
Whenever the Rhizome database is opened (by the server or by a CLI
invocation), if the database file does not exist but there is a
'rhizome.db' file in the legacy location, then move that file (and any
accompanying "blob" and "hash" sub-directories) to the configured
location.
Update copyright notices.
Add some servald config set-up functions to 'testdefs_routing.sh' and
'testdefs_swift.sh' that can be re-used in the Swift Routing API tests.
Improve some commenting on existing functions, for consistency.
Replace it with the -Wno-implicit-fallthrough CFLAGS_SQLITE option in
configure.ac, so it no longer causes compile errors with older compilers
that don't support that warning.
Whenever the daemon re-loads its config, it keeps any socket-based
(AF_INET) SOCK_DGRAM interfaces open that are matched by the new config,
but it used to close and then re-open all local (AF_UNIX) interfaces and
file-based interfaces such as SOCK_FILE dummy files and SOCK_STREAM
device files. This made it very difficult to develop the new
'routejava' test suite, because the Java API test harness always causes
a config re-load when it sets the REST API user/password in the config,
which caused the local socket interfaces to bounce, which interfered
with the reachability of nodes.
Now, local socket and file-based interfaces remain up after a config
re-load, as long as they are still matched by the new configuration.
Added INFO messages for interfaces that remain up after a config
re-load, and tweak interface up/down INFO messages to be consistent.
