When piping the output of a program into lnav, the data would
be dumped to the terminal on exit so that it would not be
lost. Since that is a bit noisy, the temp file used to store
the data is now left in .lnav so that it can be reopened later.
Older stdin captures are automatically removed after a day.
Also took the opportunity to start using filesystem::path more.
Fixes#436
The generic format picks up extra text in the timestamp, which
caused the date time scanner to reject some matches. I think
it's okay to allow for some trailing junk, so it no longer
ensures the size of the match is the same as the input.
* Remove all the wordexp related unused headers and code.
* stdout does not appear to be reassignable on OpenBSD.
* defining _WCHAR_H_CPLUSPLUS_98_CONFORMANCE_ seems to fix the const
correctness related redefinitions between wchar.h imported from
the system libc and the one imported by clang++ STL.
Fix warnings like:
ptimec.hh: In function ‘void ftime_i(char*, off_t&, ssize_t, const exttm&)’:
ptimec.hh:387:57: warning: format ‘%lld’ expects argument of type ‘long long int’, but argument 4 has type ‘uint64_t {aka long unsigned int}’ [-Wformat=]
snprintf(&dst[off_inout], len - off_inout, "%lld", t);
^
In this case we know that the `t` returned by `tm2sec` is a number of
seconds, so a long should be long enough, and anyway there is already
code checking for overflow.
Fixes#368.
https://github.com/tstack/lnav/issues/368
* This change removes the external dependency on OpenSSL, which should
allow the code to compile on Mac OS X without specifying the
MAC_OS_X_VERSION_MIN_REQUIRED non-sense.
* Spookyhash should be faster than SHA-1's Merkle-Damgård construction.
* Spookyhash is only 128-bits as opposed to SHA-1's 160-bits so memory
consumption should be less. I doubt it will lead to too many
collisions.
* The docs are updated to reflect the fact that OpenSSL is not a
dependency.