Mark Allen
6226ef621d
Make sure we use the real line number
il y a 10 ans
John R. Daily
e78cc1c40d
Backwards compatibility for any code sufficiently ill-behaved to invoke 2.x do_log directly
il y a 10 ans
John R. Daily
4f94260239
Revert decision to throw a {sink, Sink} tuple at every backend; instead, special case the throttle
il y a 10 ans
Mark Allen
1906cb247d
Export get_loglevel/2 (derp)
il y a 10 ans
Mark Allen
8a178613e3
Fix errors in dispatch_log/6 parse transform
Originally I tried to bodge in a handcoded parse transform and it
did not go very well. This time, we simplify the case statement
in lager.erl and use this simplified version in the parse transform
after getting the AST from the Erlang compiler.
il y a 10 ans
Mark Allen
f233b6d342
Return sinkname when not configured from transform
il y a 10 ans
Mark Allen
c4e2520665
Fix derpy case clause with whereis
il y a 10 ans
Mark Allen
aa195121e7
Get loglevel of sink not default in transform
Previously, code would only look up the default sink's
loglevel, not the referenced sink name's loglevel.
In other words, if lager_event's loglevel was set to NONE,
it squelch all events from all sinks, and conversely if
a sink's loglevel was set to NONE, it would be silently
ignored.
il y a 10 ans
Mark Allen
566a6dc626
If Sink referenced but not running, return error
If a sink is referenced in a call, but is not configured
(by being a running process), then emit an
{error, {sink_not_configured, Sink}} tuple.
il y a 10 ans
Mark Allen
4da506b06b
Add check for same file across backends.
Throw {error, bad_config} if we find one during
handler startup.
il y a 10 ans
John R. Daily
96e3ba7095
More/fixed comments
il y a 10 ans
John R. Daily
066fa59a1c
The audit sink was always for experimentation. Remove it before it accidentally slips into a release
il y a 10 ans
John R. Daily
b0a2c4c82e
More bugfixes re: trace file handling
il y a 10 ans
John R. Daily
89ca81df2d
Bugfix: handler tuple order was incorrect
il y a 10 ans
John R. Daily
13ec782f0a
Fix the async threshold test to eliminate a race condition. It now tests something slightly different, but should be sufficient
il y a 10 ans
John R. Daily
e08fc432a8
Fix status
il y a 10 ans
John R. Daily
302767ec0f
Make dialyzer happy
il y a 10 ans
John R. Daily
d5856f2284
Add sink identification export
il y a 10 ans
John R. Daily
439a81c2ca
Introduce sleep to account for interval between writes necessary for rotation to work
il y a 10 ans
John R. Daily
9d9bf29633
Fix init (for some questionable definition of fix)
il y a 10 ans
John R. Daily
e070348290
Fix some legacy loglevel handling in tests
il y a 10 ans
John R. Daily
daf3fdeb02
Move the extra sink argument down a layer so that reinstalls work properly
il y a 10 ans
John R. Daily
a71a7ff1a7
Do not forget to clear traces from global loglevel settings
il y a 10 ans
John R. Daily
194f4aeb94
lager maps to lager_event, so audit maps to audit_event
il y a 10 ans
John R. Daily
752065eb55
Read extra sinks from configuration options
il y a 10 ans
John R. Daily
5f27739e1e
At the lowest level of the transform, allow multiple sinks
il y a 10 ans
John R. Daily
5c1a0795cf
Need dat get_loglevels/4
il y a 10 ans
John R. Daily
087e931004
error_logger was always its own sink - return to that behavior
il y a 10 ans
John R. Daily
4343bc776d
Cannot use basho_bench without a parse transform; I hope this minimal change will do it
il y a 10 ans
John R. Daily
1b14a42b5b
Follow-on to previous fix
il y a 10 ans
John R. Daily
3ae2ca020c
Mixed up application:get_env with proplists behavior
il y a 10 ans
John R. Daily
a42eda8c96
Misunderstood the inbound Module term. Should probably have specs and/or a different name. Actually represents the unique ID we want, e.g. {lager_file_handler, "foo.log"}
il y a 10 ans
John R. Daily
67565b4444
Need 2ary get_loglevel (shame it is broken tho)
il y a 10 ans
John R. Daily
564bf98248
Do not lose existing handlers
il y a 10 ans
John R. Daily
5f3f0fee6c
Match the correct tuple. Return an {ok, _} tuple so the next case statement does not get confused
il y a 10 ans
John R. Daily
fb72c10ff0
Including the config in the handler ID only confuses our ability to detect matches from the handlers list
il y a 10 ans
John R. Daily
347c9c53b4
Lager STARTS
il y a 10 ans
John R. Daily
52c753c4f8
More bugfixes
il y a 10 ans
John R. Daily
a462c046fe
Various bugfixes of various levels of desperation
il y a 10 ans
John R. Daily
8fae61af12
More config tweaks
il y a 10 ans
John R. Daily
612dd4e65a
Finish fixing get/set. Got out of the habit of compiling before committing, oops
il y a 10 ans
John R. Daily
2926dd51a1
get/2 was confusing: Sink, Key or Key, Default. Clean up by forcing into {Sink, Key} tuple
il y a 10 ans
John R. Daily
ba5548e057
update_loglevel_config now takes a sink argument
il y a 10 ans
John R. Daily
b724b6bd42
Monads...still do not know what they are but I know I need em
il y a 10 ans
John R. Daily
b88e95ea42
macro ftw
il y a 10 ans
John R. Daily
efd3164248
Stupid compiler errors
il y a 10 ans
John R. Daily
21fd439a6f
Stupid compiler errors
il y a 10 ans
John R. Daily
53488fcb1a
Trailing commas, my old enemy
il y a 10 ans
John R. Daily
0ec113132b
Traces now are assigned to a single sink as requested (or lager_event by default)
il y a 10 ans
John R. Daily
cc7e4d26d4
Start incorporating support for dedicated tracing sink
il y a 10 ans