Its motivations, compared to logback and all, are two-fold:
- easier to adjust its configuration programmatically (e.g. from command-line options),
without tangling with Java properties, etc.
- easier to have it log to a given java.io.PrintStream: when user code is running
in the Scala kernel, System.out and all are set to ad hoc java.io.PrintStream
instances, that capture the user code output. If the logging framework also sends
its logs there, these appear in the notebook, which is not what we want (we'd like
the logs to still go in the console). With this logging library, we can initially fix
the java.io.PrintStream, so that logs are indeed always sent to the console.
Linear Supertypes
AnyRef, Any
Ordering
Alphabetic
By Inheritance
Inherited
logger
AnyRef
Any
Hide All
Show All
Visibility
Public
All
Type Members
sealed abstract classLevel extends Product with Serializable with Ordered[Level]
final case classLogger extends Product with Serializable
Ad-hoc logging library.
Its motivations, compared to logback and all, are two-fold: - easier to adjust its configuration programmatically (e.g. from command-line options), without tangling with Java properties, etc. - easier to have it log to a given java.io.PrintStream: when user code is running in the Scala kernel, System.out and all are set to ad hoc java.io.PrintStream instances, that capture the user code output. If the logging framework also sends its logs there, these appear in the notebook, which is not what we want (we'd like the logs to still go in the console). With this logging library, we can initially fix the java.io.PrintStream, so that logs are indeed always sent to the console.