A Metrics implementation using com.swoop.spark.accumulators.ByKeyAdditiveAccumulator.
Convenience trait that combines the main services exposed by RecordBuilder.
Convenience trait that combines the main services exposed by RecordBuilder.
the job context
Topmost trait for record builders.
A simple, serializable exception cause that keeps the exception message and the stack trace.
Base class for record builders that produce multiple records for one input.
Base class for record builders that produce multiple records for one input.
The input type for the builder.
The input type records are built from.
The record type.
The job context type
Shared behaviors for the build context that integrates with Spark context.
Shared behaviors for the build context that integrates with Spark context.
Trait for types that have a feature mask.
Definitions for the meaning of bits in the features
bit mask of com.swoop.spark.records.Record.
Definitions for the meaning of bits in the features
bit mask of com.swoop.spark.records.Record.
The least significant 16 bits are reserved for framework use. The next 15 bits are available for application use. The most significant bit is reserved for framework use also.
A flat record environment.
Trait identifying flights: one or more related Spark jobs.
Trait for a builder that aggregates details, which will become issues.
Trait for a builder that aggregates details, which will become issues.
Under the covers, details are Throwable instances: either exceptions thrown by the application code or ThrowableMessage instances that may be thrown and caught or simply constructed using one of the convenience methods exposed by this trait.
Trait for a builder that aggregates features.
Trait for a builder with a job context.
An issue is a container for diagnostic information including both simple fields and an optional sequence of causes.
An issue is a container for diagnostic information including both simple fields and an optional sequence of causes.
The category of the issue.
It matches the with the values coming from the different types of
Issue$.ThrowableMessageType featureMask
values.
The message related to the issue.
The causes of this issue.
The issue ID.
Optional additional details for the issue. For root cause analysis purposes it is useful to to split messages we want to associate with records into a stable portion (the issue message) and a variable portion (the issue details). This enables stable grouping and counting.
Provides root cause analysis operations at the issues level.
Shared behaviors for the build context that provides services to builders but does not work directly with Spark context APIs, allowing fast Spark-less testing.
Shared behaviors for the build context that provides services to builders but does not work directly with Spark context APIs, allowing fast Spark-less testing.
Shared trait defining metrics collection services
Base class for nested record builders of single records.
Base class for nested record builders of single records.
The source input
The record data
The record
The job context
A partitioned error environment.
Implicit operations on Spark Records dataframes.
Implicit operations on Spark Records dataframes.
Most operations can be applied equally at the record level or at the level of unwound issues.
A Spark Record is an envelope around data.
A Spark Record is an envelope around data.
The data stored in the record
The input type
Base class for record builders of single records.
Base class for record builders of single records.
The source input
The record data
The record
The job context
Base class for single and composite record builders.
Base class for single and composite record builders.
The record
The job context
Base trait for the implicits that provide hints about how records are persisted.
Base trait for the implicits that provide hints about how records are persisted.
Provides root cause analysis operations at the record level.
Implicit operations on Spark Records datasets.
Implicit operations on Spark Records datasets.
Extends Option with record building sugar.
Extends scala.util.Try with record building sugar.
Extends scala.util.Try with optional values with record building sugar.
Provides root cause analysis operations at any level by automatically dispatching to RecordsDataFrame or IssuesDataFrame based on the schema of the provided dataframe.
Base trait for root cause analysis operations.
Basic implementation of DriverContext that uses AccumulatorMetrics and includes a convenience factory for creating job contexts.
Basic implementation of DriverContext that uses AccumulatorMetrics and includes a convenience factory for creating job contexts.
Basic implementation of JobContext that uses AccumulatorMetrics.
Basic implementation of JobContext that uses AccumulatorMetrics.
Spark accumulators can be used locally without registration.
A simple, serializable adapter for StackTraceElement.
Base class for throwable diagnostic messages with IDs for quick big data processing.
Base class for throwable diagnostic messages with IDs for quick big data processing.
the type of message (error, warning, info or debug)
the message
the optional message ID
the cause, if the message is being thrown as a result of an exception
Convenience trait for throwing ThrowableMessage.
Convenience conversion to make issue saving during record creation easier.
Spark Records package.
Spark Records documentation