processors

type BigQueryConfig ¶

type BigQueryConfig struct {
	JsonPemPath string
	ProjectID   string
	DatasetID   string
}

BigQueryConfig is used when init'ing new BigQueryReader instances.

type BigQueryReader ¶

type BigQueryReader struct {
	PageSize         int    // defaults to 5000
	AggregateResults bool   // determines whether to send data as soon as available or to aggregate and send all query results, defaults to false
	UnflattenResults bool   // defaults to false
	TmpTableName     string // Used when UnflattenResults is true. default to "_ratchet_tmp"
	ConcurrencyLevel int    // See ConcurrentDataProcessor
	// contains filtered or unexported fields
}

BigQueryReader is used to query data from Google's BigQuery, and it behaves similarly to SQLReader. See SQLReader docs for explanation on static vs dynamic querying.

Note: If your data set contains nested/repeated fields you will likely want to get results back "unflattened." By default BigQuery returns results in a flattened format, which duplicates rows for each repeated value. This can be annoying to deal with, so BigQueryReader provides a "UnflattenResults" flag that will handle querying in such a way to get back unflattened results. This involves using a temporary table setting and a couple of other special query settings - read the BigQuery docs related to flatten and repeated fields for more info.

func NewBigQueryReader(config *BigQueryConfig, query string) *BigQueryReader

func NewDynamicBigQueryReader(config *BigQueryConfig, sqlGenerator func(data.JSON) (string, error)) *BigQueryReader

func (r *BigQueryReader) Concurrency() int

func (r *BigQueryReader) Finish(outputChan chan data.JSON, killChan chan error)

func (r *BigQueryReader) ForEachQueryData(d data.JSON, killChan chan error, forEach func(d data.JSON))

func (r *BigQueryReader) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *BigQueryReader) String() string

type BigQueryWriter ¶

type BigQueryWriter struct {
	ConcurrencyLevel int // See ConcurrentDataProcessor
	// contains filtered or unexported fields
}

BigQueryWriter is used to write data to Google's BigQuery. If the table you want to write to already exists, use NewBigQueryWriter, otherwise use NewBigQueryWriterForNewTable and the desired table structure will be created when the client is initiated.

func NewBigQueryWriter(config *BigQueryConfig, tableName string) *BigQueryWriter

func NewBigQueryWriterForNewTable(config *BigQueryConfig, tableName string, fields map[string]string) *BigQueryWriter

func (w *BigQueryWriter) Concurrency() int

func (w *BigQueryWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (w *BigQueryWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (w *BigQueryWriter) String() string

func (w *BigQueryWriter) WriteBatch(queuedRows []map[string]interface{}) (err error)

type CSVTransformer ¶

type CSVTransformer struct {
	Parameters util.CSVParameters
}

CSVTransformer converts data.JSON objects into a CSV string object and sends it on to the next stage. In use-cases where you simply want to write to a CSV file, use CSVWriter instead.

CSVTransformer is for more complex use-cases where you need to generate CSV data and perhaps send it to multiple output stages.

func NewCSVTransformer() *CSVTransformer

func (w *CSVTransformer) Finish(outputChan chan data.JSON, killChan chan error)

func (w *CSVTransformer) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (w *CSVTransformer) String() string

type CSVWriter ¶

type CSVWriter struct {
	Parameters util.CSVParameters
}

CSVWriter is handles converting data.JSON objects into CSV format, and writing them to the given io.Writer. The Data must be a valid JSON object or a slice of valid JSON objects. If you already have Data formatted as a CSV string you can use an IoWriter instead.

func NewCSVWriter(w io.Writer) *CSVWriter

func (w *CSVWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (w *CSVWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (w *CSVWriter) String() string

type FileReader ¶

type FileReader struct {
	// contains filtered or unexported fields
}

FileReader opens and reads the contents of the given filename.

func NewFileReader(filename string) *FileReader

func (r *FileReader) Finish(outputChan chan data.JSON, killChan chan error)

func (r *FileReader) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *FileReader) String() string

type FtpWriter ¶

type FtpWriter struct {
	// contains filtered or unexported fields
}

FtpWriter type represents an ftp writter processor

func NewFtpWriter(host, username, password, path string) *FtpWriter

func (f *FtpWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (f *FtpWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (f *FtpWriter) String() string

type FuncTransformer ¶

type FuncTransformer struct {
	Name             string // can be set for more useful log output
	ConcurrencyLevel int    // See ConcurrentDataProcessor
	// contains filtered or unexported fields
}

FuncTransformer executes the given function on each data payload, sending the resuling data to the next stage.

While FuncTransformer is useful for simple data transformation, more complicated tasks justify building a custom implementation of DataProcessor.

func NewFuncTransformer(transform func(d data.JSON) data.JSON) *FuncTransformer

func (t *FuncTransformer) Concurrency() int

func (t *FuncTransformer) Finish(outputChan chan data.JSON, killChan chan error)

func (t *FuncTransformer) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (t *FuncTransformer) String() string

type HTTPRequest ¶

type HTTPRequest struct {
	Request *http.Request
	Client  *http.Client
}

HTTPRequest executes an HTTP request and passes along the response body. It is simply wrapping an http.Request and http.Client object. See the net/http docs for more info: https://golang.org/pkg/net/http

func NewHTTPRequest(method, url string, body io.Reader) (*HTTPRequest, error)

func (r *HTTPRequest) Finish(outputChan chan data.JSON, killChan chan error)

func (r *HTTPRequest) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *HTTPRequest) String() string

type IoReader ¶

type IoReader struct {
	Reader     io.Reader
	LineByLine bool // defaults to true
	BufferSize int
	Gzipped    bool
}

IoReader wraps an io.Reader and reads it.

func NewIoReader(reader io.Reader) *IoReader

func (r *IoReader) Finish(outputChan chan data.JSON, killChan chan error)

func (r *IoReader) ForEachData(killChan chan error, foo func(d data.JSON))

func (r *IoReader) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *IoReader) String() string

type IoReaderWriter ¶

type IoReaderWriter struct {
	IoReader
	IoWriter
}

IoReaderWriter performs both the job of a IoReader and IoWriter. It will read data from the given io.Reader, write the resulting data to the given io.Writer, and (if the write was successful) send the data to the next stage of processing.

IoReaderWriter is composed of both a IoReader and IoWriter, so it supports all of the same properties and usage options.

func NewIoReaderWriter(reader io.Reader, writer io.Writer) *IoReaderWriter

func (r *IoReaderWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (r *IoReaderWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *IoReaderWriter) String() string

type IoWriter ¶

type IoWriter struct {
	Writer     io.Writer
	AddNewline bool
}

IoWriter wraps any io.Writer object. It can be used to write data out to a File, os.Stdout, or any other task that can be supported via io.Writer.

func NewIoWriter(writer io.Writer) *IoWriter

func (w *IoWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (w *IoWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (w *IoWriter) String() string

type MySQLWriter ¶

type MySQLWriter struct {
	TableName        string
	OnDupKeyUpdate   bool
	OnDupKeyFields   []string
	ConcurrencyLevel int // See ConcurrentDataProcessor
	BatchSize        int
	// contains filtered or unexported fields
}

MySQLWriter handles INSERTing data.JSON into a specified SQL table. If an error occurs while building or executing the INSERT, the error will be sent to the killChan.

Note that the data.JSON must be a valid JSON object or a slice of valid objects, where the keys are column names and the the values are the SQL values to be inserted into those columns.

For use-cases where a MySQLWriter instance needs to write to multiple tables you can pass in SQLWriterData.

func NewMySQLWriter(db *sql.DB, tableName string) *MySQLWriter

func (s *MySQLWriter) Concurrency() int

func (s *MySQLWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (s *MySQLWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *MySQLWriter) String() string

type Passthrough ¶

type Passthrough struct {
	// contains filtered or unexported fields
}

Passthrough simply passes the data on to the next stage. We have to set a placeholder field - if we leave this as an empty struct we get some properties for comparison and memory addressing that are not desirable and cause comparison bugs (see: http://dave.cheney.net/2014/03/25/the-empty-struct)

func NewPassthrough() *Passthrough

func (r *Passthrough) Finish(outputChan chan data.JSON, killChan chan error)

func (r *Passthrough) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *Passthrough) String() string

type PostgreSQLDeleterWriter ¶

type PostgreSQLDeleterWriter struct {
	TableName        string
	OnDupKeyUpdate   bool
	OnDupKeyIndex    string // The conflict target: see https://www.postgresql.org/docs/9.5/static/sql-insert.html
	OnDupKeyFields   []string
	ConcurrencyLevel int // See ConcurrentDataProcessor
	BatchSize        int
	// contains filtered or unexported fields
}

PostgreSQLDeleterWriter handles DELETING data and then INSERTing data.JSON into a specified SQL table. If an error occurs while deleting or executing the INSERT, the error will be sent to the killChan.

deleteSql must be a string to just be executed.

Note that the data.JSON must be a valid JSON object or a slice of valid objects, where the keys are column names and the the values are the SQL values to be inserted into those columns.

For use-cases where a PostgreSQLDeleterWriter instance needs to write to multiple tables you can pass in SQLWriterData.

Note that if `OnDupKeyUpdate` is true (the default), you *must* provide a value for `OnDupKeyIndex` (which is the PostgreSQL conflict target).

func NewPostgreSQLDeleterWriter(db *sql.DB, tableName string, deleteSql string) *PostgreSQLDeleterWriter

func (s *PostgreSQLDeleterWriter) Concurrency() int

func (s *PostgreSQLDeleterWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (s *PostgreSQLDeleterWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *PostgreSQLDeleterWriter) String() string

type PostgreSQLWriter ¶

type PostgreSQLWriter struct {
	TableName        string
	OnDupKeyUpdate   bool
	OnDupKeyIndex    string // The conflict target: see https://www.postgresql.org/docs/9.5/static/sql-insert.html
	OnDupKeyFields   []string
	ConcurrencyLevel int // See ConcurrentDataProcessor
	BatchSize        int
	// contains filtered or unexported fields
}

PostgreSQLWriter handles INSERTing data.JSON into a specified SQL table. If an error occurs while building or executing the INSERT, the error will be sent to the killChan.

Note that the data.JSON must be a valid JSON object or a slice of valid objects, where the keys are column names and the the values are the SQL values to be inserted into those columns.

For use-cases where a PostgreSQLWriter instance needs to write to multiple tables you can pass in SQLWriterData.

Note that if `OnDupKeyUpdate` is true (the default), you *must* provide a value for `OnDupKeyIndex` (which is the PostgreSQL conflict target).

func NewPostgreSQLWriter(db *sql.DB, tableName string) *PostgreSQLWriter

func (s *PostgreSQLWriter) Concurrency() int

func (s *PostgreSQLWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (s *PostgreSQLWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *PostgreSQLWriter) String() string

type RegexpMatcher ¶

type RegexpMatcher struct {

	// Set to true to log each match attempt (logger must be in debug mode).
	DebugLog bool
	// contains filtered or unexported fields
}

RegexpMatcher checks if incoming data matches the given Regexp, and sends it on to the next stage only if it matches. It is using regexp.Match under the covers: https://golang.org/pkg/regexp/#Match

func NewRegexpMatcher(pattern string) *RegexpMatcher

func (r *RegexpMatcher) Finish(outputChan chan data.JSON, killChan chan error)

func (r *RegexpMatcher) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (r *RegexpMatcher) String() string

type SCP ¶

type SCP struct {
	Port        string // e.g., "2222" -- only send for non-standard ports
	Object      string // e.g., "/path/to/file.txt"
	Destination string // e.g., "user@host:/path/to/destination/"

}

SCP executes the scp command, sending the given file to the given destination.

func NewSCP(obj string, destination string) *SCP

func (s *SCP) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SCP) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SCP) Run(killChan chan error)

type SQLExecutor ¶

type SQLExecutor struct {
	// contains filtered or unexported fields
}

SQLExecutor runs the given SQL and swallows any returned data.

It can operate in 2 modes: 1) Static - runs the given SQL query and ignores any received data. 2) Dynamic - generates a SQL query for each data payload it receives.

The dynamic SQL generation is implemented by passing in a "sqlGenerator" function to NewDynamicSQLExecutor. This allows you to write whatever code is needed to generate SQL based upon data flowing through the pipeline.

func NewDynamicSQLExecutor(dbConn *sql.DB, sqlGenerator func(data.JSON) (string, error)) *SQLExecutor

func NewSQLExecutor(dbConn *sql.DB, sql string) *SQLExecutor

func (s *SQLExecutor) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SQLExecutor) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SQLExecutor) String() string

type SQLPassthroughExecutor ¶

type SQLPassthroughExecutor struct {
	// contains filtered or unexported fields
}

SQLPassthroughExecutor runs the given SQL AND passes original data through to the next stage.

It can operate in 1 mode: 1) Static - runs the given SQL query and ignores any received data.

func NewSQLPassthroughExecutor(dbConn *sql.DB, sql string, timesMaxRun int) *SQLPassthroughExecutor

func (s *SQLPassthroughExecutor) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SQLPassthroughExecutor) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SQLPassthroughExecutor) String() string

type SQLReader ¶

type SQLReader struct {
	BatchSize         int
	StructDestination interface{}
	ConcurrencyLevel  int // See ConcurrentDataProcessor
	// contains filtered or unexported fields
}

SQLReader runs the given SQL and passes the resulting data to the next stage of processing.

It can operate in 2 modes: 1) Static - runs the given SQL query and ignores any received data. 2) Dynamic - generates a SQL query for each data payload it receives.

The dynamic SQL generation is implemented by passing in a "sqlGenerator" function to NewDynamicSQLReader. This allows you to write whatever code is needed to generate SQL based upon data flowing through the pipeline.

func NewDynamicSQLReader(dbConn *sql.DB, sqlGenerator func(data.JSON) (string, error)) *SQLReader

func NewSQLReader(dbConn *sql.DB, sql string) *SQLReader

func (s *SQLReader) Concurrency() int

func (s *SQLReader) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SQLReader) ForEachQueryData(d data.JSON, killChan chan error, forEach func(d data.JSON))

func (s *SQLReader) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SQLReader) String() string

type SQLReaderMySQLWriter ¶

type SQLReaderMySQLWriter struct {
	SQLReader
	MySQLWriter
	ConcurrencyLevel int // See ConcurrentDataProcessor
}

SQLReaderMySQLWriter performs both the job of a SQLReader and MySQLWriter. This means it will run a SQL query, write the resulting data into a MySQL database, and (if the write was successful) send the queried data to the next stage of processing.

SQLReaderMySQLWriter is composed of both a SQLReader and MySQLWriter, so it supports all of the same properties and usage options (such as static versus dynamic SQL querying).

func NewDynamicSQLReaderMySQLWriter(readConn *sql.DB, writeConn *sql.DB, sqlGenerator func(data.JSON) (string, error), writeTable string) *SQLReaderMySQLWriter

func NewSQLReaderMySQLWriter(readConn *sql.DB, writeConn *sql.DB, readQuery, writeTable string) *SQLReaderMySQLWriter

func (s *SQLReaderMySQLWriter) Concurrency() int

func (s *SQLReaderMySQLWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SQLReaderMySQLWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SQLReaderMySQLWriter) String() string

type SQLReaderPostgreSQLWriter ¶

type SQLReaderPostgreSQLWriter struct {
	SQLReader
	PostgreSQLWriter
	ConcurrencyLevel int // See ConcurrentDataProcessor
}

SQLReaderPostgreSQLWriter performs both the job of a SQLReader and PostgreSQLWriter. This means it will run a SQL query, write the resulting data into a PostgreSQL database, and (if the write was successful) send the queried data to the next stage of processing.

SQLReaderPostgreSQLWriter is composed of both a SQLReader and PostgreSQLWriter, so it supports all of the same properties and usage options (such as static versus dynamic SQL querying).

func NewDynamicSQLReaderPostgreSQLWriter(readConn *sql.DB, writeConn *sql.DB, sqlGenerator func(data.JSON) (string, error), writeTable string) *SQLReaderPostgreSQLWriter

func NewSQLReaderPostgreSQLWriter(readConn *sql.DB, writeConn *sql.DB, readQuery, writeTable string) *SQLReaderPostgreSQLWriter

func (s *SQLReaderPostgreSQLWriter) Concurrency() int

func (s *SQLReaderPostgreSQLWriter) Finish(outputChan chan data.JSON, killChan chan error)

func (s *SQLReaderPostgreSQLWriter) ProcessData(d data.JSON, outputChan chan data.JSON, killChan chan error)

func (s *SQLReaderPostgreSQLWriter) String() string

type SQLWriterData ¶

type SQLWriterData struct {
	TableName  string      `json:"table_name"`
	InsertData interface{} `json:"insert_data"`
}

SQLWriterData is a custom data structure you can send into a MySQLWriter stage or a PostreSQLWriter stage if you need to specify TableName on a per-data payload basis. No extra configuration is needed to use SQLWriterData, each data payload received is first checked for this structure before processing.