CSV to JSON

Package etl_pipeline is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package json2csv provides JSON to CSV functions.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package quandl provides easy access to the Quandl API It provides methods for getting response from Quandl in several formats. Basic usage looks like this: and will return a native Go object. To use the data in the response, iterate through its Data property: To receive a raw response from Quandl (CSV, JSON, XML) you can use: To pass options to the Quandl API, use something like this:

Command csv2json produces JSON from CSV

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

This package implements reading the json from ssllabs-scan output and generating a csv file. This file contains func for generating the report

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package json2csv provides JSON to CSV functions.

Ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package json2csv provides JSON to CSV functions.

Package peanut writes tagged data structs to disk in a variety of formats. Its primary purpose is to provide a single consistent interface for easy, ceremony-free persistence of record-based struct data. Each distinct struct type is written to an individual file (or table), automatically created, each named according to the name of the struct. Field/column names in each file/table are derived from struct tags. All writers use the same tags. Currently supported formats are CSV, TSV, Excel (.xlsx), JSON Lines (JSONL), and SQLite. Additional writers are also provided to assist with testing and debugging. Mutiple writers can be combined using MultiWriter. All writers have the same basic interface: a Write method, that can take any appropriately tagged struct; a Close method, which should be called to successfully complete writing; and a Cancel method, which should be called to abort writing and clean-up, in the event of an error or cancellation. It is safe to make mulltiple calls to Cancel, and it is safe to call Close after having previously called Cancel. All writers output their files atomically — that is to say: all output is written to a temporary location and only moved to the final output location when Close is called, meaning the output folder never contains any partially written files. Structs to be used with peanut must have appropriately tagged fields, for example: Fields without tags do not get written as output. First create a writer, for example: Next, write some records to it: When successfully completed: Or, to abort the whole operation in the event of an error or cancellation while writing records: Multiple writers can be combined using MultiWriter: Here w will write records to CSV files, Excel files, and a logger. Behaviour is undefined for types with the same name but in different packages, such as package1.Foo and package2.Foo. Supported datatypes for struct fields: string, bool, float32, float64, int, int8, int16, int32, int64, uint, uint8, uint16, uint32, uint64. Pointer following and nested structs are currently unsupported. Tagging a field that has an unsupported datatype will result in a error when Write is called.

This contains a highly customizable general-purpose flow exporter. For building either "go build", "go install", or the program provided in go-flows-build can be used. The latter allows for customizing builtin modules and can help with building modules as plugins or combined binaries. This flow exporter can convert network packets into flows, extract features, and export those via various file formats. In general the flow exporter reads the feature definitions (i.e. a description of which features to extract, and the flow key), and builds an execution graph (this can be viewed with the option callgraph) from this definition. Packets are then read from a source and processed in the following pipeline: () parts in the pipeline are fixed, [] parts can be configured via the specification, {} can be configured via the specification and provided via modules, and everything else can be provided from a module and configured from the command line. source is a packet source, which must provide single packets as []byte sequences and metadata like capture time, and dropped/filtered packets. The []byte-buffer can be reused for the next packet. For examples look at modules/sources. filter is a packet filter, which must return true for a given packet if it should be filtered out. For examples look at modules/filters. parse is a fixed step that parses the packet with gopacket. label is an optional step, that can provide an arbitrary label for every packet. For examples look at modules/labels. key is a fixed step that calculates the flow key. Key parameters can be configured via the specification. table, flow, record are fixed steps that are described in more detail in the flows package. merge merges the output from every table into one record stream. The feature step calculates the actual feature values. Features can be provided via modules, and the selection of which features to calculate must be provided via the specification. Features are described in more detail in the flows package. If a flow ends (e.g. because of timeout, or tcp-rst) it gets exported via the exported, which must be provided as a module and configured via the command line. For examples look at modules/exporters. The whole pipeline is executed concurrently with the following four subpipelines running concurrently: The "table"-pipeline exists n times, where n can be configured on the command line. Packets are divided onto the different "table"-pipelines according to flow-key. WARNING: Due to this concurrent processing flow output is neither order nor deterministic (without sorting)! To ensure deterministic output, flow output can be order by start time, stop time (default), or export time. Specification files are JSON files based on the NTARC format (https://nta-meta-analysis.readthedocs.io/en/latest/). Only version 2 files can be used. It is also possible to use a simpler format, if a paper specification is not needed. Simpleformat specification: V2-formated file: Unlike in the NTARC specification active_timeout and idle_timeout MUST be specified (there are no defaults). If bidirectional is true, every flow contains packets from both directions. key features give a list of features, which are used to compute a flow key. features is a formated list of features to export. This list can also contain combinations of features and operations (https://nta-meta-analysis.readthedocs.io/en/latest/features.html). Only single pass operations can ever be supported due to design restrictions in the flow exporter. In addition to the features specified in the nta-meta-analysis, two addional types of features are present: Filter features which can exclude packets from a whole flow, and control features which can change flow behaviour like exporting the flow before the end, restarting the flow, or discarding the flow. _per_packet allows exporting one flow per packet. If _allow_zero is true, then packets are accepted, where one of the parts of the flow key would be zero (e.g. non-IP packets for flow keys that contain IP-Addresses). If _expire_TCP is set to false, no TCP-based expiry is carried out (e.g. RST packets). TCP expiry is only carried out if at least the five-tuple is part of the flow key. A list of supported features can be queried with "./go-flows features" The examples directory contains several example flow specifications that can be used. The general syntax on the command line is "go-flows run <commands>" where <commands> is a list of "<verb> <which> [options] [--]" sequences. <verb> can be one of features, export, source filter, or label, and <which> is the actual module. The options can be queried from the help of the different modules (e.g. go-flows <verb>s <which>; e.g. go-flows exporters ipfix). Example: The following list describes all the different things contained in the subdirectories. Features most follow the conventions in https://nta-meta-analysis.readthedocs.io/en/latest/features.html, which states that names must follow the ipfix iana assignments (https://www.iana.org/assignments/ipfix/ipfix.xhtml), or start with an _ for common features or __ for uncommon ones. Feature names must be camelCase. The flow exporter has the full list of ipfix iana assignments already builtin which means that for these features one needs to only specifiy the name - all type information is automatically added by the flow extractor. For implementing features most of the time flows.BaseFeature is a good start point. Features need to override the needed methods: Start(*EventContext) gets called when a flow starts. Do cleanup here (features might be reused!). MUST call flows.BaseFeature.Start from this function! Event(interface{}, *EventContext, interface{}) gets called for every packet belonging to the current flow Stop(FlowEndReason, *EventContext) gets called when a flow finishes (before export) SetValue(new interface{}, when *EventContext, self interface{}) Call this one for setting a value. It stores the new value and forwards it to all dependent features. Less commonly used functions See also documentation of subpackage flows for more details about which base to choose. A simple example is the protocolIdentifier: This feature doesn't need a Start or Stop (since both functions don't provide a packet). For every packet, it checks, if the protocolIdentifier has already been set, and if it hasn't been, it sets a new value. The new value provided to Event will always be a packet.Buffer for features that expect a raw packet. For other features, this will be the actual value emitted from other features. E.g. for the specification the minfeature will receive the uint8 emitted by this feature. The final component missing from the code is the feature registration. This has to be done in init with one of the Register* functions from the flows packet. For the protocolIdentifier this looks like the following: Since protocolIdentifier is one of the iana assigned ipfix features, RegisterStandardFeature can be used, which automatically adds the rest of the ipfix information element specification. The second argument is what this feature implementation returns which in this case is a single value per flow - a FlowFeature. The third argument must be a function that returns a new feature instance. The last argument specifies the input to this features, which is a raw packet. The flows package contains a list of implemented types and Register functions. For more examples have a look at the provided features. Common part of sources/filters/labels/exporters Sources, filters, labels, and exportes must register themselves with the matching Register* function: where a name and a short description have to be provideded. The helpX function gets called if the help for this module is invoked and must write the help to os.Stderr. The newX function must parse the given arguments and return a new X. This function must have the following signature: name can be a provided name for the id, but can be empty. opts holds the parameters from a JSON specification or util.UseStringOption if args need to be parsed. args holds the rest of the arguments in case it is a command line invocation. Needed arguments must be parsed from this array and the remaining ones returned (arguments). If successful the created module must be returned as ret - otherwise an error. This function must only parse arguments and prepare the state of the module. Opening files etc. must happen in Init() All modules must fulfill the util.Module interface which contains an Init and an ID function. ID must return a string for the callgraph representation (most of the time a combination of modulename|parameter). Init will be called during intialization. Side effects like creating files must happen in Init and not during the new function! Examples of the different modules can be found in the modules directory. Sources must implement the packet.Source interface: ReadPacket gets called for reading the next packet. This function must return the layer type, the raw data of a single packet, capture information, how many packets have been skipped and filtered since the last invocation, or an error. Stop might be called asynchronously (be careful with races) to stop an ongoing capture. After or during this happening ReadPacket must return io.EOF as error. This function is only called to stop the flow exporter early (e.g. ctrl+c). data is not kept around by the flow exported which means, the source an reuse the same data buffer for every ReadPacket. Filters must implement the packet.Filter interface: Matches will be called for every packet with the capture info and the raw data as argument. If this function returns false, then the current packet gets filtered out (i.e. processing of this packet stops and the next one is used). Don't hold on to data! This will be reused for the next packet. Labels must implement the packet.Label interface: This function can return an arbitrary value as label for the packet (can also be nil for no label). If the label source is empty io.EOF must be returned. Exporters must implement the flow.Exporter interface: Fields, Export, Finish will never be called concurrently, are expected to be blocking until finished, and, therefore, don't need to take care about synchronization. The Fields function gets called before processing starts and provides a list of feature names that will be exported (e.g. the csv exporter uses this to create the csv header). Export gets called for every record that must be exported. Arguments are a template for this list of features, the actual features values, and an export time. Finish will be called after all packets and flows have been processed. This function must flush data and wait for this process to finish.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package freegeoip provides an API for searching the geolocation of IP addresses. It uses a database that can be either a local file or a remote resource from a URL. Local databases are monitored by fsnotify and reloaded when the file is either updated or overwritten. Remote databases are automatically downloaded and updated in background so you can focus on using the API and not managing the database. Also, the freegeoip package provides http handlers that any Go http server (net/http) can use. These handlers can process IP geolocation lookup requests and return data in multiple formats like CSV, XML, JSON and JSONP. It has also an API for supporting custom formats.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.

Package ratchet is a library for performing data pipeline / ETL tasks in Go. The main construct in Ratchet is Pipeline. A Pipeline has a series of PipelineStages, which will each perform some type of data processing, and then send new data on to the next stage. Each PipelineStage consists of one or more DataProcessors, which are responsible for receiving, processing, and then sending data on to the next stage of processing. DataProcessors each run in their own goroutine, and therefore all data processing can be executing concurrently. Here is a conceptual drawing of a fairly simple Pipeline: In this example, we have a Pipeline consisting of 3 PipelineStages. The first stage has a DataProcessor that runs queries on a SQL database, the second is doing custom transformation work on that data, and the third stage branches into 2 DataProcessors, one writing the resulting data to a CSV file, and the other inserting into another SQL database. In the example above, Stage 1 and Stage 3 are using built-in DataProcessors (see the "processors" package/subdirectory). However, Stage 2 is using a custom implementation of DataProcessor. By using a combination of built-in processors, and supporting the writing of any Go code to process data, Ratchet makes it possible to write very custom and fast data pipeline systems. See the DataProcessor documentation to learn more. Since each DataProcessor is running in it's own goroutine, SQLReader can continue pulling and sending data while each subsequent stage is also processing data. Optimally-designed pipelines have processors that can each run in an isolated fashion, processing data without having to worry about what's coming next down the pipeline. All data payloads sent between DataProcessors are of type data.JSON ([]byte). This provides a good balance of consistency and flexibility. See the "data" package for details and helper functions for dealing with data.JSON. Another good read for handling JSON data in Go is http://blog.golang.org/json-and-go. Note that many of the concepts in Ratchet were taken from the Golang blog's post on pipelines (http://blog.golang.org/pipelines). While the details discussed in that blog post are largely abstracted away by Ratchet, it is still an interesting read and will help explain the general concepts being applied. There are two ways to construct and run a Pipeline. The first is a basic, non-branching Pipeline. For example: This is a 3-stage Pipeline that queries some SQL data in stage 1, does some custom data transformation in stage 2, and then writes the resulting data to a SQL table in stage 3. The code to create and run this basic Pipeline would look something like: The second way to construct a Pipeline is using a PipelineLayout. This method allows for more complex Pipeline configurations that support branching between stages that are running multiple DataProcessors. Here is a (fairly complex) example: This Pipeline consists of 4 stages where each DataProcessor is choosing which DataProcessors in the subsequent stage should receive the data it sends. The SQLReader in stage 2, for example, is sending data to only 2 processors in the next stage, while the Custom DataProcessor in stage 2 is sending it's data to 3. The code for constructing and running a Pipeline like this would look like: This example is only conceptual, the main points being to explain the flexibility you have when designing your Pipeline's layout and to demonstrate the syntax for constructing a new PipelineLayout.