# Pathling

> Tools for FHIR analytics, built on Apache Spark. Includes Python, R, Scala and Java libraries, plus a FHIR server.


## roadmap

We are continually adding new features to the various different components of Pathling.

- [Roadmap](/roadmap.md): We are continually adding new features to the various different components of Pathling.

## docs

Pathling is a set of tools that make it easier to use FHIR and clinical terminology within health data analytics.

- [Overview](/docs.md): Pathling is a set of tools that make it easier to use FHIR and clinical terminology within health data analytics.

### server

A FHIR R4 analytics server implementing SQL on FHIR and Bulk Data Access.

- [Server](/docs/server.md): A FHIR R4 analytics server implementing SQL on FHIR and Bulk Data Access.
- [Server](/docs/server/2.0.1.md): A FHIR R4 analytics server implementing SQL on FHIR and Bulk Data Access.
- [Admin UI](/docs/server/2.0.1/admin-ui.md): The admin UI provides a web interface for managing Pathling Server.
- [Authorization](/docs/server/2.0.1/authorization.md): Pathling can perform the role of a resource server within the OpenID Connect framework.
- [Caching](/docs/server/2.0.1/caching.md): Pathling implements ETag-based cache validation, which enables clients to skip processing of queries when the underlying data has not changed.
- [Configuration](/docs/server/2.0.1/configuration.md): Configuration options for the Pathling server.
- [Kubernetes](/docs/server/2.0.1/deployment/kubernetes.md): Instructions for deploying Pathling server on Kubernetes using Helm.
- [Synchronization](/docs/server/2.0.1/deployment/synchronization.md): Keep Pathling in sync with a source FHIR server using scheduled jobs and the $import-pnp operation.
- [Getting started](/docs/server/2.0.1/getting-started.md): Quick start guide for running Pathling Server with Docker.
- [Batch](/docs/server/2.0.1/operations/batch.md): The batch operation allows you to perform multiple create, update, and delete operations in a single request.
- [Bulk submit](/docs/server/2.0.1/operations/bulk-submit.md): The bulk submit operation allows external systems to push FHIR data to the server using a staged submission workflow.
- [Create, read, update, and delete](/docs/server/2.0.1/operations/crud.md): Pathling implements the create, read, update, and delete operations from the FHIR REST API.
- [Export](/docs/server/2.0.1/operations/export.md): The export operation allows FHIR data to be exported from the server in bulk using the FHIR Bulk Data Access specification.
- [Import](/docs/server/2.0.1/operations/import.md): The import operation allows FHIR data to be imported into the server from external sources such as S3, HDFS, or local filesystem.
- [Import (ping and pull)](/docs/server/2.0.1/operations/import-pnp.md): The import-pnp operation allows bulk data to be imported via a ping and pull workflow, where the server fetches data from a remote FHIR bulk export endpoint.
- [Search](/docs/server/2.0.1/operations/search.md): Pathling supports both standard FHIR search parameters and FHIRPath-based search for filtering resources.
- [Run SQL query](/docs/server/2.0.1/operations/sql-run.md): The sqlquery-run operation executes a SQL query against materialised ViewDefinition tables and returns tabular results.
- [Export view](/docs/server/2.0.1/operations/view-export.md): The view export operation executes ViewDefinitions against FHIR data and exports the results to files in NDJSON, CSV, or Parquet format.
- [Run view](/docs/server/2.0.1/operations/view-run.md): The view run operation executes a ViewDefinition against FHIR data and returns tabular results in NDJSON or CSV format.
- [Admin UI](/docs/server/admin-ui.md): The admin UI provides a web interface for managing Pathling Server.
- [Authorization](/docs/server/authorization.md): Pathling can perform the role of a resource server within the OpenID Connect framework.
- [Caching](/docs/server/caching.md): Pathling implements ETag-based cache validation, which enables clients to skip processing of queries when the underlying data has not changed.
- [Configuration](/docs/server/configuration.md): Configuration options for the Pathling server.
- [Kubernetes](/docs/server/deployment/kubernetes.md): Instructions for deploying Pathling server on Kubernetes using Helm.
- [Synchronization](/docs/server/deployment/synchronization.md): Keep Pathling in sync with a source FHIR server using scheduled jobs and the $import-pnp operation.
- [Getting started](/docs/server/getting-started.md): Quick start guide for running Pathling Server with Docker.
- [Batch](/docs/server/operations/batch.md): The batch operation allows you to perform multiple create, update, and delete operations in a single request.
- [Bulk submit](/docs/server/operations/bulk-submit.md): The bulk submit operation allows external systems to push FHIR data to the server using a staged submission workflow.
- [Create, read, update, and delete](/docs/server/operations/crud.md): Pathling implements the create, read, update, and delete operations from the FHIR REST API.
- [Export](/docs/server/operations/export.md): The export operation allows FHIR data to be exported from the server in bulk using the FHIR Bulk Data Access specification.
- [Import](/docs/server/operations/import.md): The import operation allows FHIR data to be imported into the server from external sources such as S3, HDFS, or local filesystem.
- [Import (ping and pull)](/docs/server/operations/import-pnp.md): The import-pnp operation allows bulk data to be imported via a ping and pull workflow, where the server fetches data from a remote FHIR bulk export endpoint.
- [Jobs](/docs/server/operations/jobs.md): The jobs operation lists the asynchronous jobs held in the server's in-memory registry, so a client can monitor and cancel their work without retaining every kick-off response.
- [Search](/docs/server/operations/search.md): Pathling supports both standard FHIR search parameters and FHIRPath-based search for filtering resources.
- [Export](/docs/server/operations/sql-export.md): The sql-export operation asynchronously exports one or more ViewDefinition, SQLQuery or SQLView subjects to downloadable files.
- [Run](/docs/server/operations/sql-run.md): The sql-run operation synchronously executes a ViewDefinition, SQLQuery or SQLView and returns the result.
- [SQL on FHIR](/docs/server/sql-on-fhir.md): How Pathling Server runs SQL on FHIR ViewDefinitions, SQLQueries and SQLViews, resolves their dependencies and exposes value sets and concept maps to SQL.
- [Terminology](/docs/server/terminology.md): How to configure Pathling Server to use a remote FHIR terminology server, with or without authentication, or a local terminology store built with the Pathling libraries.

### fhirpath

Pathling implements FHIRPath functionality to aid in querying and constructing views over FHIR data.

- [FHIRPath](/docs/fhirpath.md): Pathling implements FHIRPath functionality to aid in querying and constructing views over FHIR data.
- [Extension functions](/docs/fhirpath/extension-functions.md): Extension functions and data types implemented in Pathling in addition to the standard FHIRPath and FHIR specifications.
- [FHIR-specific functions](/docs/fhirpath/fhir-functions.md): FHIR-specific functions defined in the FHIR specification as additional FHIRPath functions for use with FHIR data.
- [SQL on FHIR functions](/docs/fhirpath/sql-on-fhir-functions.md): Functions defined in the SQL on FHIR specification for producing keys that join the rows of one view to another.

### libraries

Pathling provides a set of libraries that provide assistance with using FHIR and terminology services from Apache Spark applications and data science workflows.

- [Libraries](/docs/libraries.md): Pathling provides a set of libraries that provide assistance with using FHIR and terminology services from Apache Spark applications and data science workflows.
- [Command line interface](/docs/libraries/cli.md): The Pathling command line interface surfaces the Python library's functionality - data conversion, SQL on FHIR views, FHIRPath evaluation, bulk export, terminology operations, Python scripting, and an interactive console - through scriptable commands.
- [Configuration](/docs/libraries/cli/configuration.md): Reference for the Pathling command line interface configuration file, listing every key, its matching flag and its default.
- [Adding display terms to codes](/docs/libraries/examples/display-terms.md): Tutorial demonstrating how to add human-readable display terms to medical codes using Pathling's terminology functions.
- [FHIR server synchronisation](/docs/libraries/examples/fhir-server-sync.md): Examples of synchronising your analytic data store with a FHIR server.
- [Grouping and analysing SNOMED CT data](/docs/libraries/examples/grouping-snomed.md): Tutorial demonstrating how to group and analyse SNOMED CT data using Python with the Pathling library and a terminology server.
- [Querying MIMIC-IV data](/docs/libraries/examples/mimic-iv.md): Example of running queries over the MIMIC-IV on FHIR dataset using the Pathling libraries.
- [Prostate cancer risk factors](/docs/libraries/examples/prostate-cancer.md): Example of running SQL on FHIR queries using the Pathling libraries.
- [Recursive traversal with repeat](/docs/libraries/examples/questionnaire-repeat.md): Example of using the repeat directive to flatten nested QuestionnaireResponse items.
- [Streaming queries](/docs/libraries/examples/streaming.md): Examples of running streaming queries over FHIR data using the Pathling libraries.
- [Classifying medicines with a concept map](/docs/libraries/examples/terminology-in-sql.md): Tutorial demonstrating how to join FHIR data to the mappings of a concept map in Spark SQL, following the SQL on FHIR Terminology in SQL page.
- [FHIRPath](/docs/libraries/fhirpath.md): The Pathling library provides functions for converting FHIRPath expressions into Spark Columns for filtering and extracting data.
- [Installation](/docs/libraries/installation.md): Instructions for installing the Pathling libraries for Python, R, Scala, and Java.
- [Databricks installation](/docs/libraries/installation/databricks.md): Instructions for installing the Pathling library on a Databricks cluster.
- [Pre-release versions](/docs/libraries/installation/prerelease.md): Instructions for using a pre-release version of the Pathling library.
- [Spark configuration](/docs/libraries/installation/spark.md): Instructions for configuring Apache Spark to use the Pathling library.
- [Windows installation](/docs/libraries/installation/windows.md): Instructions for installing the Pathling libraries on Windows.
- [Data in and out](/docs/libraries/io.md): The Pathling library can read and write data in a variety of different formats.
- [Cloud storage](/docs/libraries/io/cloud-storage.md): Configure the Pathling library to read and write FHIR data in Amazon S3, Azure Blob Storage, Azure Data Lake Storage and Google Cloud Storage.
- [FHIR encoders](/docs/libraries/io/encoders.md): The Pathling library can be used to transform FHIR Bundles or NDJSON into Spark data sets.
- [Parquet specification](/docs/libraries/io/schema.md): A scheme for representing FHIR resources within a Parquet schema.
- [Running SQL on FHIR views](/docs/libraries/running-queries.md): The Pathling library can be used to query datasets of FHIR resources using SQL on FHIR views. This is useful for creating tabular views of FHIR data for use in analytic tools.
- [Search](/docs/libraries/search.md): The Pathling library provides functions for converting FHIR search expressions into Spark Columns for filtering resources.
- [Terminology functions](/docs/libraries/terminology.md): The Pathling library provides a set of functions for querying a FHIR terminology server from within your queries and transformations.
- [Local terminology mode](/docs/libraries/terminology/local.md): Pathling can evaluate the terminology functions against a local terminology store built from SNOMED CT and FHIR terminology content, with no network dependency.
- [Terminology server support](/docs/libraries/terminology/server.md): How to point the Pathling library at a FHIR terminology server, tune the HTTP client and cache, and authenticate to a protected server.
- [Terminology in SQL](/docs/libraries/terminology/sql.md): How to expose a FHIR value set or concept map as a table that Spark SQL can join to, following the SQL on FHIR Terminology in SQL page.