Skip to content

Latest commit

 

History

History
75 lines (53 loc) · 2.19 KB

README.md

File metadata and controls

75 lines (53 loc) · 2.19 KB
Raw

CQL REPL

A REPL CLI for interacting with the CQL engine. This can be a useful tool for quickly iterating while authoring CQL measures.

Warning: When using these tools with protected health information (PHI), please be sure to follow your organization's policies with respect to PHI.

Running

To build the program from source run the following from the root of the repository (note you must have Go installed):

go build cmd/repl/repl.go

This will build the repl binary and write it out in your current directory. You can then run the binary:

./repl
  • If an invalid statement is input, inputting an empty line clears the current cache
  • Typing in exit will quit the program.
  • To continue a statement across multiple lines type a backslash () character before submitting that line

ex:

define multiline_example: \
  @2014 + 1 year

Note: While writing multiple statements this way is technically supported the output ordering of the results is currently undefined.

You may also pass in additional flags to repl in order to load some FHIR data or load an external input CQL library, for example:

./repl \
  -cql_file="path/to/my/file.cql" \
  -bundle_file="path/to/my/patient_or_resource.json"

See the Flags section below for more.

Flags

bundle_file -- Optional

The path to a single bundle FHIR resource (often a patient). This is the resource file the cql will evaluate against for all executions in the REPL insance.

cql_file -- Optional

The path to a CQL file. The contents of this file are seeded into the current context and outputs are displayed before the execution of the first REPL loop.

valuesets_dir -- Optional

The path to a directory containing json definitions of FHIR valuesets.

Future Enhancements

  • Right now the repl can't tell if an expression is unfinished or incorrect. define test: is an unfinished statement that would be nice if we could handle separately from invalid statements.
  • Ability to run against multiple bundle files. This would require some work to ensure the output for this were readable.
  • Ability to clear the current context and load a different bundle resource to evaluate against.