This is a Visual Studio Code extension for working with HL7 v2.x files. It provides basic syntax highlighting, along with the following features:
- display field description when mouse is hovered over a field.
- highlight user specified fields in the message.
- mask out identifying fields in the message.
- send a HL7 message to a remote host.
- receive HL7 messages sent from a remote host.
- extract the value of a field from all open files.
- display fields from a single segment as a list.
- split a HL7 batch file into a separate file per message.
- confirm all required fields in the message contain values.
- extract all similar from the file to a new document.
- correct messages that are missing a line break delimiter between segments
- search for a field in the message by location or description
Note: The extension is automatically activated for files with a .hl7 file extension. If viewing files without a .hl7 file extension, you will need to manually specify that the file is a HL7 file. Click on the current language (e.g. 'PlainText') in the right hand side of the status bar, and enter 'hl7' as the language. It is recommended to rename files to use a .hl7 extension for ease of use.
- Segment, field, component, sub component and repeat separators, and segment names are colourised.
Note: The default Visual Studio Code dark and light themes don't highlight the field separators, I've found the Monkai & Solarize themes work well.
When the mouse is hovered over a field, the field name and location is displayed in a pop-up tooltip. If the file has a .hl7 extension this will be applied when the file loads, other wise it will need to be manually activated via:
- Press F1 --> HL7 Tools: Identify Fields
By default only the first 200 segments of each message will include hover descriptions (performance can be impacted applying this to large hl7 files). This limit is configurable via the following user preference.
"hl7tools.MaxLinesForFieldDescriptions"
Custom Segment Support
Support for custom 'Z' segments can be included by providing a custom schema definition. The custom schema file must follow the same format as the default segment schemas included in this extension ([extension folder]\schema\2.x\segments.js). Each segment fields requires values for "datatype", "desc", and "opt" fields. Set the value of the 'CustomSegmentSchema' user preference to reference the file containing the custom segment schema (full path and file name). e.g.
"hl7tools.CustomSegmentSchema": c:\schemas\CustomSchema.js
The custom schema should not modify segments defined in the HL7 v2.x specification, this is intended for 'Z' segment definitions only. It is assumed all custom segments will use standard data types.
The custom segments schema will be available to all extension commands that use field descriptions.
This command prompts for a HL7 field location (e.g. PID-3), or partial field description (e.g. name). The corresponding field(s) are then highlighted in the editor. The field highlighting will be applied to other HL7 messages when they become the active document. The background colour applied to highlighted fields can be configured via the user preference "hl7tools.highlightBackgroundColor". This preference requires a RGBA colour value, specified as: rgba(red, green, blue, alpha). The alpha parameter is a number between 0.0 (fully transparent) and 1.0 (fully opaque).
- Press F1 --> HL7 Tools: Highlight Field
- Enter the reference to the segment and field location (e.g. PID-3), or enter part of the field name (e.g. patient).
- The field is highlighted. If part of a field name was entered, all matching fields will be highlighted. e.g. 'Patient' would match to 'Patient ID List', 'Patient Name', etc. Selecting a new field will remove the previously highlighted field. To remove highlighting run the command 'HL7 Tools: Clear Highlighted Field', or enter a blank value for a field location.
Note: the field highlighting may be shifted by a character if the document end of line character is changed from EOL preference defined by the editor settings.
This command sends the current message to a remote host (via TCP using MLLP framing).
- Press F1 --> HL7 Tools: Send Message
- Enter the destination hostname (or IP address) and the port number. Click on the SendMessage button.
- Optionally select a socket encoding (defaults to UTF-8, or user defined default configured in settings file)
- Optionally require TLS for the connection to the remote host (remote host cert chain must be valid & trusted)
- The status will be displayed in the results text box. If an ACK message is returned it is also displayed. If no ACK is returned, the connection will time out and close (default timeout is 5 seconds). Status information will be cleared each time a message is sent.
Note: TLS connections will fail if the remote certificate name does not match the hostname provided, or the issuer of the certificate is not trusted. The trusted CA list is hardcoded by node.js, it is not based on workstation trusted CAs. If connecting to an endpoint using an certificate issued from an internal (enterprise) CA, then the user preferences will need to specify the root CA public key so that the remote certificate is trusted. This is set via the following preference containing an array of paths to all custom trusted root CAs (base64 encoded CER files):"
"hl7tools.TrustedCertificateAuthorities":[
"c:\\trustedCA\\enterprise-root-ca.cer",
"c:\\trustedCA\\another-root-ca.cer"
]
Other user preferences applicable to this function include:
// The TCP connection timeout (in seconds) when sending a HL7 message.
"hl7tools.ConnectionTimeout": 10
// The default socket encoding to apply when sending/receiving messages
// Select from "UTF-8" or "ISO-8859-1" (default to UTF-8 if not set)
"hl7tools.SocketEncoding: "UTF-8"
If the "hl7tools.FavouriteRemoteHosts" setting is populated you will be able to select the hosts from the 'favourites' drop-down list.
The 'favourite' endpoints are defined in settings.json - similar to the example below. Note the property names are case sensitive. It is an array of objects, each object defining the remote Hostname (string), Port (number), and a Description (string) of the endpoint. The UseTLS option (boolean) is optional.
// list of favourite endpoints
"hl7tools.FavouriteRemoteHosts": [
{
"Description": "Dev Server",
"Hostname": "127.0.0.1",
"Port": 5000,
"UseTLS: true"
},
{
"Description": "Test Server",
"Hostname": "127.0.0.1",
"Port": 6000
}
]
This command listens for HL7 messages sent from a remote host (via TCP using MLLP framing). All messages received are displayed in the editor as new documents.
- Press F1 --> HL7 Tools: Start Message Listener
- To stop receiving messages, Press F1 --> HL7 Tools: Stop Message Listener
User preferences applicable to this function include:
// Send a ACK in response to messages received (HL7 Tools: Start Message Listener).
"hl7tools.SendACK": true
// Send a specific ACK Code (MSA-1) in the response ("AA"|"AE"|"AR"|"CA"|"CE"|"CR"). Defaults to "CA"
"hl7tools.AckCode": "AA"
This function lists all fields from the currently selected segment in a hierarchial list. If a field contains components they will be displayed beneath the parent field. Any repeating field values are included.
- Select the segment in the message using the cursor.
- Press F1 --> HL7 Tools: Display Segment Fields.
- The selected segment's fields will be displayed in the output window.
- Press F1 --> HL7 Tools: Mask Identifiers.
- Common patient and next of kin identifiers will be masked with a '*' character. The masked message will be displayed in a new window, the original message will not be modified.
Note: The first identifier in the Patient Identifier List (PID-3) will remain, all additional identifiers in this list will be masked. It is assumed this ID isn't a named identifier and is still useful if messages need to be exchanged to troubleshoot issues. The aim is to mask common user identifiable fields such as name, address, etc. The full list of fields masked are:
- PID-3 (all repeats except for the first), PID-4 to PID-17, PID-19 to PID-23, PID-26 to PID-28
- NK1-2, NK1-4 to NK1-7, NK1-10 to NK1-16, NK1-19, NK1-20, NK1-25 to NK1-33, NK1-35, NK1-37, NK1-38
- All IN1 fields after IN1-2
- All IN2 fields after IN2-1
- All GT1 fields after GT1-2
If a single file contains multiple HL7 messages, this function splits each message into a new document. Each document is opened in the editor as an untitled file, it will be the user's responsibility to save the files. Large batch files will take longer to process.
- Press F1 --> HL7 Tools: Split HL7 Batch Files.
This command will open a copy of all segments the same name the one currently selected in the message. e.g. if the cursor is in the NK1 segment, all NK1 segments in the current file will be copied to a new window. Suited more for files containing multiple HL7 messages.
- move the cursor to the segment type you want to extract
- Press F1 --> HL7 Tools: Extract Matching Segments
This command is intended to be run when an imported file does not include delimiters between segments (i.e. segments are run together without a line break between each segment). This command will identify segments and add in line breaks. If line breaks are already present this will have no effect, it is only intended to correct malformed files.
- Press F1 --> HL7 Tools: Add Linebreaks to Segments
To activate this command every time a file is the active file in the editor, set the following user preference to true (defaults to false)
// Apply the command 'Add Linebreak to Segments' every time a file is active in the editor
"hl7tools.AddLinebreakOnActivation": true
Note: This is now implemented as a document formatter for HL7 files to align with vscode recommended practices. This still works as a separate function, but the recommended approach is to use the 'Format Document' function.
This command extracts values for a specified field from all open files.
Note:As a work around for a Visual Studio Code limitation, all open files will need to be cycled through the first time this command is run for any session.
- Press F1 --> HL7 Tools: Extract Fields from all Messages
- Wait for all open files to be cycled through (first time per session only)
- When prompted, enter the field to extract (e.g. PID-3)
This command confirms all required fields (as defined by the HL7 specification) are present in the message and contain values. If any required fields do not contain values, the field location and line number are reported in an output window.
Note: This does not consider conditional fields, nor does it examine the data type of any required fields to confirm if the value is valid.
- Press F1 --> HL7 Tools: Check Required Fields
- If all required fields are populated, an information message will appear. If any are missing, the field location and the corresponding line number from the file will be displayed in an output window.
Search the current message fields matching either a field description (e.g. 'Patient Name'), or a field location (e.g. 'PV1-3').
- Press F1 --> HL7 Tools: Find Field
- Enter the reference to the field location (e.g. PID-3), or enter part of the field name (e.g. patient).
- The matching field will be selected, with the cursor moved to the beginning of the field.
- Press F1 --> HL7 Tools: Find Next Field to find further matching fields in the message.
Press F1 and enter the ext install hl7tools command.
Clone the GitHub repository under your local extensions folder:
- Windows:
%USERPROFILE%\.vscode\extensions - Mac / Linux:
$HOME/.vscode/extensions
You can submit your issues and feature requests on the GitHub issues page. Test files, or detailed descriptions that can be used to reproduce the issue are appreciated.
- The HL7 syntax highlighting was sourced from https://github.com/craighurley/sublime-hl7-syntax
- the HL7 segment descriptions (segment.js) was extracted from http://github.com/fernandojsg/hl7-dictionary.
- Thanks to https://github.com/sherwanikhans for assistance on the 'Add Linebreak To Segments' command.
- Workaround to search all open files sourced from https://github.com/atishay/vscode-allautocomplete.
- User preference to specify the MSA-1 values in ACK messages was contributed by @jacobMedisoft (https://github.com/jacobMedisoft/)