Stream: shorthand
Topic: SUSHI 0.12.0 (Ballot Edition)
Chris Moesel (Apr 09 2020 at 22:10):
SUSHI 0.12.0 aligns with the FHIR Shorthand 0.12.0 (May 2020 ballot) specification. All FHIR Shorthand features specified in the balloted language reference are supported in this release of SUSHI.
SUSHI 0.12.0 contains the following enhancements and bug fixes:
- Introduces streamlined usage for IG Publisher workflows
- Allows multiline strings for concept definitions (#319)
- Supports using an alias as the value in fixed value rules (#87)
- Sets minimum cardinality to 1 for fixed values that are discriminator paths (#301)
- Checks for supported FHIR version and minimally verifies that FHIR cache is properly loaded
- Fixes comment insertion bug that caused XML parsing issues in IG Publisher (#338)
- Fixes handling of nested required fixed values when generating instances (#300)
- Fixes parser support for non-breaking spaces (#315)
- Adds more random punny expressions for the CLI summary report
See the SUSHI 0.12.0 Release Notes for more details.
Bob Milius (Apr 12 2020 at 19:24):
I'm getting a new error with this version of sushi and using this fsh:
Alias: GeneGroupCS = http://www.genenames.org/genegroup
Alias: LNC = http://loinc.org
Profile: MyReport
Parent: DiagnosticReport
Id: mydiagnosticreport
Title: "My Diagnostic Report"
Description: "Profile of Diagnostic Report"
* code.coding ^slicing.discriminator.type = #pattern
* code.coding ^slicing.discriminator.path = "system"
* code.coding ^slicing.rules = #open
* code.coding ^slicing.description = "slicing system"
* code.coding contains
MasterGeneticPanel 1..1 and
HLAGeneGroup 0..1
* code.coding[MasterGeneticPanel].system = "http://loinc.org" (exactly)
* code.coding[MasterGeneticPanel].code = #81247-9 (exactly) // "Master HL7 genetic variant reporting panel"
* code.coding[HLAGeneGroup].system = "http://www.genenames.org/genegroup" (exactly)
* code.coding[HLAGeneGroup].code = #588 (exactly) // HLA Gene Group
Instance: MyReportExample
InstanceOf: mydiagnosticreport
Usage: #example
Title: "My Report Example"
Description: "Example of Report"
* code.coding[HLAGeneGroup] = GeneGroupCS#588 // "Histocompatibility complex (HLA)"
* code.coding[MasterGeneticPanel] = LNC#81247-9 // "Master HL7 genetic variant reporting panel"
* status = #final
* subject.type = "Patient"
* subject.identifier.system = "http://example.org/mysubject_identifers"
* subject.identifier.value = "1234"
The error from sushi is
error Cannot resolve element from path: subject.coding[MasterGeneticPanel].system
It seems like it's confusing the code.coding slice with subject.
I don't get this error from v0.11.1 or v0.11.2
Nick Freiter (Apr 13 2020 at 12:32):
This looks like a bug, I've logged it here https://github.com/FHIR/sushi/issues/357, and will work on getting a fix out. For the time being, does this actually change the output you get? From my experimentation it looked like it does not. So until we get the fix out feel free to just ignore that error message, assuming the output isn't actually changing.
Bob Milius (Apr 13 2020 at 13:55):
Actually they result in different builds. The one with 0.11.2 gives no errors, but 0.12.0 gives me 12.
with 0.11.2
Finished. 00:11.0988. Validation output in /Users/bob/src/fsh/test/build/output/qa.html Errors: 0 Warnings: 0 Info: 2 (00:31.0397) Done
with 0.12.0
Finished. 00:10.0723. Validation output in /Users/bob/src/fsh/test/build/output/qa.html Errors: 12 Warnings: 0 Info: 4 (00:27.0661) Done
Nick Freiter (Apr 13 2020 at 14:19):
Ah ok I see. Sorry about that. The fix is in progress and shouldn't be too difficult, but in the interim it's probably best to just stick with 0.11.2 if that is working well for you.
Bob Milius (Apr 13 2020 at 18:59):
actually I need to go back to 0.11.1 since my index is xml.
Chris Moesel (Apr 15 2020 at 19:13):
@ Bob Milius - this is fixed in the newly released SUSHI 0.12.1. Thanks for your patience!
Last updated: Apr 12 2022 at 19:14 UTC
