Stream: committers
Topic: local build error
Melva Peters (Jul 16 2016 at 16:07):
I've updated to the revision 9008 but am getting the following error "Compile java failed". I'm running this on a Mac. I've tried to delete the .class files using find . -name ‘*.class’ -type f -delete. Does anyone have any suggestions?
Lloyd McKenzie (Jul 16 2016 at 16:25):
Can you paste the full compile message?
Melva Peters (Jul 16 2016 at 16:26):
[java] Errors: 0 Warnings: 71 Hints: 903 [java] [java] ===Resources with FMM forced to 0=== [java] 40.042 85sec 936MB [java] ERROR @ Java Compile warning: [options] bootstrap class path not set in conjunction with -source 1.6 (src = Publisher) [java] ERROR @ Java Compile /Users/mpeters/Documents/FHIR/trunk/build/implementations/java/org.hl7.fhir.dstu3/src/org/hl7/fhir/dstu3/model/NutritionOrder.java:3329: error: cannot find symbol [java] return ResourceType.NutritionOrder; [java] ^ [java] symbol: variable NutritionOrder [java] location: class org.hl7.fhir.dstu3.model.ResourceType (src = Publisher) [java] ERROR @ Java Compile /Users/mpeters/Documents/FHIR/trunk/build/implementations/java/org.hl7.fhir.dstu3/src/org/hl7/fhir/dstu3/model/DiagnosticOrder.java:2041: error: cannot find symbol [java] return ResourceType.DiagnosticOrder; [java] ^ [java] symbol: variable DiagnosticOrder [java] location: class org.hl7.fhir.dstu3.model.ResourceType (src = Publisher) [java] ERROR @ Java Compile Note: Some input files use or override a deprecated API. (src = Publisher) [java] ERROR @ Java Compile Note: Recompile with -Xlint:deprecation for details. (src = Publisher) [java] ERROR @ Java Compile Note: Some input files use unchecked or unsafe operations. (src = Publisher) [java] ERROR @ Java Compile Note: Recompile with -Xlint:unchecked for details. (src = Publisher) [java] This was a Full Build 0.119 85sec 936MB [java] FHIR build failure @ Sat, Jul 16, 2016 09:26-0700 0.72 86sec 943MB [java] Error: Compile java failed [java] java.lang.Exception: Compile java failed [java] at org.hl7.fhir.tools.publisher.Publisher.produceSpecification(Publisher.java:1946) [java] at org.hl7.fhir.tools.publisher.Publisher.execute(Publisher.java:623) [java] at org.hl7.fhir.tools.publisher.Publisher.main(Publisher.java:472)
BUILD FAILED
/Users/mpeters/Documents/FHIR/trunk/build/build.xml:30: The following error occurred while executing this line:
/Users/mpeters/Documents/FHIR/trunk/build/tools/java/org.hl7.fhir.tools.core/build.xml:123: Java returned: 1
Melva Peters (Jul 16 2016 at 16:53):
Thanks to @Lloyd McKenzie I've fixed my build. If you run into this issue, try deleting implementations/java and then doing an update and re-running
Grahame Grieve (Jul 16 2016 at 21:09):
when you rename resources, you must add the old name of the resource to '[removed-resources]' in fhir.ini
Grahame Grieve (Jul 16 2016 at 21:09):
then this problem will not happen
Eric Haas (Jul 17 2016 at 13:49):
Aha that is why... I have always had to delete the file directly when renaming
Eric Haas (Jul 18 2016 at 00:38):
this is an error I have not seen before: ERROR @ Java Compile warning: [options] bootstrap class path not set in conjunction with -source 1.6 (src = Publisher) after a successful partial build
Eric Haas (Jul 18 2016 at 00:39):
plus many more
Grahame Grieve (Jul 18 2016 at 00:39):
I haven't seen that before, but that's just a warning, not an error/
Grahame Grieve (Jul 18 2016 at 00:44):
it's what follows that are your errors
Eric Haas (Jul 18 2016 at 01:54):
I'll try reloading tools. first
Eric Haas (Jul 18 2016 at 02:20):
now the build says
" ==============!! Running without terminology server !!=============="
and hangs at this [java] Use Tx Server from BWS for value set http://hl7.org/fhir/ValueSet/kos-title on http://nema.org/dicom/dicm
Eric Haas (Jul 18 2016 at 02:21):
I'm definately not going to make the freeze deadline - too many issues like this I'm afraid
Grahame Grieve (Jul 18 2016 at 02:25):
I've just restarted my server - should fix the hang
Eric Haas (Jul 18 2016 at 02:25):
thanks
Eric Haas (Jul 18 2016 at 02:25):
and lke magic ...
Grahame Grieve (Jul 18 2016 at 02:26):
how badly going to miss the deadline?
Eric Haas (Jul 18 2016 at 02:29):
Trying to update a single resoruce to align with the workflow template has taken most of day. I think this one is the toughest - Diagnostic Order. but I'll need to see how the other's go.
Grahame Grieve (Jul 18 2016 at 02:30):
ok
Eric Haas (Jul 18 2016 at 02:30):
[java] c: /Users/ehaas/Documents/FHIR/build/implementations/java/org.hl7.fhir.dstu3/src/org/hl7/fhir/dstu3/model/DiagnosticRequest.java:294: error: '}' expected
[java] @Child(name = "context", type = {Encounter.class, Episode of Care.class}, order=9, min=0, max=1, modifier=false, summary=true)
[java]
and many more like these - I don;t know where to look - the partial build works
Grahame Grieve (Jul 18 2016 at 02:32):
"Episode of Care" - looks like you have that somewhere in your source as a Resource Type on a Reference
Grahame Grieve (Jul 18 2016 at 02:32):
instead of EpisodeOfCare
Eric Haas (Jul 18 2016 at 02:34):
Indeed, I would never found that
Grahame Grieve (Jul 18 2016 at 02:34):
it should be picked up by the resource validator. I'll have to look into why
Lloyd McKenzie (Jul 18 2016 at 03:27):
I'm finally getting to PC changes. I'll continue committing as long as you'll let me, but I expect most, if not all, of the workflow alignment changes will need to be handled as a "note to balloters". Project work that I'd hoped would be done Thursday finished an hour ago :(
Michelle (Moseman) Miller (Apr 21 2017 at 21:34):
I am getting the following LOCAL build error
[java] ERROR: Unable to process warnings: Unable to create warnings file - one or more profiles-* files unavailable or invalid
I'm curious if this error is preventing me from seeing a more meaningful error. I can get this error to go away if I remove one of the files I edited (condition-spreadsheet.xml), so I know the deeper root cause is in that file somewhere. That said, I don't see any insightful errors telling me what might be wrong with the spreadsheet.
Grahame Grieve (Apr 21 2017 at 21:36):
@Lloyd McKenzie I thought we fixed that?
Lloyd McKenzie (Apr 21 2017 at 21:45):
Well, we fixed it in that it'll now spew to console all messages it knows about in an unsorted/unorganized fashion, but we can't get it to run the pretty formatting process if the build errors out before the profiles-* files are created.
Michelle (Moseman) Miller (Apr 21 2017 at 21:46):
I'm guessing it might be due to a new element added without a corresponding example (for FMM=3 resource), but still would be nice (i.e. more efficient) to have a meaningful error to help me troubleshoot.
Grahame Grieve (Apr 21 2017 at 21:52):
can you send your whole log? there should still be the actual error message in there
Michelle (Moseman) Miller (Apr 21 2017 at 22:40):
OK, I think I finally found it. The main gotcha for me was my cmd window buffer wasn't set high enough to see the error much, much, much earlier in the log. Darn RIM mapping is preventing me from committing the change.
[java] INFORMATION @ Condition.stage.type: RIM Mapping is required (src = ResourceValidator)
[java] ERROR @ Condition: Resource Condition (FMM=3) cannot have an FMM level >2 (3) if it has informational hints (src = ResourceValidator)
Last updated: Apr 12 2022 at 19:14 UTC
