[BUG][Python] Adding custom model template with Python generator throws exception

[phemmer]

Bug Report Checklist

• Have you provided a full/minimal spec to reproduce the issue?
• Have you validated the input using an OpenAPI validator (example)?
• Have you tested with the latest master to confirm the issue still exists?
• Have you searched for related issues/PRs?
• What's the actual output vs expected output?
• [Optional] Sponsorship to speed up the bug fix or feature request (example)

Description

When adding a custom model template with the Python generator, an exception is thrown.

openapi-generator version

5.0.0

OpenAPI declaration file content or url

# api.yaml
openapi: 3.0.3
info:
  title: API
  description: API
  version: 1.0.0
servers:
  - url: 'https://localhost:80'
paths: {}
components:
  schemas:
    Test:
      type: object

Generation Details

Following the instructions from https://github.com/OpenAPITools/openapi-generator/blob/master/docs/customization.md:

# config.yaml
templateDir: templates
files:
  model-stub.mustache:
    templateType: Model
    destinationFilename: .pyi

$ java -jar openapi-generator-cli-5.0.0.jar generate -g python -c config.yaml -i api.yaml -o gen/
[main] INFO  o.o.codegen.DefaultGenerator - Generating with dryRun=false
[main] WARN  o.o.c.ignore.CodegenIgnoreProcessor - Output directory does not exist, or is inaccessible. No file (.openapi-generator-ignore) will be evaluated.
[main] INFO  o.o.codegen.DefaultGenerator - OpenAPI Generator: python (client)
[main] INFO  o.o.codegen.DefaultGenerator - Generator 'python' is considered experimental.
[main] INFO  o.o.c.l.PythonLegacyClientCodegen - Environment variable PYTHON_POST_PROCESS_FILE not defined so the Python code may not be properly formatted. To define it, try 'export PYTHON_POST_PROCESS_FILE="/usr/local/bin/yapf -i"' (Linux/Mac)
[main] INFO  o.o.c.l.PythonLegacyClientCodegen - NOTE: To enable file post-processing, 'enablePostProcessFile' must be set to `true` (--enable-post-process-file for CLI).
[main] INFO  o.o.c.languages.PythonClientCodegen - generateAliasAsModel is hard coded to true in this generator. Alias models will only be generated if they contain validations or enums
Exception in thread "main" java.lang.IllegalStateException: Duplicate key SupportingFile[templateFile='__init__model.mustache', folder='openapi_client/models', destinationFilename='__init__.py', canOverwrite=true]
	at java.util.stream.Collectors.lambda$throwingMerger$0(Collectors.java:133)
	at java.util.HashMap.merge(HashMap.java:1254)
	at java.util.stream.Collectors.lambda$toMap$58(Collectors.java:1320)
	at java.util.stream.ReduceOps$3ReducingSink.accept(ReduceOps.java:169)
	at java.util.ArrayList$ArrayListSpliterator.forEachRemaining(ArrayList.java:1384)
	at java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:482)
	at java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:472)
	at java.util.stream.ReduceOps$ReduceOp.evaluateSequential(ReduceOps.java:708)
	at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234)
	at java.util.stream.ReferencePipeline.collect(ReferencePipeline.java:566)
	at org.openapitools.codegen.DefaultGenerator.processUserDefinedTemplates(DefaultGenerator.java:941)
	at org.openapitools.codegen.DefaultGenerator.generate(DefaultGenerator.java:869)
	at org.openapitools.codegen.cmd.Generate.execute(Generate.java:432)
	at org.openapitools.codegen.cmd.OpenApiGeneratorCommand.run(OpenApiGeneratorCommand.java:32)
	at org.openapitools.codegen.OpenAPIGenerator.main(OpenAPIGenerator.java:66)

Steps to reproduce

1. Create the above api.yaml.
2. Create the above config.yaml.
3. Create an empty templates directory.
4. Run the above generate command.

Expected result

Should create a gen/openapi_client/model/test.pyi file according to templates/model-stub.mustache.

Related issues/PRs

Suggest a fix

It would appear that the use of templateType: model is causing the generator to add __init__model.mustache to the file list multiple times. There does not appear to be a way to prevent this.
As I am not redefining or providing a template for __init__model.mustache, it should not be trying to add it again.

[spacether]

So I am also unable to get this working:
Spec file (ensures the model is written)

# api.yaml
openapi: 3.0.3
info:
  title: API
  description: API
  version: 1.0.0
servers:
  - url: 'https://localhost:80'
paths:
  /pathA:
    post:
      operationId: pathA
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/Test'
        required: true
      responses:
        200:
          description: success
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Test'
components:
  schemas:
    Test:
      type: object
      properties:
        id:
          type: integer
        name:
          type: string
      # Both properties are required
      required:
        - id
        - name

Config file:

templateDir: templates_python
files:
    model_new.mustache:
        templateType: Model
        destinationFilename: Model.py

This did not work for python but it did work for Java
A work around is to use all the template files, and to modify the model template only

By eye it looks like the error is happening because we add these files twice:

• https://github.com/OpenAPITools/openapi-generator/blob/master/modules/openapi-generator/src/main/java/org/openapitools/codegen/languages/PythonClientCodegen.java#L130

• openapi-generator/modules/openapi-generator/src/main/java/org/openapitools/codegen/languages/PythonLegacyClientCodegen.java

  Line 291 in 008ddfb

  supportingFiles.add(new SupportingFile("__init__model.mustache", packagePath() + File.separatorChar + modelPackage, "__init__.py"));

[DavideFarina96]

Version: openapi-generator-cli 5.0.1-SNAPSHOT (using the provided docker container)

I am facing the same issue, but when adding a config file to generate a different controller, instead of a model. The code gets generated correctly with the default configurations, but it throws an error when using the following config file:

templateDir: /local/generated_code/openapi_server
files:
  controller.mustache:
    templateType: API
    destinationFilename: .py

Error:

[main] WARN  o.o.codegen.DefaultCodegen - Empty operationId found for path: PUT /_node-configs/{ncgroup}/{ncid}/_udata. Renamed to auto-generated operationId: _node_configsNcgroupNcidUdataPUT
Exception in thread "main" java.lang.IllegalStateException: Duplicate key SupportingFile[templateFile='__init__.mustache', folder='openapi_server/controllers', destinationFilename='__init__.py', canOverwrite=true]
        at java.util.stream.Collectors.lambda$throwingMerger$0(Collectors.java:133)
        at java.util.HashMap.merge(HashMap.java:1253)
        at java.util.stream.Collectors.lambda$toMap$58(Collectors.java:1320)
        at java.util.stream.ReduceOps$3ReducingSink.accept(ReduceOps.java:169)
        at java.util.ArrayList$ArrayListSpliterator.forEachRemaining(ArrayList.java:1374)
        at java.util.stream.AbstractPipeline.copyInto(AbstractPipeline.java:481)
        at java.util.stream.AbstractPipeline.wrapAndCopyInto(AbstractPipeline.java:471)
        at java.util.stream.ReduceOps$ReduceOp.evaluateSequential(ReduceOps.java:708)
        at java.util.stream.AbstractPipeline.evaluate(AbstractPipeline.java:234)
        at java.util.stream.ReferencePipeline.collect(ReferencePipeline.java:499)
        at org.openapitools.codegen.DefaultGenerator.processUserDefinedTemplates(DefaultGenerator.java:942)
        at org.openapitools.codegen.DefaultGenerator.generate(DefaultGenerator.java:870)
        at org.openapitools.codegen.cmd.Generate.execute(Generate.java:432)
        at org.openapitools.codegen.cmd.OpenApiGeneratorCommand.run(OpenApiGeneratorCommand.java:32)
        at org.openapitools.codegen.OpenAPIGenerator.main(OpenAPIGenerator.java:66)

[DavideFarina96]

One workaround to use the templates is the following:

• Get the default template files you want to use
• Modify them to suit your needs
• Put these templates (only the modified ones is enough) in a templates folder (any name will work)
• When generating the code, instead of passing the configuration defining the templateDir, pass the path to the directory with the -t parameter. For example, in my case (using the docker container), this looks like this:

docker-entrypoint.sh generate \
-i /local/GitLab_repos/SnapshotDatabase/snapshot-api/fs_api.yaml \
-g python-flask \
-t /local/templates/ \
-o /local/fs_api_gen

This workaround lets you use the templates you want, bypassing the issue which arises when using the config file to define where to get the templates from

[phemmer]

Note that only works when you want to modify/replace the existing template. The use case I had that triggered this issue was generating a separate file.