Skip to content

Conversation

@DavidGrath
Copy link
Contributor

@DavidGrath DavidGrath commented Mar 18, 2025

Added test givenObjectHasAdditionalPropertiesWhenGenerateThenIndexSignatureNotUsedToGenerateMethodName to verify that it works. Also opened the newly generated folder in VSCode and found no syntax errors.
@joscha please help review

to close #20195

PR checklist

  • Read the contribution guidelines.
  • Pull Request title clearly describes the work in the pull request and Pull Request description provides details about how to validate the work. Missing information here may result in delayed response from the community.
  • Run the following to build the project and update samples:
    ./mvnw clean package || exit
    ./bin/generate-samples.sh ./bin/configs/*.yaml || exit
    ./bin/utils/export_docs_generators.sh || exit
    
    (For Windows users, please run the script in Git BASH)
    Commit all changed files.
    This is important, as CI jobs will verify all generator outputs of your HEAD commit as it would merge with master.
    These must match the expectations made by your contribution.
    You may regenerate an individual generator by passing the relevant config(s) as an argument to the script, for example ./bin/generate-samples.sh bin/configs/java*.
    IMPORTANT: Do NOT purge/delete any folders/files (e.g. tests) when regenerating the samples as manually written tests may be removed.
  • File the PR against the correct branch: master (upcoming 7.x.0 minor release - breaking changes with fallbacks), 8.0.x (breaking changes without fallbacks)
  • If your PR is targeting a particular programming language, @mention the technical committee members, so they are more likely to review the pull request.

@joscha
Copy link
Contributor

joscha commented Mar 19, 2025

refs #20195

Copy link
Contributor

@joscha joscha left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎉

|sortModelPropertiesByRequiredFlag|Sort model properties to place required parameters before optional parameters.| |true|
|sortParamsByRequiredFlag|Sort method arguments to place required parameters before optional parameters.| |true|
|sourceFolder|source folder for generated code| |OpenAPI/src|
|sourceFolder|source folder for generated code| |OpenAPI\src|
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We should revert these - probably because regeneration was done on windows?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hello, thank you for the feedback.
I also think it was because of Windows. I'll try using a Linux machine instead to effectively undo my own changes

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you can use WSL instead

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you. Done

you can use WSL instead

{{^isEnumRef}}
{{^withoutRuntimeChecks}}
formParams.append('{{baseName}}', new Blob([JSON.stringify({{{dataType}}}ToJSON(requestParameters['{{paramName}}']))], { type: "application/json", }));
formParams.append('{{baseName}}', new Blob([JSON.stringify({{returnType}}ToJSON(requestParameters['{{paramName}}']))], { type: "application/json", }));
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we add a fixture (generated sample) to the repository that shows this change? Or have the samples not been regenerated?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the only real changes I got from running generate-samples were CRLF/LF changes.
Tried again recently and this is pretty much all the output:
image

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@joscha Okay, so I retried on WSL and it seems the package-lock.json was the only real change

@DavidGrath
Copy link
Contributor Author

Good Day @joscha ,
Do you think it's good to go now?

Copy link
Contributor

@joscha joscha left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice. Will need @macjohnny to merge.

@wing328 wing328 changed the title Fix issue #20195 [typescript-fetch] to fix incorrect parsing with additional properties Mar 23, 2025
@wing328 wing328 added this to the 7.13.0 milestone Mar 23, 2025
@macjohnny macjohnny merged commit 0becb3f into OpenAPITools:master Mar 24, 2025
15 checks passed
{{^isEnumRef}}
{{^withoutRuntimeChecks}}
formParams.append('{{baseName}}', new Blob([JSON.stringify({{{dataType}}}ToJSON(requestParameters['{{paramName}}']))], { type: "application/json", }));
formParams.append('{{baseName}}', new Blob([JSON.stringify({{returnType}}ToJSON(requestParameters['{{paramName}}']))], { type: "application/json", }));

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this seems to be only working when additionalProperties exists

it is causing an issue where the param (paramName here) has it's own schema

for the case shown below, it puts in RequestPayload but instead it should be RequestContext

e.g.

requestBody:
  description: entity for which custom field will be added such as fund, deal, asset, investor etc
  content:
    multipart/form-data:
      schema:
        $ref: "#/components/schemas/RequestPayload"

...

RequestPayload:
  type: object
  properties:
    context:
      $ref: "#/components/schemas/RequestContext"
  ...

RequestContext:
  type: object
  properties:
    ...

@macjohnny @joscha

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@paras-fundwave thanks for reporting! would you want to send a PR to fix that?

Copy link

@paras-fundwave paras-fundwave Jul 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sure thing!

while I'll familiarise myself with the codebase, can you help confirm the approach I have in mind

we'll need to have 2 outputs (for this line) depending on whether additionalProperties exists

from this PR I can see I'll need to add test-cases too, anything else I'm missing that can help but might not be mentioned in the contributing.md?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sg, no objections, although i didnt look into the issue in detail

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks like how datatype is processed has changed too in b/w v7.0.0 and v7.1.0

reverting to dataType is still leads to malformed code - broken v7.1.0 onwards

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, I see. It seems like an inverted mustache section might come in handy. Were you able to spot any other issues around this? Thanks for pointing this out

Copy link
Contributor

@adrianhj adrianhj Jul 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seeing similar issues on 7.14.0 with a discriminated schema where returnType is null and ends up generating invalid code: new Blob([JSON.stringify(ToJSON(...)], ...), dataType generates as expected with new Blob([JSON.stringify(MyDiscriminatedSchemaToJSON(...)], ...)

7.12.0 generates OK, 7.13.0 onwards is broken.

I haven't read up on the original issue fully -- but as the line in question is part of building up the request, using the return type for serializing request parameters seems wrong?

Copy link
Contributor

@adrianhj adrianhj Jul 10, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Diving into the test I don't think the assertion is right.

The schema for the request body is:

"multipart/form-data": {
  "schema": {
    "type": "object",
    "properties": {
      "paRes": {
        "type": "object",
        "additionalProperties": {
          "type": "string"
        },
        "description": "Form sent for Authentication"
      }
    }
  },
  "encoding": {
    "paRes": {
      "style": "form"
    }
  }
}

ResponseOfString is a schema used as the response:

"responses": {
  "200": {
    "description": "OK",
    "content": {
      "application/json": {
        "schema": {
          "$ref": "#/components/schemas/ResponseOfString"
        }
      }
    }
  }
}

The code that gets generated to append the part for the form in the request:

if (requestParameters['paRes'] != null) {
            formParams.append('paRes', new Blob([JSON.stringify(ResponseOfStringToJSON(requestParameters['paRes']))], { type: "application/json", }));
                    }

But that is surely wrong, as ResponseOfString is the response of the operation, not related to the request body?

I have raised a PR to revert: #21542 (FYI @macjohnny)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[BUG] [typescript-fetch] Incorrect parsing with additional properties

6 participants