[python-experimental] Support schema property which has $ref to 'oneOf' schema

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.

[sebastien-rosset]

This PR adds support for use cases when a schema property is referencing another schema which is itself a oneOf.

PR checklist

• Read the contribution guidelines.
• If contributing template-only or documentation-only changes which will change sample output, build the project before.
• Run the shell script(s) under ./bin/ (or Windows batch scripts under.\bin\windows) to update Petstore samples related to your fix. This is important, as CI jobs will verify all generator outputs of your HEAD commit, and these must match the expectations made by your contribution. You only need to run ./bin/{LANG}-petstore.sh, ./bin/openapi3/{LANG}-petstore.sh if updating the code or mustache templates for a language ({LANG}) (e.g. php, ruby, python, etc).
• File the PR against the correct branch: master, 4.3.x, 5.0.x. Default: master.
• Copy the technical committee to review the pull request if your PR is targeting a particular programming language.

[sebastien-rosset]

@spacether , I don't think we currently have any sample or unit test that exercises the following use case: a schema property is referencing a oneOf schema. I've added a new Drawing schema that has a "mainShape" and that main shape is a $ref to Shape. I have also add a unit test that currently fails. IMO that unit test should pass.

[spacether]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

[sebastien-rosset]

@sebastien-rosset please take a look at this branch which I made from this PR.
In it I add the function is_valid_type to check if input models are present in the discriminator in the required data type.
So if we expect Shape and we pass in ScaleneTrianagle, it checks the discriminator in Shape to see if it can find ScaleneTrianagle.

My branch's update makes it so we can instantiate Drawings with oneOf class inputs for Shape and Triangle types.
How about including that fix here?

That's awesome, let me check.

[sebastien-rosset]

I have tested your proposed changes, this seems to work, which is great.
I did have to make a small change to convert the dict_keys to a list before accessing the item at index 0.

[sebastien-rosset]

There is a dependency on #6121 being merged before we can handle the 'null' type in oneOf. If we have a ShapeOrNull schema which is oneOf [type: 'null', Triangle, Quadrilateral] and in Drawing we have a secondary_shape property, then we should be able to assign None to Drawing.secondary_shape, whereas assigning None to Drawing.main_shape should raise an exception.

[spacether]

Thanks for the PR. Looks good to me.