Models with different cases are mapped to the same file

[raymondfeng]

We can define models as Pet and pet. The runtime can handle that and treat them as different models. But the workspace maps both of them to pet.json. As a result, we lost one of two model definitions.

I understand the file mapping is designed to support operating system such as Mac or Windows which has case-insensitive file names by default.

Should we make model names case insensitive too?

[bajtos]

Here is my idea how to address this issue:

• Add a new property ModelDefinition.fileName, with the default value computed from the model name
• Modify ModelDefinition.getPath to use this new property
• Add a validation to enforce unique fileName within each facet

Should we make model names case insensitive too?

Not until we resolve strongloop/loopback#397, as the current workaround relies on case-sensitive model names. Once we have a proper solution, I am all for making model names case insensitive.

[ritch]

+1 for ModelDefinition.fileName.

[raymondfeng]

I had the impression that we already have the configFile property internally?

[bajtos]

configFile is a computed internal property, it includes facet path, etc. The API consumers must not depend on such details. That's why I want to add a fileName property that represents the last path part only, which is exactly what API consumers are interested in.

[ritch]

We might also be able to add ensureUnique('fileName', {within: 'facet'}) (forgot the exact syntax)...

[bajtos]

We might also be able to add ensureUnique('fileName', {within: 'facet'}) (forgot the exact syntax)...

👍

[bajtos]

On second though, are all these auto-computed properties worth the effort? Also the solution proposed here creates more work for workspace consumers, as they have to be aware of the new fileName property and implement some mechanism for providing a custom file name when the auto-generated one is not unique.

What if we implemented strongloop/loopback#397 first and then added validation that the model name must be unique within the facet, case-insensitively?

@ritch @raymondfeng

[bajtos]

@ritch @raymondfeng ping, what's your opinion on the comment above?

[bajtos]

I am proposing to implement a validation rule that the model name must me unique case-insensitively within the fact. It is the easiest possible solution to fix the problem and it will not require any extra work in downstream projects like Arc and generator-loopback.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

This issue has been closed due to continued inactivity. Thank you for your understanding. If you believe this to be in error, please contact one of the code owners, listed in the CODEOWNERS file at the top-level of this repository.