Added automated schema to md generation

[bismuth01]

Added a pnpm script schema2md which generates .md files for all schemas in src/resources/schema and outputs them in src/resources/schema2md.

[ravenac95]

/validate b4f2701

[oso-prs]

Validation Results

⛔ Found 1 errors ⛔

commit b4f27016f87c561bf74a632fc09f873bcbcc8c23

---

Overview

0xbfbaa134ed5da64eca6b6fd39c04b273b6bb600d

• ⚠️ missing validator for factory on arbitrum_one

• ✅ 0xbfbaa134ed5da64eca6b6fd39c04b273b6bb600d is a 'contract' on arbitrum_one

0xe4666f0937b62d64c10316db0b7061549f87e95f

• ✅ 0xe4666f0937b62d64c10316db0b7061549f87e95f is a 'contract' on arbitrum_one

0x0ed37f546c7b70cd787eee94a521f24cd6653442

• ⚠️ missing validator for factory on arbitrum_one

• ✅ 0x0ed37f546c7b70cd787eee94a521f24cd6653442 is a 'contract' on arbitrum_one

0xb45913337aca4988b08a61388c25e48ba5f6227c

• ✅ 0xb45913337aca4988b08a61388c25e48ba5f6227c is a 'contract' on arbitrum_one

0xb5087f95643a9a4069471a28d32c569d9bd57fe4

• ✅ 0xb5087f95643a9a4069471a28d32c569d9bd57fe4 is a 'contract' on arbitrum_one

https://defillama.com/protocol/timeless-finance

• ✅ https://defillama.com/protocol/timeless-finance is a valid DefiLlama URL

• ✅ https://defillama.com/protocol/timeless-finance is a valid DefiLlama slug

https://defillama.com/protocol/bunni

• ✅ https://defillama.com/protocol/bunni is a valid DefiLlama URL

• ✅ https://defillama.com/protocol/bunni is a valid DefiLlama slug

https://defillama.com/protocol/fluid

• ❌ https://defillama.com/protocol/fluid is not a valid DefiLlama slug

• ✅ https://defillama.com/protocol/fluid is a valid DefiLlama URL

[ravenac95]

Hrm Untitled array is maybe not the most useful. I'm not sure that having Arrays as a top level item is generally useful either. I think we should instead only document the objects inside the array if they're object types and not any built in type.

[ravenac95]

Is this something we could change in jsonschema2md?

[bismuth01]

I don't think there is anything more jsonschema2md provides for such customization. They also have an example array schema with it's respective generated array readme and array items readme.

A script can be created to remove the generated array schema readme, but removing it from the base README can't seem to be automated.

However, I think custom scripts can be made to do the job if only a fixed types of tags are used.

[ryscheng]

These markdown files are not very useful right now, considering most fields are either "Untitled" or "Unknown".
Do you have a plan on how to address this? If not, I suggest just closing this PR

[bismuth01]

I can make a custom python script to create the required kind of readme for schemas if I am given some specifications. But opening a separate issue will be better to adhere to the topics of the issues and PRs.