If the input schema is complex enough, then Schemathesis might be stuck and eventually die with out-of-memory.
Example schema - https://gist.githubusercontent.com/piotr-iohk/66270947ddc8fd31540b0377153fa411/raw/5af278fcf02414da17f4f8c622dcf182638034c1/gistfile1.txt
Many schemas from https://github.com/APIs-guru/openapi-directory cause the same effect.
My current assumption is that it is caused by the hypothesis-jsonschema canonicalisation approach, but I need to check details deeper.
https://github.com/Zac-HD/hypothesis-jsonschema/pull/69 might help.
Update. I was wrong that the problem is with the data generation layer. It actually happens when Schemathesis tries to inline all references
If the input schema is complex enough, then Schemathesis might be stuck and eventually die with out-of-memory.
Example schema - https://gist.githubusercontent.com/piotr-iohk/66270947ddc8fd31540b0377153fa411/raw/5af278fcf02414da17f4f8c622dcf182638034c1/gistfile1.txt
Many schemas from https://github.com/APIs-guru/openapi-directory cause the same effect.
My current assumption is that it is caused by the
hypothesis-jsonschemacanonicalisation approach, but I need to check details deeper.https://github.com/Zac-HD/hypothesis-jsonschema/pull/69 might help.
Update. I was wrong that the problem is with the data generation layer. It actually happens when Schemathesis tries to inline all references