Added log for avoid silent incoherence between user input and app behaviour - non-existing fields

[Intrinsical-AI]

Added:

        # Solr default behavior, passing nonexistent fields results in a silent failure with no logging -- added logging
        self.schema_fields = {field['name'] for field in requests.get(urljoin(self.endpoint.encoded_string(), 'schema/fields')).json()['fields']}
        log.debug(f"Schema fields loaded: {len(self.schema_fields)} fields found.")

This: 1)notice in the logs which fields are present in the index. 2) Subset only the valid ones. Might need aditional logic to handle empty cases. let me know if you agree with the approach.

[Intrinsical-AI]

Test are not passing now since we have filtered the fields.. Incoming commit

[Intrinsical-AI]

Now test are passing. We stored a set of strings as a class attribute (self.schema_fields, that requests the Solr index (given by the endpoint parameter), the present fields.

Included debug logs following the constructor pattern.

[Intrinsical-AI]

@dantuzi Lmk what you think and I'll rebase if needed

[dantuzi]

I don't see self.schema_fields used somewhere. Am I wrong?

[dantuzi]

If this is needed only in this small piece of code, schema_fields should be a local variable

[Intrinsical-AI]

I think it might be useful for checking the payload when building the queries to the engine

[dantuzi]

It looks like this entire PR is for writing this string in the logs. Why do you think this is useful?

[Intrinsical-AI]

I as developer / user rather knowing when some fields are present or not, than just executing and expect to behave nice. The problem was the silent omission of fields if not present on the schema

[Intrinsical-AI]

Another option could be implementing this directly on the _search method. Maybe makes more sense. But I'd like to know when the searched fields are not present