Open shonfeder opened 1 year ago
The current formatting for ITF traces makes them hard for a human to read, simply due to unnecessary line breaks and indentation. E.g., here's an example of two states from a trace in the current formatting
{ "#meta": { "index": 0 }, "Names": { "#set": [ "bar", "baz", "foo" ] }, "paths": { "#set": [ [ { "tag": "Dir", "value": "." } ] ] } }, { "#meta": { "index": 1 }, "Names": { "#set": [ "bar", "baz", "foo" ] }, "paths": { "#set": [ [ { "tag": "Dir", "value": "." }, { "tag": "Dir", "value": "foo" } ], [ { "tag": "Dir", "value": "." } ] ] } },
And here's the same trace formatted by Emac's json-mode:
json-mode
{ "#meta": { "index": 0 }, "Names": { "#set": ["bar", "baz", "foo"] }, "paths": { "#set": [[{ "tag": "Dir", "value": "." }]] } }, { "#meta": { "index": 1 }, "Names": { "#set": ["bar", "baz", "foo"] }, "paths": { "#set": [ [ { "tag": "Dir", "value": "." }, { "tag": "Dir", "value": "foo" } ], [{ "tag": "Dir", "value": "." }] ] } },
I assume this is limited by ujson formatting? Perhaps we can just rely on people using external tools to format their JSON?
ujson
The current formatting for ITF traces makes them hard for a human to read, simply due to unnecessary line breaks and indentation. E.g., here's an example of two states from a trace in the current formatting
And here's the same trace formatted by Emac's
json-mode
: