[chipkent]

Preview only. This is a draft in the fork showing the exact text of the PR planned for massive-com/client-python. It will be closed, not merged, once the upstream PR is opened.

Fixes #<issue: F5c>

Builds on #12 (BadResponse.status).

When a 200 response body is not valid JSON (for example an HTML page from a proxy, or a truncated body), BaseClient._get and _paginate_iter log the decode error and return []. A get_* call returns [] instead of a model, and a list_* call yields nothing if the first page is bad, or stops early if a later page is bad. Either way the caller gets missing data with no error.

• _decode now raises BadResponse with status=resp.status and the first 200 bytes of the body; the json error is chained as __cause__.
• _get and _paginate_iter drop their try/except and let it propagate. These are the only two places that decode a response.
• The BadResponse docstring now covers this case.

Valid responses are unaffected. A call that used to return [] for a body it could not decode now raises BadResponse, the exception callers already handle for non-200 responses.

try:
    trades = list(client.list_trades("AAPL"))
except BadResponse as e:
    print(e.status, e)  # 200 Invalid JSON in response: <html>...

Tests in test_rest/test_bad_response.py use pook to serve a non-JSON body to a get_* call, to the first page of a list_* call, and to its second page. All three fail without the fix.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

_{Comment @coderabbitai help to get the list of available commands.}