fix: GzipParser passes through non-gzipped data instead of raising BadGzipFile

[devin-ai-integration]

Summary

GzipParser.parse documented that "if the data is not gzipped, pass the data to the inner parser as is", but never implemented it — it unconditionally wrapped the stream in gzip.GzipFile, so any layer that isn't actually gzip crashes with BadGzipFile: Not a gzipped file (...).

This bites declarative connectors that stack decoders, because create_gzip_decoder builds a header-selected decoder whose fallback strips exactly one gzip layer:

# GzipDecoder -> GzipDecoder -> CsvDecoder  becomes:
#   header match -> GzipParser(GzipParser(CsvParser))   # 2 gzip layers assumed
#   fallback     -> GzipParser(CsvParser)               # exactly 1 gzip layer assumed

When a response intermittently arrives with fewer gzip layers than configured (or as an unexpected format, e.g. a zip archive with magic b'PK'), the rigid parser raises BadGzipFile and the stream fails. This surfaced as an intermittent BadGzipFile: Not a gzipped file (b'PK') on a custom App Store Connect connector's subscription_event_report stream.

Fix: peek the leading bytes (via io.BufferedReader, which does not consume the stream and works uniformly over urllib3 raw responses, GzipFile, and BytesIO) and only gzip-decompress when the gzip magic number is present; otherwise hand the stream straight to the inner parser.

buffered = data if isinstance(data, io.BufferedReader) else io.BufferedReader(data)
if buffered.peek(2)[:2] == b"\x1f\x8b":
    with gzip.GzipFile(fileobj=buffered, mode="rb") as gz:
        yield from self.inner_parser.parse(gz)
else:
    yield from self.inner_parser.parse(buffered)

This makes nested/fallback gzip decoding tolerant of 0, 1, or N gzip layers and is fully backward-compatible when data is gzipped. Note: a genuinely zip-compressed (b'PK') payload still needs a ZipfileDecoder for that layer — this change stops the hard crash but does not unzip.

Tests

• New unit tests in test_composite_decoder.py: single GzipParser passthrough on uncompressed data; nested GzipParser(GzipParser(CsvParser)) handling both one and two gzip layers; header-based fallback to a GzipParser on a non-gzipped response.
• Existing decoder + streaming memory-usage tests still pass.

Link to Devin session: https://app.devin.ai/sessions/862184b2ca174e4483b2cc1b5264e978

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[github-actions]

👋 Greetings, Airbyte Team Member!

Here are some helpful tips and reminders for your convenience.

💡 Show Tips and Tricks

Testing This CDK Version

You can test this version of the CDK using the following:

# Run the CLI from this branch:
uvx 'git+https://github.com/airbytehq/airbyte-python-cdk.git@devin/1783966700-gzip-parser-passthrough#egg=airbyte-python-cdk[dev]' --help

# Update a connector to use the CDK from this branch ref:
cd airbyte-integrations/connectors/source-example
poe use-cdk-branch devin/1783966700-gzip-parser-passthrough

PR Slash Commands

Airbyte Maintainers can execute the following slash commands on your PR:

• /autofix - Fixes most formatting and linting issues
• /poetry-lock - Updates poetry.lock file
• /test - Runs connector tests with the updated CDK
• /prerelease - Triggers a prerelease publish with default arguments
• /poe build - Regenerate git-committed build artifacts, such as the pydantic models which are generated from the manifest JSON schema in YAML.
• /poe <command> - Runs any poe command in the CDK environment

📚 Show Repo Guidance

Helpful Resources

• CDK API Reference

📝 Edit this welcome message.

[github-actions]

PyTest Results (Fast)

4 132 tests  +5   4 120 ✅ +5   7m 47s ⏱️ +6s
    1 suites ±0      12 💤 ±0 
    1 files   ±0       0 ❌ ±0 

Results for commit 73f1521. ± Comparison against base commit dd9558c.

[github-actions]

PyTest Results (Full)

4 135 tests  +5   4 123 ✅ +5   11m 42s ⏱️ + 4m 27s
    1 suites ±0      12 💤 ±0 
    1 files   ±0       0 ❌ ±0 

Results for commit 73f1521. ± Comparison against base commit dd9558c.