Skip to content

support "conent-encoding: br" #599

New issue
New issue
Open
#620
Open
support "conent-encoding: br"#599
#620
@elonzh

Description

@elonzh
elonzh
opened on Jul 26, 2021

Currently vcrpy doesn't decompress response in Brotli compression format, but Brotli is well supported in HTTP clients like requests, httpx.

vcrpy/vcr/filters.py

Lines 127 to 162 in c79a06f

def decode_response(response):
"""
If the response is compressed with gzip or deflate:
1. decompress the response body
2. delete the content-encoding header
3. update content-length header to decompressed length
"""
def is_compressed(headers):
encoding = headers.get("content-encoding", [])
return encoding and encoding[0] in ("gzip", "deflate")
def decompress_body(body, encoding):
"""Returns decompressed body according to encoding using zlib.
to (de-)compress gzip format, use wbits = zlib.MAX_WBITS | 16
"""
if encoding == "gzip":
return zlib.decompress(body, zlib.MAX_WBITS | 16)
else: # encoding == 'deflate'
return zlib.decompress(body)
# Deepcopy here in case `headers` contain objects that could
# be mutated by a shallow copy and corrupt the real response.
response = copy.deepcopy(response)
headers = CaseInsensitiveDict(response["headers"])
if is_compressed(headers):
encoding = headers["content-encoding"][0]
headers["content-encoding"].remove(encoding)
if not headers["content-encoding"]:
del headers["content-encoding"]
new_body = decompress_body(response["body"]["string"], encoding)
response["body"]["string"] = new_body
headers["content-length"] = [str(len(new_body))]
response["headers"] = dict(headers)
return response

https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Content-Encoding#directives

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions