Fixed gzip automatic content-type assignment and added automatic compression header configuration #251
Conversation
…ression setting This implements the fix for detecting the proper content-type even when the file has the ".gz" extension. It further makes sure the compression headers are set properly if a "gz." file is detected, but the compression headers weren't explicitly set by the user.
…tent encoding header
src/microdot/microdot.py
Outdated
| @@ -774,7 +774,10 @@ def send_file(cls, filename, status_code=200, content_type=None, | |||
| first. | |||
| """ | |||
| if content_type is None: | |||
| ext = filename.split('.')[-1] | |||
| if filename.endswith('.gz'): | |||
There was a problem hiding this comment.
| if filename.endswith('.gz'): | |
| if compressed and filename.endswith('.gz'): |
I think it is best to rely on the compression argument here.
src/microdot/microdot.py
Outdated
| @@ -786,7 +789,7 @@ def send_file(cls, filename, status_code=200, content_type=None, | |||
| if max_age is not None: | |||
| headers['Cache-Control'] = 'max-age={}'.format(max_age) | |||
|
|
|||
| if compressed: | |||
| if compressed or filename.endswith('.gz'): | |||
There was a problem hiding this comment.
| if compressed or filename.endswith('.gz'): | |
| if compressed: |
I think this is fine as it was. If the user passes compressed=False then the response should not be compressed.
tests/test_response.py
Outdated
| self.assertEqual(res.status_code, 200) | ||
| self.assertEqual(res.headers['Content-Type'], 'text/plain') | ||
|
|
||
| res = Response.send_file('tests/files/test.txt.gz') |
There was a problem hiding this comment.
This test should use octet-stream as content file, because the compressed argument is False by default.
There should be another test that passes compressed=True and only then the we want the content type to be text and the content-encoding header to be present.
|
I do not like the idea of making it "automatic". If the call has |
…ng to the feedback.
|
Your suggestions make sense. I have modified the microdot.py and test files accordingly. |
This implements the fix for detecting the proper content-type even when the file has the ".gz" extension. It further makes sure the compression headers are set properly if a "gz." file is detected, but the compression headers weren't explicitly set by the user.