You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Accept: text/html,*/* headers in request cause binary file (zip) downloads to be corrupted. Content type of response content-type: "application/octet-stream" is not being used to process the response correctly. These headers are automatically being set by Chrome. If I instead test the same request with curl ... -H 'Accept: */*' alternatively, I am able to download the same zip via the proxy without corrupting it.
Resulting downloads are larger than original files on the server. And unzip -t indicates it is a corrupted archive. In fact, using cat a.zip | iconv -f utf-8 -t utf-8 > /dev/null test on the corrupted file reveals they are in a valid UTF-8 format. This suggests that proxy may not be honoring the content-type response coming back from the web server. Instead it may be using "text/html" as the content type to process files being downloaded.
Attempting to rewrite the Accept headers to */* using following method also fails.
Accept: text/html,*/*
headers in request cause binary file (zip) downloads to be corrupted. Content type of responsecontent-type: "application/octet-stream"
is not being used to process the response correctly. These headers are automatically being set by Chrome. If I instead test the same request withcurl ... -H 'Accept: */*'
alternatively, I am able to download the same zip via the proxy without corrupting it.Here is the sequence of events:
Resulting downloads are larger than original files on the server. And
unzip -t
indicates it is a corrupted archive. In fact, usingcat a.zip | iconv -f utf-8 -t utf-8 > /dev/null
test on the corrupted file reveals they are in a valid UTF-8 format. This suggests that proxy may not be honoring the content-type response coming back from the web server. Instead it may be using "text/html" as the content type to process files being downloaded.Attempting to rewrite the Accept headers to
*/*
using following method also fails.Here is the code used to proxy the requests:
The text was updated successfully, but these errors were encountered: