Detect gzip encoding to manually decompress response, but 'Content-Encoding' header missing

问题

I am using net/http library in 'Go' to make an HTTP GET request. In the response, i get 12 headers. But when i run the exact same query through postman, i get 16 headers. One of those missing is 'Content-Encoding'. I understand this must be a CORS issue.

But since i have not set the header Accept-Encoding: gzip in my request, and i am still getting the gzip encoding in response, the Go transport is not automatically decompressing the response for me. So, i need to be able to manually detect the encoding and then decompress it. But, i cannot detect if the 'Content-Encoding' header is missing in the response.

Here is my code where i try to do this:

func calcDistanceAndDurationWithUberApi(originLat float64, originLon float64, destinationLat float64, destinationLon float64) (float64, float64, error) {

    endpoint := "https://api.uber.com/v1.2/estimates/price"
    parameters := fmt.Sprintf("?start_latitude=%v&start_longitude=%v&end_latitude=%v&end_longitude=%v", originLat, originLon, destinationLat, destinationLon)

    req, err := http.NewRequest("GET", endpoint + parameters, nil)
    if err != nil {
        return 0, 0, err
    }

    req.Header.Add("Authorization", "Token " + getUberApiKey())
    req.Header.Add("Accept-Language", "en_US")
    req.Header.Add("Content-Type", "application/json")

    httpClient := &http.Client{}
    resp, err := httpClient.Do(req)
    if err != nil {
        return 0, 0, err
    }
    if resp.StatusCode != 200 {
        return 0, 0, errors.NotFound("Response: %v", resp.StatusCode)
    }
    defer resp.Body.Close()

    pretty.Println("- REQUEST: ")
    pretty.Println(req)

    // Check if server sent gzipped response. Decompress if yes.
    var respReader io.ReadCloser
    switch resp.Header.Get("Content-Encoding") {
    case "gzip":
        fmt.Println("Content-Encoding is gzip")
        respReader, err = gzip.NewReader(resp.Body)
        defer respReader.Close()
    default:
        fmt.Println("Content-Encoding is Not gzip")
        respReader = resp.Body
    }

    pretty.Println("- RESPONSE HEADER: ")
    pretty.Println(resp.Header)

    pretty.Println("- RESPONSE BODY: ")
    pretty.Println(respReader)

    return 0, 0, nil
}

The response status is '200 OK'. Here is the output (Response):

- RESPONSE HEADER: 
http.Header{
    "Content-Language":          {"en"},
    "Cache-Control":             {"max-age=0"},
    "X-Uber-App":                {"uberex-nonsandbox", "optimus"},
    "Strict-Transport-Security": {"max-age=604800", "max-age=2592000"},
    "X-Content-Type-Options":    {"nosniff"},
    "Date":                      {"Fri, 19 May 2017 07:52:17 GMT"},
    "Content-Geo-System":        {"wgs-84"},
    "Connection":                {"keep-alive"},
    "X-Frame-Options":           {"SAMEORIGIN"},
    "X-Xss-Protection":          {"1; mode=block"},
    "Server":                    {"nginx"},
    "Content-Type":              {"application/json"},
}
- RESPONSE BODY: 
&http.gzipReader{
body: &http.bodyEOFSignal{
    body: &http.body{
        src: &internal.chunkedReader{
            r:  &bufio.Reader{
                buf: {0x48, 0x54, .......... }

回答1:

I gave in to the stubbornness of the uber api and added another request header, req.Header.Add("Accept-Encoding", "gzip").

Now i am getting the response header "Content-Encoding": "gzip", although i am still getting an undecipherable response body, but that's beyond the scope of this question.

来源：https://stackoverflow.com/questions/44064556/detect-gzip-encoding-to-manually-decompress-response-but-content-encoding-hea

标签

http-headers

cors

httpresponse

content-encoding