Response

Definition

Responses can be used to capture values to perform subsequent requests, or add asserts to HTTP responses. Response on requests are optional, a Hurl file can just consist of a sequence of requests.

A response describes the expected HTTP response, with mandatory version and status, followed by optional headers, captures, asserts and body. Assertions in the expected HTTP response describe values of the received HTTP response. Captures capture values from the received HTTP response and populate a set of named variables that can be used in the following entries.

Example

hurl

GET https://example.org
HTTP 200
Last-Modified: Wed, 21 Oct 2015 07:28:00 GMT
[Asserts]
xpath "normalize-space(//head/title)" startsWith "Welcome"
xpath "//li" count == 18

Structure

<div class="hurl-structure-schema"> <div class="hurl-structure"> <div class="hurl-structure-col-0"> <div class="hurl-part-0"> HTTP 200 </div> <div class=" hurl-part-1"> content-length: 206 accept-ranges: bytes user-agent: Test </div> <div class="hurl-part-2"> [Captures] ... </div> <div class="hurl-part-2"> [Asserts] ... </div> <div class="hurl-part-3"> {

        &nbsp;&nbsp;"type": "FOO",

        &nbsp;&nbsp;"value": 356789,

        &nbsp;&nbsp;"ordered": true,

        &nbsp;&nbsp;"index": 10

        }
    </div>
</div>
<div class="hurl-structure-col-1">
    <div class="hurl-request-explanation-part-0">
        <a href="/docs/asserting-response.html#version-status">Version and status (mandatory if response present)</a>
    </div>
    <div class="hurl-request-explanation-part-1">

<a href="/docs/asserting-response.html#headers">HTTP response headers</a> (optional) </div> <div class="hurl-request-explanation-part-2">

    </div>
    <div class="hurl-request-explanation-part-2">
        <a href="/docs/capturing-response.html">Captures</a> and <a href="/docs/asserting-response.html#asserts">asserts</a> (optional sections, unordered)
    </div>
    <div class="hurl-request-explanation-part-2">
      

      

      

      

    </div>
    <div class="hurl-request-explanation-part-3">
        <a href="/docs/asserting-response.html#body">HTTP response body</a> (optional)
    </div>
</div>

</div> </div>

Capture and Assertion

With the response section, one can optionally capture value from headers, body, or add assert on status code, body or headers.

Body compression

Hurl outputs the raw HTTP body to stdout by default. If response body is compressed (using br, gzip, deflate), the binary stream is output, without any modification. One can use --compressed option to request a compressed response and automatically get the decompressed body.

Captures and asserts work automatically on the decompressed body, so you can request compressed data (using Accept-Encoding header by example) and add assert and captures on the decoded body as if there weren't any compression.

Timings

HTTP response timings are exposed through Hurl structured output (see --json), HTML report (see --report-html) and JSON report (see --report-json).

On each response, libcurl response timings are available:

time_namelookup: the time it took from the start until the name resolving was completed. You can use --resolve to exclude DNS performance from the measure.
time_connect: The time it took from the start until the TCP connect to the remote host (or proxy) was completed.
time_appconnect: The time it took from the start until the SSL/SSH/etc connect/handshake to the remote host was completed. The client is then ready to send its HTTP GET request.
time_starttransfer: The time it took from the start until the first byte was just about to be transferred (just before Hurl reads the first byte from the network). This includes time_pretransfer and also the time the server needed to calculate the result.
time_total: The total time that the full operation lasted.

All timings are in microsecond.

<a href="https://blog.cloudflare.com/a-question-of-timing/"><small>Courtesy of Cloudflare</small></a>

</div>