curl/tests/data/test38
Daniel Stenberg 5d9fc28fa7 Modified the default HTTP headers used by libcurl:
A) Normal non-proxy HTTP:

 - no more "Pragma: no-cache" (this only makes sense to proxies)

B) Non-CONNECT HTTP request over proxy:

 - "Pragma: no-cache" is used (like before)
 - "Proxy-Connection: Keep-alive" (for older style 1.0-proxies)

C) CONNECT HTTP request over proxy:

 - "Host: [name]:[port]"
 - "Proxy-Connection: Keep-alive"
2005-05-11 09:52:59 +00:00

60 lines
1007 B
Plaintext

<info>
<keywords>
HTTP
HTTP GET
Resume
FAILURE
</keywords>
</info>
# Server-side
<reply>
<data nocheck=1>
HTTP/1.0 200 Mooo
Date: Mon, 13 Nov 2000 13:41:09 GMT
Server: myown/1.0
Connection: close
todelooooo lalalala yada yada, we know nothing about ranges ;-)
</data>
</reply>
# Client-side
<client>
<server>
http
</server>
<name>
HTTP resume request without server supporting it
</name>
<command option="no-output">
http://%HOSTIP:%HTTPPORT/want/38 -C - -i -o log/fewl.txt
</command>
<file name="log/fewl.txt">
This text is here to simulate a partly downloaded file to resume
download on.
</file>
</client>
# Verify data after the test has been "shot"
<verify>
<errorcode>
33
</errorcode>
<strip>
^User-Agent:.*
</strip>
<protocol>
GET /want/38 HTTP/1.1
Range: bytes=78-
Host: 127.0.0.1:%HTTPPORT
Accept: */*
</protocol>
# the download target file must remain untouched
<file name="log/fewl.txt">
This text is here to simulate a partly downloaded file to resume
download on.
</file>
</verify>