curl/tests/data/test26

<testcase>
<info>
<keywords>
HTTP
HTTP GET
</keywords>
</info>
# Server-side
<reply>
<data>
HTTP/1.0 200 Mooo swsclose
Server: test-server/fake
Connection: close

*flopp*
</data>
</reply>

# Client-side
<client>
<server>
http
</server>
 <name>
specify more -o than URLs
 </name>
 <command>
http://%HOSTIP:%HTTPPORT/want/26 -o - -o -
</command>
</client>

# Verify data after the test has been "shot"
<verify>
<strip>
^User-Agent: curl.*
</strip>
<protocol>
GET /want/26 HTTP/1.1
User-Agent: curl/7.8.1-pre3 (sparc-sun-solaris2.7) libcurl 7.8.1-pre3 (OpenSSL 0.9.6a) (krb4 enabled)
Host: %HOSTIP:%HTTPPORT
Accept: */*

</protocol>
</verify>
</testcase>
Convert (most of) the test data files into genuine XML. A handful still are not, due mainly to the lack of support for XML character entities (e.g. & => & ). This will make it easier to validate test files using tools like xmllint, as well as edit and view them using XML tools. 2007-01-23 10:25:56 +08:00			`<testcase>`
keywords added 2005-04-17 07:15:19 +08:00			`<info>`
			`<keywords>`
			`HTTP`
			`HTTP GET`
			`</keywords>`
			`</info>`
nine new test cases 2001-08-07 20:42:23 +08:00			`# Server-side`
			`<reply>`
			`<data>`
Venkat Akella found out that libcurl did not like HTTP responses that simply responded with a single status line and no headers nor body. Starting now, a HTTP response on a persistent connection (i.e not set to be closed after the response has been taken care of) must have Content-Length or chunked encoding set, or libcurl will simply assume that there is no body. To my horror I learned that we had no less than 57(!) test cases that did bad HTTP responses like this, and even the test http server (sws) responded badly when queried by the test system if it is the test system. So although the actual fix for the problem was tiny, going through all the newly failing test cases got really painful and boring. 2006-11-25 21:32:04 +08:00			`HTTP/1.0 200 Mooo swsclose`
nine new test cases 2001-08-07 20:42:23 +08:00			`Server: test-server/fake`
			`Connection: close`

			`flopp`
			`</data>`
			`</reply>`

			`# Client-side`
			`<client>`
Each test case now specifies which server(s) it needs, without relying on the test number. 2003-05-01 04:25:39 +08:00			`<server>`
			`http`
			`</server>`
nine new test cases 2001-08-07 20:42:23 +08:00			`<name>`
			`specify more -o than URLs`
			`</name>`
			`<command>`
Now the test servers and test cases can run on a custom port number. There's no fixed port numbers in use anymore. Starting now, the default ports the servers use are 8990 - 8993. There's no option to modify these yet, but changing the $base option in the top of the runtests.pl script. 2004-09-08 16:08:38 +08:00			`http://%HOSTIP:%HTTPPORT/want/26 -o - -o -`
nine new test cases 2001-08-07 20:42:23 +08:00			`</command>`
A minor "syntax error" in numerous test files corrected 2005-01-26 05:45:03 +08:00			`</client>`
nine new test cases 2001-08-07 20:42:23 +08:00
			`# Verify data after the test has been "shot"`
			`<verify>`
strip off user-agent before checking protocol 2001-08-09 19:58:56 +08:00			`<strip>`
			`^User-Agent: curl.*`
			`</strip>`
nine new test cases 2001-08-07 20:42:23 +08:00			`<protocol>`
			`GET /want/26 HTTP/1.1`
			`User-Agent: curl/7.8.1-pre3 (sparc-sun-solaris2.7) libcurl 7.8.1-pre3 (OpenSSL 0.9.6a) (krb4 enabled)`
Replaced 127.0.0.1 with %HOSTIP where possible 2007-09-15 03:32:31 +08:00			`Host: %HOSTIP:%HTTPPORT`
Modified the default HTTP Accept: header to only be Accept: / 2004-02-09 20:46:41 +08:00			`Accept: /`
nine new test cases 2001-08-07 20:42:23 +08:00
			`</protocol>`
			`</verify>`
Convert (most of) the test data files into genuine XML. A handful still are not, due mainly to the lack of support for XML character entities (e.g. & => & ). This will make it easier to validate test files using tools like xmllint, as well as edit and view them using XML tools. 2007-01-23 10:25:56 +08:00			`</testcase>`