1<testcase> 2<info> 3<keywords> 4HTTP 5HTTP POST 6</keywords> 7</info> 8# 9# Server-side 10<reply> 11<data> 12HTTP/1.1 302 OK 13Date: Thu, 09 Nov 2010 14:49:00 GMT 14Server: test-server/fake 15Location: 5990001 16Content-Length: 6 17Connection: close 18Content-Type: text/html 19 20-foo- 21</data> 22<data1> 23HTTP/1.1 200 OK 24Date: Thu, 09 Nov 2010 14:49:00 GMT 25Server: test-server/fake 26Transfer-Encoding: chunked 27Connection: close 28Content-Type: text/html 29 3032 31this data is slightly larger than the first piece 32 330 34 35</data1> 36 37<datacheck> 38HTTP/1.1 302 OK 39Date: Thu, 09 Nov 2010 14:49:00 GMT 40Server: test-server/fake 41Location: 5990001 42Content-Length: 6 43Connection: close 44Content-Type: text/html 45 46HTTP/1.1 200 OK 47Date: Thu, 09 Nov 2010 14:49:00 GMT 48Server: test-server/fake 49Transfer-Encoding: chunked 50Connection: close 51Content-Type: text/html 52 53this data is slightly larger than the first piece 54</datacheck> 55 56</reply> 57 58# Client-side 59<client> 60<server> 61http 62</server> 63# tool is what to use instead of 'curl' 64<tool> 65lib599 66</tool> 67 68# Bug 3510057 pointed out that when following a location to a larger chunked 69# encoded page, the first size would still be used in the progress callback 70 <name> 71HTTP GET with progress callback and redirects changing content sizes 72 </name> 73 <command> 74http://%HOSTIP:%HTTPPORT/599 75</command> 76</client> 77 78# 79# Verify data after the test has been "shot" 80<verify> 81 82</verify> 83</testcase> 84