This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
<test-case name="fn-unparsed-text-054" xmlns="http://www.w3.org/2010/09/qt-fots-catalog"> <description>Test the stability of the "unparsed-text" function</description> <created by="Michael Kay" on="2016-12-28"/> <test><![CDATA[ for $t1 in unparsed-text('http://date.jsontest.com') return every $i in 1 to 50 satisfies unparsed-text(translate(concat('http://date.jsontest.com', $i), '0123456789', '')) eq $t1 ]]></test> <result> <assert-true/> </result> </test-case> This test seems to depend on resolving an external http resource. There are two problems with this: (1) implementations aren't required to support this (2) this makes the test suite depend on external resources
Those criticisms are valid, but can you think of any other way of constructing a test case that demonstrates that the result is stable when the contents of the relevant resource are constantly changing?
No, unfortunately I can not. I suppose this may be yet another untestable aspect of our specifications.
If you really want to keep it, can you put it under a feature flag? I don't want my test harness to depend on an internet connection.
The WG agreed to add a new feature flag and this has been done. <dependencies> <dependency type="feature" name="remote_http"/> </dependencies>
This test case may return »This application is temporarily over its serving quota. Please try again later.«. @Michael: Could the number of http call be reduced?