This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
When trying to check http://anqam.com, the library aborts with the following trace: Exception in thread "main" org.w3c.mwi.mobileok.basic.TestException: Can't find doc content at org.w3c.mwi.mobileok.basic.AbstractCSSTestImplementation.getDocContentLineNumber(AbstractCSSTestImplementation.java:150) at org.w3c.mwi.mobileok.basic.AbstractCSSTestImplementation.generateResultsOnCSSContent(AbstractCSSTestImplementation.java:66) at org.w3c.mwi.mobileok.basic.AbstractCSSTestImplementation.runTest(AbstractCSSTestImplementation.java:56) at org.w3c.mwi.mobileok.basic.Tester.runTests(Tester.java:75) at org.w3c.mwi.mobileok.basic.Tester.main(Tester.java:211) For some reason, the (for sure existing) content of the document isn't written in the moki document. Earlier in the processing, the message "ATTENTION: No document" appears, which apparently isn't emitted by our code - this may be lead to follow.
The warning is from our code -- not sure if I like it, but whatever. The problem here is that the page is larger than 50KB so we ignore the body. I am ... not sure why I picked 50KB. That seems low, as a limit guarding against massive documents. I'll make it 500KB. The resulting error is a little funny because we don't have a document, but proceed with the test. This is a more general problem. I think we just don't run tests without a body, right? Again, this is just a question of what garbage we put out when we get garbage in. Partial, meaningless results are better than partial, meaningless results plus an exception.
I committed the updated to CVS. I also added code that will prevent the tests from running if there is no primary document -- but will still ensure that the "MAIN_DOCUMENT" test registers a failure.