You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I couldn't find a conformance suite at analogous to the W3C DOM conformance tests at http://www.saxproject.org/, but Elliotte Rusty Harold has a conformance suite at http://cafeconleche.org/SAXTest/. The fixtures are XML and the test suite can probably be very easily duplicated in JavaScript.
As for the whinging in #4, if the SAX conformance test doesn't say anything about error handling, we could write a fixture generator in Java, one that parses XML and emits errors, with line and column number and we could use that as a fixture to determine when the SAX parser should emit an error. Simply choose the best Java based SAX parser according Elliotte's conformance suite and make that parser the fixture.
As an aside, have a look at the test output of a Travis CI run of Timezone, and toward the end you'll see a test with 35950 assertions, one for the millisecond of and the millisecond before every clock transition in IANA Timezone Database. I run it again against the minified source. I'm mentioning this as an example of a fixture based test.
Update: Fixed link to Timezone Travis CI output.
The text was updated successfully, but these errors were encountered:
I couldn't find a conformance suite at analogous to the W3C DOM conformance tests at http://www.saxproject.org/, but Elliotte Rusty Harold has a conformance suite at http://cafeconleche.org/SAXTest/. The fixtures are XML and the test suite can probably be very easily duplicated in JavaScript.
Note that I'm using fixtures to mean data driven and apply the Ruby definition of fixture.
As for the whinging in #4, if the SAX conformance test doesn't say anything about error handling, we could write a fixture generator in Java, one that parses XML and emits errors, with line and column number and we could use that as a fixture to determine when the SAX parser should emit an error. Simply choose the best Java based SAX parser according Elliotte's conformance suite and make that parser the fixture.
As an aside, have a look at the test output of a Travis CI run of Timezone, and toward the end you'll see a test with 35950 assertions, one for the millisecond of and the millisecond before every clock transition in IANA Timezone Database. I run it again against the minified source. I'm mentioning this as an example of a fixture based test.
Update: Fixed link to Timezone Travis CI output.
The text was updated successfully, but these errors were encountered: