This is a continuation of issue #75 which I closed because it pointed to two tests with false positives, one of which has been resolved.
But test document 10 still gives a false positive for WCAG 4.1.2 test 15. The algorithm points to three distinct lines which are clearly headers (at least to the human eye) and says they appear to be paragraphs.
Even if the font does not give clear clues, there are other signs which do. Each is a single word line with a number, and each one is indented. I think we need to look a little more at how the algorithm is tuned.
This is a continuation of issue #75 which I closed because it pointed to two tests with false positives, one of which has been resolved.
But test document 10 still gives a false positive for WCAG 4.1.2 test 15. The algorithm points to three distinct lines which are clearly headers (at least to the human eye) and says they appear to be paragraphs.
Even if the font does not give clear clues, there are other signs which do. Each is a single word line with a number, and each one is indented. I think we need to look a little more at how the algorithm is tuned.