|
Not a single one of those will prevent the book from working on any decent epub reading software. They are all examples of the mostly useless schema validation checks I talked about previously. And run the autofix html tool in the editor to fix half the html related "errors" automatically.
What people often fail to grasp is that HTML parsing is now a robustly solved problem. Regardless of what mess you make of your HTML, all modern HTML renders will parse it with the same algorithm to give you the same final result. That algorithm is codified in the HTML 5 standard. Therefore, all you need to do is check your html against a single rendering engine and you are largely done. XHTML and it's accompanying schema checks are trying to solve a problem that no longer exists.
The problems that are very relevant today are deviations in different HTML renderers from what the standard specifies (and note I said renderer, not parser) or areas the standard does not cover. And epubcheck cannot help you with either of those, since, by its very nature it only checks against the standard.
About the only use I have ever found for epubcheck is to detect typos in code areas made during hand crafting an epub file. This can often point out mistakes that you would otherwise only detect by actually looking at the book in a renderer. Your last example, of preserveAspectRatio is one such case.
|