Quote:
Originally Posted by Sirtel
No, it doesn't. I don't know, maybe it did 15 years ago or so, and it sure changes some things in the css, but it doesn't "trash" anything. I use calibre conversion all the time, because it does many useful things - converts inline styles to css, merges identical css classes, simplifies unnecessarily convoluted css and so on. It's an extremely rare occurrence when it "trashes" anything.
Yes, if you want your css to be identical to the original, conversion is not for you. But that's a far cry from trashing.
|
To me, flattening the structure & merging CSS stylesheets can result in the stylesheets being merged in the wrong order. In one recent case, there were three stylesheets. After the merge, the second stylesheet had been merged first and this did rather interesting things to the cascade.
As for converting inline styles to CSS? That is a trivial task and I seldom see inline styles. The only time I worry about merging near identical CSS classes is on those rare occasions when I am converting an omnibus which has separate stylesheets for each book. As for simplifying complex CSS, calibre does a decent job of trying to do that. Again, there are too many failures (IMHO) that require manual correction.
I won't even get into what calibre did to accessibility code though I haven't tested that in a while. Perhaps it has improved.
For me, it is not a rare occasion when calibre does nasties (IMHO) to the structure of an ePub when converting from azw3/KF8 to ePub, it's most ebooks that requires manual touchup to correct issues such as the simplified CSS not centring headers/subheaders that were centred through using multiple wrapped divs. I tend to suspect many of the issues that annoy me would probably not be annoying to most people if they even noticed them.