Hacker News Reader

Comment by zepolen | Hacker News Reader

With extensive experience with both print and web all I can say is:

HTML and PDF were designed for two entirely different things, PDF being much more strict in this respect (as it can make assumptions about display medium that HTML cannot).

Thinking you can go from HTML to PDF with correct results is a very wrong assumption.

View on HN

Agreed for arbitrary HTML, but there a lot of webapps (including one I've written) where it's useful to be able to produce PDF output. When you control the HTML and can test that it converts to PDF the way you want, this can be much faster than trying to generate PDFs from raw Postscript or whatever.

Exactly. This has countless applications. Vastly simpler to create an HTML report, and convert to PDF, than to program said report in some sort of PDF library. You can use your framework's template library, and leverage existing skills. Your designers can design reports as easily as pages.

On my Mac, the kerning was wacked on Arial. Probably something that can be done about that. But all in all - highly impressive.

I found that the font rendering with this was a little weird in general. Arial was pretty horrible, other fonts were a bit better but letter spacing was still always a bit off (this was on a PC btw, so it's not just Mac). I was never able to figure out what was going on, so we ended up going with a for-pay html-to-pdf library because it didn't have the same font rendering issues.

Alos, the point of working with HTML is to let the app users to modify directly their report in the HTML and then still be able to download the PDF.

Think about an invoicing tool. You could have the invoice in HTML with some kind of rela pdf look.. users could for example delete rows, change amounts, ... and still be able to download the invoice in PDF just after.

Reply on news.ycombinator.com