Have this been discussed? Which direction are you going to rotate?

We discussed this when I was working on the PDFium API.  I added a TODO in my code checked in to rotate pages in the future.  Hence I created this bug to keep track of the work needed.

My plan as shown in the CL sent for review is to rotate 90 degree counterclockwise, and that's what Safari does as well.  However, Acrobat reader Mac OS version doesn't rotate pages at all when I tried to print a portrait landscape mixed orientation PDF.

Discussed with Weifang, and we would like to rotate pages to maximize paper usage for N-up.

The PDF document's orientation is determined by the majority pages' orientation.  The rest of pages will be rotated if there are different from the PDF doc's orientation.

How do we know that's what the user wants? Especially given the size of Chrome's user base and their diverse needs?

In the past, we have made product decisions where we thought we know what's best for the user. e.g. with duplex printing, we went for simplicity and only implemented long-side duplex. Then users file bugs and demand short-side duplex. ( bug 169120 , bug 842063)

Both Acrobat Reader and Mac Preview.app let the user choose. How about we only maximize the page usage when the user selects "Fit to page"?

When the source is PDF, chrome doesn't provide an option to change orientation.  Do we want to add that in the future?  If the answer is yes, we may defer this work.  If the answer is No, it's OK that we only maximize the page usage when the user selects "Fit to page".

I personally like the way how Acrobat Reader works, having a check box for auto-rotate pages within each sheet.  It also has a "Fit" checkbox for fit-to-page.