On Thu, Jun 28, 2012 at 7:18 AM, Kang-Hao (Kenny) Lu <kennyluck@csail.mit.edu> wrote: > I am having trouble understanding the algorithm here, and I'll > appreciate some help, for example, a pointer to a previous discussion. We're having trouble understanding what we were trying to say too, so you're not alone. > 9.2. Line Length Determination > > # 3. Determine the flex base size and hypothetical main size of each > # item: > # > # ... > # Otherwise, if the flex container's main-axis is parallel to the > # item's inline-axis, lay out the item using the available space and > # its flex basis if the item is inflexible, > > Question 1: What does use the available space *and* its flex basis mean? > Using its flex basis as the available space? Using the maximum of > available space from the flex container and the flex basis? Yeah, we need to clarify this. > Question 2: What is the flex base size of > > <div style="display: flex; width: fit-content"><!--floated container--> > <div style="flex: 0 0 50%;"> > </div> > </div> > > The flex item is inflexible and the percentage is resolved against an > flex container with indefinite width and hence indefinite by definition. > How do you resolve 50% at this point? That's undefined per 2.1. We need to fix this for *all* display types. > # or ‘auto’ otherwise, treating ‘auto’ as ‘max-content’... > # If the flex basis is ‘fill-available’, or ‘fit-content’, and > # the flex container is being sized under a min-content or > # max-content main-size constraint, size the item under that > # constraint instead. The flex base size is the item's resulting > # measure. > > Question 3: What does "sized under a min-content or max-content > main-size constraint" mean? What are the examples besides "width: > min-content" and "width: max-content" on the flex container? Those terms are defined in a commented out section of the algorithm. We're not sure where to put the definitions; they're applicable to a lot more than just Flexbox. We definitely need to do something about this, though. ~TJReceived on Friday, 29 June 2012 00:25:45 GMT
This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:55 GMT