[css3-flexbox] issues resolved in FTF

Several flexbox issues have been discussed at CSSWG FTF. Most issues apply to both current specification [1] and a recent proposal [2]. Convergence of the two proposal is work in progress.

RESOLVED:

ISSUE-138<http://www.w3.org/Style/CSS/Tracker/issues/138>   absolute positioning of children of flexbox<http://www.w3.org/Style/CSS/Tracker/issues/138>
Absolute positioned elements are wrapped into anonymous blocks as normal inline content, similar to the same case in tables.

ISSUE-139<http://www.w3.org/Style/CSS/Tracker/issues/139>   pack:start in reverse direction<http://www.w3.org/Style/CSS/Tracker/issues/139>
"box-pack:start" refers to logical start. E.g. in horizontal layout, if "box-direction:reverse" is right-to-left, "box-pack:start" aligns content to the right.

ISSUE-140<http://www.w3.org/Style/CSS/Tracker/issues/140>   max-height on flexbox children<http://www.w3.org/Style/CSS/Tracker/issues/140>
On a child of a flexbox , 'max-length' specified in transverse direction ('max-height' in horizontal flexbox or 'max-width' in vertical flexbox) has the effect of limiting the size of the element. It changes the effective size of the element for the purposes of calculating the size of the flexbox; it has no unusual effects beyond that.
If in a flexbox with "box-align:stretch" one or more children have 'max-height' or 'max-width' that make their dimension in transverse direction smaller than other children, such children are aligned as if with "box-align:before"

ISSUE-141<http://www.w3.org/Style/CSS/Tracker/issues/141>   baseline of an inline-box<http://www.w3.org/Style/CSS/Tracker/issues/141>
Baseline of a flexbox is the baseline of its first child (in source order, in the first ordinal group). This doesn't depend on direction, orientation or 'box-lines' property.
NOTE: this simple rule may need to be adjusted if use cases are found where getting baseline from the first child is not the best choice. To be updated as needed.

Within a flexbox, baselines of children are determined in the same way as when applying baseline alignment to table cells (baseline of an element with multiple lines of text is the baseline of its first line).

ISSUE-142<http://www.w3.org/Style/CSS/Tracker/issues/142>   Does box-ordinal-group change drawing order?<http://www.w3.org/Style/CSS/Tracker/issues/142>
Flexbox children are rendered in order of ordinal groups (a.k.a. box tree order). Appendix E may or may not need an update (if it is defined to refer to "box tree" it automatically requires to render in ordinal group order).

ISSUE-143<http://www.w3.org/Style/CSS/Tracker/issues/143>   Does "shrink-to-fit" width of a flexbox child depend on available space?<http://www.w3.org/Style/CSS/Tracker/issues/143>
Shrink-to-fit calculations are performed in infinite space.
In terms of CSS3 box model [3], shrink-to-fit width and height should be using 'max-content' rather than 'fit-content'

ISSUE-144<http://www.w3.org/Style/CSS/Tracker/issues/144>   distribution of negative available space<http://www.w3.org/Style/CSS/Tracker/issues/144>
Not resolved. Agreed that simply applying flex calculation with the negative sign rarely makes sense. A better solution to be investigated. Advice: explore TeX box model.

ISSUE-145<http://www.w3.org/Style/CSS/Tracker/issues/145>   inline-block children of flexbox<http://www.w3.org/Style/CSS/Tracker/issues/145>
Change how non-block children of flexbox are handled
Before: all inline content (including text, inline elements, inline blocks) is wrapped in to anonymous blocks. Then these anonymous blocks become individual flexbox items.
After: only *non-replaced* inline elements are wrapped into anonymous blocks. Replaced elements (such as buttons and images) are always individual flexbox items.

REMAINING ISSUE:  http://www.w3.org/Style/CSS/Tracker/products/28

References:
[1] July 2009 spec http://dev.w3.org/csswg/css3-flexbox/
[2] new proposal http://www.xanthir.com/:wih
[3] CSS3 Basic Box Model:  http://www.w3.org/Style/Group/css3-src/css3-box/Overview.html