Closed Bug 35393 Opened 25 years ago Closed 25 years ago

feature: Need span for XUL grids

Categories

(Core :: XUL, enhancement, P1)

enhancement

Tracking

()

RESOLVED WONTFIX

People

(Reporter: german, Assigned: eric)

Details

(Whiteboard: 5 days 5/5/00)

We do need span (like in tables) for XUL grids ebfore we can re-create most of the common dialogs, properties dialogs and prefs panels among others. This is a pre-requisite for skinnability for NS 6.
setting apprpriate keywords and owners
Assignee: trudelle → evaughan
Keywords: beta2, skins
Summary: Need span for XUL grids → [skins][beta2]Need span for XUL grids
Blocks: 29160
setting P1 for M16
Priority: P3 → P1
Summary: [skins][beta2]Need span for XUL grids → Need span for XUL grids
Target Milestone: --- → M16
targeting
Status: NEW → ASSIGNED
Whiteboard: 3 days
setting completion date
Whiteboard: 3 days → 3 days 4/19/00
Whiteboard: 3 days 4/19/00 → 3 days 4/24/00
Keywords: nsbeta2
No longer blocks: 29160
adding feature tag
Keywords: beta2
Summary: Need span for XUL grids → feature: Need span for XUL grids
Reseting completetion date.
Whiteboard: 3 days 4/24/00 → 3 days 4/27/00
Whiteboard: 3 days 4/27/00 → 3 days 4/29/00
updating status whiteboard with latest estimates
Whiteboard: 3 days 4/29/00 → 5 days 5/5/00
Consider using nested <grid>s instead. We've used that successfully in Composer's Spell Check dialog.
Span turn out to be very math intensive. To do them right is not easy. I can probably get the 90% case but it will take time. Unfortunately more than we have for m16. However it is possible to get most of what you want just by nesting them. I suggest we try and work around them. And schedule them to be finished for a future milestone. This will allow me to work on more important bugs. Such as splitters not collapsing. And stack z order problems. -E
Status: ASSIGNED → RESOLVED
Closed: 25 years ago
Resolution: --- → WONTFIX
Can you point me to some details about the math intensive stuff? Sounds interesting.
I have to sit and write it down. Most of the work I did was just with some engineers here. It seems simple at first but once you truly understand the whole problem. It turns out be not quite so simple. The really issue it stating the problem correcly. And for that you have to understand boxes intimately. So the plan to to write a white paper on boxes and then state the problem and let people have crack at it. But essentually the problem as to do would overlapping constaints in grids that cause couped equations. Here is an example: 3 columns. Columns 1 has a flex of 1, 2 a flex of 2, and 3 a flex of 3. Now you have now you have 3 rows. In the first row you have 3 buttons one in each column. And each button has a max size of 20px. Then you have a second row. It has 2 cells. On that spans cells 1 and 2 and a second that in in column 3. the first one has a max size of 100px. Then you have a third row with 3 cells in it. Cell one in in column1 but cell 2 spans columns 2 and 3. It has a max size of say 100px. Now you need to detemine the min, max, and preferred sizes for each column. Now assuming that all these cells can have all three types of contraints such as min, max, and pref. As well as different flexes. This is easy to solve if there are no overlapping constrains. If there are like above the solution is not so simple.
Verify? Doesn't look someone's still going about this....
Keywords: skins
You need to log in before you can comment on or make changes to this bug.