| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Jan Holesovsky <kendy@suse.cz>
|
|
Checking whether an empty TableCell Element is a header row is now in 2 cases:
* | e | h | h | h | h | ... => The empty cell should be a header
This case occurs when this row is the first row in the table.
* | h | p | p | e | p | ... => The empty cell should not be a header
This case occurs when this row is not the first row in the table,
but the first column is the row header.
where | is a column seperator, e is an empty cell, h is a header cell
(which has role="tablehead" attribute), and p is not a header cell.
Note that parsing occurs left-to-right, so isTableHeader depends on
the last TableCell Element in that row. I assume that if the last element
is a header, that row should be a header row.
Currently this code gives correct behaviour, but checking whether a row
is the first row might be more correct.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The paragraph ids are duplicate among files :-(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Still it is not perfect for duplicates of 'branch' attribute in bookmarks, but
what can we do...
|
|
|
|
It is usually (== everywhere but 2 or 3 files) the only level 1 heading in the
file, and additionaly exactly the same as the page title. When we skip that,
we get much nicer table of content, better structure, etc.
|
|
|
|
|
|
|
|
|
|
|
|
This is more dynamic, and can be controlled by the help authors, instead of
relying on new builds when something help-related changes.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
