|
[
Permalink
| « Hide
]
Patrick Durusau added a comment - 06/Jul/09 01:00 PM
Proposal filed.
The original text is peculiar. It is the row being collapsed that has it be invisible, according to preceding text.
The observation should be about how rows are made collapsed, not about how collapsed rows become invisible. A better solution here would be to strike the statement "Collapsed rows have been made invisible by ... the UI directly," since it has no normative or descriptive significance here. In contrary to Dennis proposal I keep the sentence as I see descriptive significance. It explains that collapsed rows are always invisible due to direct user interaction.
In other words: When bringing the two kind of row invisibility side by side, it spoils the effect when taking one away. Svante, I agree with what you want to make clear. But neither the original text nor the change accomplish that.
I suggest that the entire sub-topic be replaced by the following: """ The table:visibility attribute specifies the conditions under which the table row is visible. The values of the attribute are: * visible: the row is visible * collapse: the row is not visible * filter: the row is visible if it satisfies the table filter condition (section 8.7) The default is "visible". """ The wording for the "filter" case may need to be adjusted depending on what is supposed to happen if there is no filter condition. That is, it is only visible if a filter selects it or is it only invisible if there is a filter and it fails to select the row.
Or is it stranger than that and the filter case is implementation-dependent? Dennis and I agreed on a new resolution.
The proposed resolution appears to be a significant change from teh desciption in 1.0:
In 1.0 it is clear that a row with table:visibility-filter is invisible. In the proposal such a row may or may not be visible. Note that this proposal also seems t differ from the way it is implemented in some implementations: In an OOo 3.2 generated document with a filter only those rows covered by the filter that are not visible have table:visibility="filter", the visible rows in the filter are missing that attribute. This is especially of concern for consumers that want to display the file only since the proposal require that it can evaluate the table condition while the plain reading of the original 1.0 does not require such evaluation. In CD5 of ODF 1.2 the description of table:visibility=filter is:
filter: a row or column is not visible as the result of applying a filter. This also differs from this proposed resolution. Perhaps a better resolution would be to say: "filter: the row is not visible as it does not satisfy the table filter condition (section 8.7) " Andreas,
That's a great catch. I would have never figured out that this is an ex post facto indication that the row is hidden because the filter condition is already not satisfied (or was not satisfied when the document was produced). I agree. We need to discuss this to make sure the ODF 1.2 condition is as it was for ODF 1.0/1.1 and that this is the way to describe it. Thanks. You are absolutely right, Andreas.
No need of discussion here, it was earlier defined as you stated it and as well in ODF 1.2 defined. Aligned wording to ODF 1.2, to minimize merge conflicts when errata might be applied. "filter: a row is not visible as the result of applying a filter (see section 8.7)" Fixed the resolution (& errata) and removed the discussion tag from topic. Thanks for pointing this out, Svante This resolution also resolves the defect-report item assigned
The consolidted resolution is as provided in CD04-rev07. Errata 02 - ODF 1.2 Reconciliation. There are differences in wording in ODF 1.2 although the sense is the same. One cleanup is also valuable for ODF 1.2 and is reported as a task.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
Bug
Applied
Major
ISO/IEC JTC 1/SC 34 N 1078 : DEFECT REPORT NUMBER JP2-13
