OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

oic message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [oic] Table header


I think the problem is the wording.  It is a fact that <table:table-rows> and <table:table-columns> cannot contain header elements, neither directly nor indirectly (via groups, since those can't occur either).

That's simple.

Here's a way to clarify how row/column headers work:

With regard to how headers work, the basic rule is that those rows/columns of a table that are introduced as row/column headers, respectively, must be in contiguous rows/columns, respectively.  once a row/column header has been presented, it will be repeated at the corresponding heading edge if the table is continued vertically onto another page or is continued horizontally onto another page.  In a spreadsheet, both forms of continuation are commonplace.  

What's missing in this rule is some behavior about groups and what one might want it to be in the future.  (1) I would make the following implementation defined: What happens with a group-introduced row/column header when the group has no non-header rows/columns on the beginning of a new page beyond where the groups row/column headers were established.  (2) I would recommend that group-introduced row/column headers that are not contiguous with the previously-introduced row/column headers *should* be ignored as headers and be treated as ordinary rows/columns by default.  (Some implementations do very strange things already, so I wouldn't say *shall*.)

-----Original Message-----
From: oic@lists.oasis-open.org [mailto:oic@lists.oasis-open.org] On Behalf Of Andreas J. Guelzow
Sent: Tuesday, November 08, 2011 13:32
To: oic@lists.oasis-open.org
Subject: Re: [oic] Table header

On Tue, 2011-11-08 at 13:55 -0700, Svante Schubert wrote:
> table:table-columns seems to be a redundant element as none repetition
> should be the default.

Yes. I am pretty sure there are other redundant elements and attributes
in ODF. I do not see a problem with that. 

> 
> Why is the header not mandatory first in the table?

Imagine you start a table with several "header" rows but on following
pages you only want to repeat say the last two of those "header" rows
since they would suffice for recognition of the columns.

>         Is anyone aware of an table application (not necessary an ODF
>         application) that is able to have header in arbitrary places?
>         
Excel used to be able to do that (it has been 2 decades since I have
used it but I suspect it is still possible.) Gnumeric can have a single
set of header rows in arbitrary places.
>         
>         And if headers can be on arbitrary positions, what would be
>         the position of the header on the next page?

On the pages following the natural occurrence of the header rows they
appear at the top of the page.

>         The first, similar position as on the first page or last
>         position?

For example, say header rows are rows 50 & 51.

page 1 shows rows 1 to 30, the header rows do not appear on this page
yet.

page 2 shows rows 31 to 60, the header rows appear in their natural
position, ie. between rows 49 and 52.

page 3 shows rows 50,51,61 to 88, with the header rows appearing at the
top of the page

Second example, say header rows are rows 60 & 61:

page 1 shows rows 1 to 30, the header rows do not appear on this page
yet.

page 2 shows rows 31 to 60, the header row 60 appears in its natural
position, ie. after row 59.

page 3 shows rows 60,61 to 88, with the header row 60 appearing at the
top of the page and header row 61 in its natural position, ie before row
62 (and after the repeated header row 60).

page 4 shows rows 60,61,89 to 116, with the header rows appearing at the
top of the page.

Within a spreadsheet program this for example allows the statistical
analysis to precede the data on the same sheet and in a print out the
data immediately to follow the end of the of the analysis.

Andreas

 
-- 
Andreas J. Guelzow, PhD, FTICA
Concordia University College of Alberta


---------------------------------------------------------------------
To unsubscribe, e-mail: oic-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: oic-help@lists.oasis-open.org



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]