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


Help: OASIS Mailing Lists Help | MarkMail Help

office message

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

Subject: [OASIS Issue Tracker] Commented: (OFFICE-3435) ODF 1.2 CD05 Part 26.11.13 Incredible HYPERLINK Function

    [ http://tools.oasis-open.org/issues/browse/OFFICE-3435?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=23055#action_23055 ] 

Dennis Hamilton commented on OFFICE-3435:

There are three proposals:

1. Ditch it (the one I recorded as the resolution), since Andreas was the only one who commented on the options I offered in the proposal, and which is objected to by Eike.

2 Do nothing, which Andreas and I certainly don't want to see, as far as I can tell.

3. Describe it in some neutral way that allows it to be implemented without completely savaging the way evaluation is specified to work.

I believe the thread we are on is related to case (3).  The function-result value is delivered as the result for use in the evaluator (and if the last step of the evaluation, delivering it as the computed result after any necessary conversion).  What remains is figuring out how the IRI value is delivered to the Host separately at the point the IRI value is determined for the individual HYPERLINK evaluation.  I think that is the open issue around option (3) discussion.  Whatever we do, only simple cases will likely work the same between OO.o, Gnumeric, and Excel.  There is evidently a wide variance among the current implementations of this function.

My view is that whatever works for (3), it works by the OpenFormula evaluator delivering the IRIs that arise in this manner to the HOST for HOST-dependent behavior resolution, separate from the delivery of the evaluator final computation result.

Note that my examples around (3) are also intended to reveal the extent to which current implementations allow the IRI parameter to itself be the result of computation or must they be string literals always?  Andreas also has good cases when the HYPERLINK invocation is not the outermost operation of the evaluation.

> ODF 1.2 CD05 Part 2 6.11.13 Incredible HYPERLINK Function 
> ----------------------------------------------------------
>                 Key: OFFICE-3435
>                 URL: http://tools.oasis-open.org/issues/browse/OFFICE-3435
>             Project: OASIS Open Document Format for Office Applications (OpenDocument) TC
>          Issue Type: Bug
>          Components: Needs Discussion, OpenFormula, Part 2 (Formulas)
>    Affects Versions: ODF 1.2 CD 05
>            Reporter: Dennis Hamilton
>            Assignee: Dennis Hamilton
>             Fix For: ODF 1.2 CD 06
> ODF 1.2 CD05 Part 2 has an inexplicable definition for the HYPERLINK function.
> The description asserts that there is an effect on a cell.
> The description assrts that there is an effect on a cell even if the function's occurrence is not needed in the evaluation of the formula.
> The description asserts that there is a coordinated effect in how an IRI is composed (or multiples delivered) from individual occurrences of the HYPERLINK function in the same expression.
> The description asserts unjustifiable requirements on the implementation of an evaluator.
> Finally, the examples of this function are used to refine the definition.  Only the simplest examples can be inferred from the definition.
> It appears that incidental consequences of a specific implementation are being appealed to, without concern for what is essential and good enough for an OpenFormula case and no matter what odd behaviors that implementations might have for cases that defy well-definedness.

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators: http://tools.oasis-open.org/issues/secure/Administrators.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


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