RE: [oic] charts in spreadsheets, ranges

[Hanssens Bart <Bart.Hanssens@fedict.be>]

Dennis,


indeed, the <draw:object xlink:href=...  is what I was referring to.


Best regards

Bart
________________________________________
From: Dennis E. Hamilton [dennis.hamilton@acm.org]
Sent: Thursday, June 25, 2009 10:16 PM
To: 'Andreas J. Guelzow'; oic@lists.oasis-open.org
Subject: RE: [oic] charts in spreadsheets, ranges

Thanks, I figured that was the case.

1. OK, I assume the discussion of table:cell-range-address and
chart:values-cell-range-address has nothing to do with the URI question,
since those attributes do not carry URI values.

2. So we are talking about the interesting case of an "embedded" chart
document whose content.xml file has an <office:body> element with
<office:chart> content which in turn contains a single <chart:chart> along
with some optional companion material.

3. This chart is being incorporated into the main document by reference
somehow, such as using a <draw:object> element xlink:href attribute with a
relative URI that references the "sub-package" that is the chart document.
So 17.5 applies.

4. There's nothing in 17.5 or the URI specification that deals with
sub-packages, only subfiles.  The first clue about sub-packages is in
section 17.7.3 (although there is no definition of what the full-path value
would be when there is no corresponding Zip item).  The second clue is in
the sample manifest following section 17.7.6, with the following remarkable
element:

   <manifest:file-entry
      manifest:media-type="application/vnd.oasis.opendocument.text"
      manifest:full-path="/"
      />

5. By anology, based on the only thin evidence in the specification,

   <manifest:file-entry
      manifest:media-type="application/vnd.oasis.opendocument.chart"
      manifest:full-path="Objects/Chart 1/"
      />

   and the like appear to follow.


 - Dennis

PS: I agree with Bart that there is no rationale for "./Object 1"
whatsoever.  It should be "Object 1/" and "./Object 1/" (resolving to the
same full-path), once the the specification is tightened to allow URI
reference to the entirety of an embedded set of package items for a
sub-document.

-----Original Message-----
From: Andreas J. Guelzow [mailto:aguelzow@pyrshep.ca]
http://lists.oasis-open.org/archives/oic/200906/msg00057.html
Sent: Thursday, June 25, 2009 11:04
To: oic@lists.oasis-open.org
Subject: RE: [oic] charts in spreadsheets, ranges

On Thu, 2009-06-25 at 10:59 -0700, Dennis E. Hamilton wrote:
http://lists.oasis-open.org/archives/oic/200906/msg00056.html
> I don't believe "Object 1/" can be referenced unless there is a
> manifest:full-path with that value in a <manifest:file-entry> of the
> manifest.xml file.  In that case, the manifest:media-type will provide the
> MIME type of the embedded package content.
>
> You need to look at 17.5 of ODF 1.1 (the counterpart in ODF 1.2 Part 2 has
> not changed yet).
>
> If you are talking about an embedded object (a single binary blob), then
its
> manifest:full-path item name should not end in "/".

Bart is not talking about an "embedded object (a single binary blob)"
but a chart described in the chart document "Object 1/". So this looks
like a subfolder with content.xml etc.

Andreas

[ ... ]
>
> -----Original Message-----
> From: Hanssens Bart [mailto:Bart.Hanssens@fedict.be]
> http://lists.oasis-open.org/archives/oic/200906/msg00055.html
> Sent: Thursday, June 25, 2009 04:15
> To: oic@lists.oasis-open.org
> Subject: [oic] charts in spreadsheets, ranges
>
> [ ... ]
>
> So, for maximum interoperability, I guess a consumer must be able to
>
> - handle both href "./Object 1" and "Object 1/" hrefs
> - draw the chart even when the cell range is missing on the
chart:plot-area
> or the chart:series
>
>
> And a producer should probably set the cell range on both elements.
>
> IIRC the correct way for the href is "Object 1/" (why is there a space by
> the way ? It is a valid IRI, but still strange), although this will break
> older implementations that rely on "./Object 1". Not sure if there's a
> work-around that's both correct and backwards-compatible...
>
>
>
> Best regards,
>
> Bart
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>
--
Andreas J. Guelzow <aguelzow@pyrshep.ca>


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php