Regarding the question of styles lost between ODF 1.2 and ODF 1.3, I have analysed the .odt files for all WDs from WD08 through to CSD02. The style that was lost (from WD13-03 onwards), and therefore affects the extraction of default attribute values, is a paragraph style: âDefault Valueâ. No character style was lost, since the character style âAttribute Valueâ is still applied where it should be. See, for example, Part 3, Â19.2 in CSD02.
From: firstname.lastname@example.org <email@example.com> On Behalf Of Svante Schubert
Sent: 28 May 2020 00:32
To: Paul Knight <firstname.lastname@example.org>
Subject: Re: [office] Re: URL to the XSLT being used to transform ODT 1.3 to HTML specification
Thanks, for the information that will help. I am going to compare the XSLT stylesheets.
If I am able to improve the XSLT and the resulting HTML outcome, would it be possible for the TC to exchange the existing none-normative HTML ODF 1.3 specification files from:
How is the situation if we want to add some characters styles to the normative ODF 1.3 ODT specification?
They existed in ODF 1.2 & were lost by regression.
They are used for automated extraction of ODF attribute default values from the ODF 1.3 ODT specification.
Does OASIS have a flow chart on evolving a specification to a standard?
These are likely editorial issue and as far as I know, only requires 14 days of review, right?
I used LibreOffice to export to html for the CS01 publication.
This is the version I used:
Version: 18.104.22.168 (x64)
Build ID: a64200df03143b798afd1ec74a12ab50359878ed
did you create the HTML ODF files as Francis assumed?
I am looking for the XSLT being used, to compare it with my old ones to fix some regressions.
Furthermore, the TC decided to enable a TC GitHub to have a better programmatic access to our files and enable automated regression tests.
What would be the next steps?
Sorry you havenât had a response to your posting of 14 May. I presume that youâre referring to CS01. I had no hand in preparing this HTML, so my guess is certainly no better than yours, but looking at the source of the HTML (itâs XHTML) Iâd say that it was all definitely converted using the XSLT built into LibreOffice, since I see the following comment at the top of each Part:
<!--This file was converted to xhtml by LibreOffice - see https://cgit.freedesktop.org/libreoffice/core/tree/filter/source/xslt for the code.-->
My guess is that Paul Knight at OASIS generated the XHTML in LibreOffice rather than by running the XSLT separately, but Patrick may know differently.
after 13 days I suspect the XSLT was taken directly from the LibreOffice that was installed on the editor's machine.
I am asking as I would like to compare that XSLT version used to create the ODF 1.3 HTML with the versions I have in my personal archive and I had been used for ODF 1.2.
Otherwise, I will compare my version with the latest version of LibreOffice.
The reason for the comparison are several regressions, that I have observed:
- Missing paragraph border
- Missing space in between heading text and heading number
- Untypical margin/padding of paragraphs
- Finally, I would like to write a script to remove the random ID of the content table (first below) and only use the methodical ID that is being inserted (second below,
for instance, the following two URLs point to the same <table:table-cell> declaration:
Dear ODF editors,
Could you please point me to the location of the latest / currently used XSL transformation stylesheets of the ODT 1.3 to HTML specification?
OASIS - Advancing open standards for the information society