Re: Now: Paragraph Direction Detection and Harmonization

[Eli Zaretskii <eliz@gnu.org>]

> From: Mohsen BANAN <list-general@mohsen.1.banan.byname.net>
> Cc: Mohsen BANAN <list-general@mohsen.1.banan.byname.net>,  emacs-bidi@gnu.org,  emacs-devel@gnu.org
> Date: Tue, 26 Apr 2011 11:24:04 -0700
> 
> I don't know if there is a PICS-Proforma for 
> http://unicode.org/reports/tr9/.
> 
> If there was, it would take the form of tables
> refering to the base spec asking for details of
> implementation with regard to each
> section/option/feature/...  of the specification.
> 
> An example of PICS-Proforma in the RFC Series is:
> http://tools.ietf.org/html/rfc1708

I don't think such a thing exists for UAX#9.  There's Section 4
"Bidirectional Conformance"; Emacs implements everything it requires
under "Full bidirectionality" class, as stated in the fragment of NEWS
I cited.  That's all there is to it.

> In the absence of a PICS-Proforma, based on your
> implementation details, you (or somebody) can go
> through the specification and include emacs-bidi
> implementation details and cross-referencing UAX#9
> sections.

The source code in bidi.c already includes such cross-references.

> Given the lack of precision for uniformity in
> determination of say paragraph direction, based on
> the emacs PICS, we can then visit our sisters at
> Mozilla and have a conversation towards more
> harmony.

The algorithm used by Emacs to determine the paragraph base direction
is exactly as prescribed by UAX#9.  What differs, under the
"high-level protocol" fire escape, is the definition of the
paragraph.  This definition is explained in the message I quoted
earlier in this thread and in the Emacs User Manual.  It is quite
simple and easy to understand, so I really see no place for any
misunderstanding.