[ekdosis] ekdosis v1.2

Sun Apr 4 18:47:18 CEST 2021

Dear Dominik,

Of course I didn't know about the existence of the Saktumiva system.
Thank you!  I've found the overall rendition absolutely admirable, to
to mention the many options that are available: IAST, Devanāgarī,
etc., plus those related to punctuation and orthographic variants.
What an amazing work!  As I understand this system, however, one is
supposed to encode the edition text in TEI from the outset, which is
fine of course, but starting from a (La)TeX source file comes with
different technical requirements.  Unfortunately, as far as I am
concerned, I can only address the "Sandhi issue" from a purely
computerized approach, namely: how are we to encode a string in which
there is a letter that must be printed in the edition text, but not in
the apparatus criticus, knowing that: 1. the argument of \lem{} is
printed in the edition text and in the apparatus criticus; 2. only
this same argument is processed as regards the TEI output?  What I'm
getting at is whether you insert this letter in \lem{} or not, you are
in trouble.  Of course, you may say:

   ... kurvan-kriyāyoga\app{
    \lem[wit={X,Y,Z}, alt={tato bhavet}]{s-tato bhavet}
    \rdg{...}}

so that only the contents of the "alt" optional argument is printed in
the apparatus criticus.  But then, what do we get in the TEI output,
knowing that this optional argument is not processed as regards TEI?
Something as:

   ... kurvan-kriyāyoga<app>
                         <lem wit="#X #Y #Z">s-tato bhavet</lem>
                         <rdg>...</rdg>
			 </app>

which is wrong.

After receiving your message, I tried to investigate further this
LaTeX-related issue.  It so happens that ekdosis provides pattern
matching functions that can be of use in such situations.  You will
find attached a PDF with comments.  It starts with a small piece of
text.  Then follow the .tex source file and the corresponding TEI xml
output produced by ekdosis.  The relevant parts in both the .tex and
the xml listings are highlighted in light gray.

What do you think of the problem and my approach?  I wouldn't want to
get completely mistaken!

With regards to sanskrit-hyphenation.tex, Dropbox says: "The item was
deleted".  Is there any other place where I can get it?

Thank you!

Best,

Robert

On Sat, Apr 03, 2021 at 01:23:44PM -0600, Dominik Wujastyk wrote:
> Dear Robert,
> 
> Great news all round.
> 
> I'm pleased to know that the Marburg (and SOAS?) Sanskritists are working
> with Ekdosis.  What you call the "Sandhi problem," by which you mean
> something mainly orthographical not phonetic (sandhi), has to a large
> extent been solved in the Saktumiva <http://saktumiva.org> system where one
> can switch writing systems at will and the system does the right thing with
> inter-word spacing from script-to-script.  Dr Charles Li is the author of
> Saktumiva, and the code is OA and on Github.
> 
> Another Sanskrit issue is hyphenation.  For XeTeX Polyglossia users, as a
> user myself, Sanskrit text is already recognized and if tagged with the
> right Sanskrit language tag Polyglossia (and Babel) will invoke the
> appropriate hyphenation rules for both Devanagari and to IAST Roman
> transliteration as well as several other scripts (Bengali, Gujarati,
> Kannada, Malayalam, and Telugu; thanks to the author Yves Codet, author of
> hyph-sa.tex).  However, these rules are simple and sometimes fail.  Over
> several years I have been building up a list of manually hyphenated
> Sanskrit words and compounds.  Importing my file (\input
> sanskrit-hyphenations.tex
> <https://www.dropbox.com/s/9n3ei6n2zy4vlx9/sanskrit-hyphenations.tex?dl=0>)
> after \begin{document} can help a lot with TeX line-breaking.
> 
> Best,
> Dominik
> 
> 
> On Sat, 3 Apr 2021 at 08:16, Robert Alessi <alessi at robertalessi.net> wrote:
> 
> > Dear members of the ekdosis list,
> >
> > As I am about to announce v1.2 of ekdosis, I realize that I never made
> > any announcement about v1.1.  It must be said that in the last few
> > months I somewhat have lost track of my emails.  I will however try to
> > catch up in the coming weeks.  In the meantime, please accept my
> > apologies for this.
> >
> > ekdosis v1.1
> > ------------
> > So v1.1 was published at the beginning of last November, with the
> > following new features:---
> > - Improved documentation with new sections, notably "Variae
> >   Quaestiones"" and "Laying Out Parallel Texts".
> > - Compliance to LaTeX 2020-10-01 (namely: inclusion of xparse
> >   and new hook management).
> > - Improved \DeclareScholar meant to build a list of persons within the
> >   <listPerson> TEI element.  A new command \DeclareSource for editions
> >   used as sources was added.  As a consequence, 'source' and 'resp'
> >   were added as optional arguments of \lem and \rdg.
> > - Editorial changes: new commands \supplied, \surplus, \sic and \gap
> >   were added among others.
> > - Some commands were renamed: \SetxmlBibResource →
> >   \AddxmlBibResource and \TeXtoTEIPatt → \TeXtoTEIPat.
> > - A new command \addentries was added for modifying the number of
> >   accepted entries on the current page.
> > - ekdosis was modified to accept any page numbering scheme. (v1.0 only
> >   worked on pages numbered with Arabic numerals.)
> > - A new command \rdgGrp for sub-variation was added.
> >
> > ekdosis v1.2
> > ------------
> > This release has just been published and should be available from the
> > mirrors in a couple of days.  The announcement text follows:---
> > - Improved support of babel/polyglossia: it is now possible to apply
> >   to any layer of the apparatus criticus a language different from the
> >   one that is selected in the edition text. This is notably important
> >   eg. for Arabic or Sanskrit where it is customary to use a
> >   transliteration scheme in the apparatus criticus.
> > - New general hooks have been added, allowing for an initial
> >   separating rule, the styling of lemma texts and/or variant readings,
> >   &c.
> > - Poetry: ekdosis can now load and use the facilities provided by the
> >   verse package. Verse lines can be numbered independently of prose
> >   text. Of course, continuous line numbering remains
> >   possible. Indentation patterns are also supported. Furthermore, a
> >   specific environment for the encoding of line groups has been
> >   added. A detailed example from Raymond MacDonald Alden's edition of
> >   Shakespeare's Sonnets from the Quarto of 1609 with variorum readings
> >   is provided in the documentation, where an effort has been made to
> >   keep the typography and the punctuation of the original edition.
> > - Conditional page breaks: the line number and optionally the page
> >   number where a page break should occur can be specified. If the
> >   conditions specified be not met, then the page break is not
> >   triggered.
> > - Much improved export to TEI xml.
> >
> > I am also very happy to announce that ekdosis is being intensively
> > tested in large projects involving Sanskrit, Syriac or Arabic edition
> > texts.  I should be able to post some news about Syriac and Arabic in
> > the coming weeks.
> >
> > As for Sanskrit, of great interest is that an Anglo-German Haṭha-Yoga
> > project is now editing with ekdosis.
> > (
> > https://www.uni-marburg.de/de/fb10/iksl/faecher/indologie/aktuelles/news/yoga-projekt
> > )
> > This work has also led to the development of "babel-iast", a mapping
> > system for Romanized Sanskrit in IAST scheme to Unicode Devanāgarī.
> > (https://github.com/radardenker/babel-iast)  At the time of writing,
> > within the framework of this project, development towards producing
> > html output from TEI xml is underway.  Do not hesitate to ask for more
> > information should you be interested in knowing more about this.
> >
> > What is good is that inter-disciplinary work always provides
> > opportunities to spot tricky issues, such as what I would call "the
> > Sandhi problem" for Sanskrit edition texts, namely rules that dictate
> > to write together in Devanāgarī words according to their final and
> > initial sound.  All of this has to be rendered properly in the edition
> > text, in the apparatus criticus in case some variant gets in the way
> > and of course in TEI and html output.  I hope to find a solution to
> > work around such issues soon.
> >
> > Best wishes,
> >
> > Robert
> >
> > --
> > Robert Alessi
> > CNRS UMR 8167 «Orient & Méditerranée» (Paris)
> > <robert.alessi at cnrs.fr>
> > _______________________________________________
> > ekdosis mailing list
> > ekdosis at robertalessi.net
> > http://www.robertalessi.net/mailman/listinfo/ekdosis
> >

> _______________________________________________
> ekdosis mailing list
> ekdosis at robertalessi.net
> http://www.robertalessi.net/mailman/listinfo/ekdosis

-------------- next part --------------
A non-text attachment was scrubbed...
Name: mwe-mk2.pdf
Type: application/pdf
Size: 42032 bytes
Desc: not available
URL: <http://www.robertalessi.net/pipermail/ekdosis/attachments/20210404/afa7d002/attachment-0001.pdf>