#111 Incorrect rendering of Vedic Sanskrit accents

[David M. Jones]

On a lark, I've been experimenting with typesetting Vedic Sanskrit,
specifically verses from the Ṛg-Veda, using the Murty Hindi font from
Tiro Typeworks. So far, XeTeX works beautifully, with two exceptions
involving the Vedic udatta and anudatta accent marks:

a) They don't attach properly to the Devanagari digits 1 and 3, and

b) They don't combine properly with the candrabindu, anusvara, and
visarga signs.

luaTeX does render these correctly, so it doesn't appear to be a
problem with the font itself.

I've included a zip file with a test file and sample output from
luaTeX and XeTeX.

Is this a bug, or am I missing something obvious?

[Arthur Reutenauer]

Hi David,

Thanks for the very clear and detailed bug report. May I suggest, since you already have the ZIP file handy, that you mention the bug to the XeTeX mailing list? We’re only a handful of people monitoring this bug tracker; on the list you may find people who would be able to analyse the problem in more detail.

The list’s address is xetex@tug.org and you’ll find more information here. You may link to the ZIP file on this ticket (https://sourceforge.net/p/xetex/bugs/111/attachment/xetex-indic-bug.zip) so as to avoid sending it to the list directly. Then again, sending it is fine if you feel like it, it’s pretty small.

P-S: I came across the first books by Murty Classical Library in my bookshop recently and really liked the typography; glad to know they’re making the fonts freely available too.

[Anonymous]

Hi, Arthur,

Thanks for your email. I've been debating whether to crosspost the
bug to the XeTeX mailing list, and this is just the encouragement I
need.

The Murty volumes are handsome, aren't they?

Best wishes,
David.

From: Arthur Reutenauer reutenauer@users.sf.net
Reply-To: "[xetex:bugs] " 111@bugs.xetex.p.re.sf.net
Date: Fri, 22 May 2015 00:06:28 +0000

Hi David,

Thanks for the very clear and detailed bug report. May I suggest, since you already have the ZIP file handy, that you mention the bug to the XeTeX mailing list? We’re only a handful of people monitoring this bug tracker; on the list you may find people who would be able to analyse the problem in more detail.

The list’s address is xetex@tug.org and you’ll find more information here. You may link to the ZIP file on this ticket (https://sourceforge.net/p/xetex/bugs/111/attachment/xetex-indic-bug.zip) so as to avoid sending it to the list directly. Then again, sending it is fine if you feel like it, it’s pretty small.

P-S: I came across the first books by Murty Classical Library in my bookshop recently and really liked the typography; glad to know they’re making the fonts freely available too.

---

** [bugs:#111] Incorrect rendering of Vedic Sanskrit accents**

Status: open
Group: Future
Created: Thu May 07, 2015 12:51 AM UTC by David M. Jones
Last Updated: Thu May 07, 2015 12:51 AM UTC
Owner: nobody

On a lark, I've been experimenting with typesetting Vedic Sanskrit,
specifically verses from the Ṛg-Veda, using the Murty Hindi font from
Tiro Typeworks. So far, XeTeX works beautifully, with two exceptions
involving the Vedic udatta and anudatta accent marks:

a) They don't attach properly to the Devanagari digits 1 and 3, and

b) They don't combine properly with the candrabindu, anusvara, and
visarga signs.

luaTeX does render these correctly, so it doesn't appear to be a
problem with the font itself.

I've included a zip file with a test file and sample output from
luaTeX and XeTeX.

Is this a bug, or am I missing something obvious?

---

Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/xetex/bugs/111/

To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/

[Anonymous]

ShreeDevi Kumar on the xetex list pointed out that I didn't understand
the relevant portion of the Unicode standard, rule R10 on page 444 of
the Unicode 7.0 specification:

http://tug.org/mailman/htdig/xetex/2015-May/026017.html

What I referred to as a "semantically invalid reordering of the
Unicode characters that happens to produce acceptable output" is, in
fact, the prescribed Unicode canonical order.

An argument could be made that the order with the svara before the
bindu should be supported as a quality-of-implementation issue, but
it's irrelevant for my use, so I'm not going to press the issue.

I believe these are still glitches:

1) The combinations of visarga + udatta and visarga + anudatta are
still badly positioned. The addition of the visarga shouldn't
change the placement of the udatta or anudatta with respect to the
base character.

2) The combinations of the udatta and anudatta with the Devanagari
digits 1 and 3 still look off to me: the udatta could be better
centered and the anudatta could be lowered a bit to separate it
from the digit.

However, these are consistent with other rendering engines we've
tried, so they would appear to be font issues. I'll take them up with
the font studio.

[Anonymous]

Hi, Arthur,

Thanks for pointing me in the right direction.

Yes, please do close the ticket.

Best wishes,
David.

[Arthur Reutenauer]

Hi David,

Many thanks for forwarding the issue to the XeTeX list, and for this comment on the ticket. If it’s satisfactory to you I’ll now close it, keeping in mind your recommendations for an optimal implementation (unlikely to happen soon, though, I’m afraid).

Best,

Arthur