On 08/07/2003 13:01, Peter Kirk wrote:Thank you, Ted.
[regarding Haralambous] ... can you remind us of the
reference and if possible the URL?
"Typesetting the Holy Bible in Hebrew, with TEX" Yannis Haralambous EuroTEX Proceedings 1994 TUGboat 15(3):174-191, September, 1994
I've found it on-line at: http://tex.loria.fr/fontes/biblical-hebrew.ps.gz http://genepi.louis-jean.com/omega/biblical-hebrew94.pdf (secondary quality image)
From this paper I have identified some further cases of multiple vowels in the printed BHS text. These do not appear in the WTS encoded text because they are cases of Ketiv and Qere where WTS has chosen to encode what is intended, not what is printed. This list cannot claim to be exhaustive because it is based in Haralambous' non-exhaustive list in pp.10-11 of his article (as at http://genepi.louis-jean.com/omega/biblical-hebrew94.pdf); in fact quite probably there are many more such cases, wherever the Qere is longer than the Ketiv. (Note that I have omitted the cantillation marks as I have no convenient way to type them; the forms in BHS are mostly the same as those in Haramlambous' list except that the *'s are omitted and the spaces between the consonants closed up.)
2 Samuel 22:8 (patah - hiriq) וִַתְגָּעַשׁ 1 Kings 9:18 (patah - sheva) וְאֶת־תְַּמֹר 2 Kings 5:25 (patah - hiriq) מֵאִַן 2 Kings 9:15 (sheva - patah) לְַגִּיד Jeremiah 18:23 (sheva - hiriq) וְִהְיוּ Ezekiel 25:9 (qamets - sheva) וְקִרְיָתְָמָה Ezekiel 46:19 (patah - hiriq) בַּיַּרְכָתִַם Daniel 2:9 (sheva - sheva) הִזְְמִנְתּוּן
Note that the sequences patah - sheva and sheva - patah both occur with a distinction of order, although if simply encoded as Unicode sequences they are canonically equivalent. This implies that the approach of defining non-standard rendering orders for such glyph combinations is not going to be adequate to deal with these unusual cases.
Many of the other cases in this list of Haralambous' are printed in BHS with a vowel at the start of the word, to the right of any base character. This is of course another challenge to the normal Unicode encoding rules.
And then there are also Haralambous' "Missing Words" (p.11), which are generally printed in BHS as blank space surrounded by vowel points.
What all of these unusual cases, as well as the case of Yerushala(y)im, have in common, it seems to me, is that a consonant (sometimes more than one) has been omitted from the word (or in some printed editions sometimes replaced by an asterisk or zero) but the vowel which goes with that consonant has been printed. This accords with what Jony Rosenne has been saying; logically there is a consonant there, although it is not visible. And it encourages me to prefer the kind of solution which Peter Constable has just outlined of inserting a character ELLIPTIC LETTER between the two adjacent vowels, or before the apparently word initial vowels. But there may be a need for two different elliptic letters, a zero width one for cases like Yerushala(y)im and a wider one for use in the "Missing Words". Or possibly the choice between these two could be left to the renderer.
-- Peter Kirk [EMAIL PROTECTED] http://web.onetel.net.uk/~peterkirk/
