IBUS table for Deseret

deseret.txt

SCIM_Generic_Table_Phrase_Library_TEXT
VERSION_1_0

BEGIN_DEFINITION

# Library of Congress romanisation of Deseret (followed except for ï, which it does not define):
#   >  https://www.loc.gov/catdir/cpso/romanization/deseret.pdf
#
# Macron accents (overlines) are represented here by a single x prefix, other accents (umlaut and
# circumflex) are represented by a doubled x prefix.
#
# Very roughly, the accents themselves correspond to:
#
# - Macron: lengthening subject to applicable vowel shifts, as follows from a magic E (sid->side).
# - Circumflex: shortening, per the cot-caught distinction.
# - Umlaut: various other variations.
#
# As far as this IME is concerned, the difference between circumflex and umlaut is inconsequential,
# due to both being represented as "xx" before the vowel.

NAME = Deseret
AUTHOR = HarJIT <harjit@harjit.moe>
NAME.en_Dsrt = 𐐔𐐯𐑅𐐨𐑉𐐯𐐻
LANGUAGES = en_Dsrt
SERIAL_NUMBER = 20180417
UUID = 0aa1f540-b1e1-4637-814f-ade9372abe14
LICENSE = LGPL

DESCRIPTION = Deseret script, based on Library of Congress romanisation; use one or two x before instead of accents.
ICON = ibus-table.svg
STATUS_PROMPT = 𐐔(𐐼𐑂)

LAYOUT = default
VALID_INPUT_CHARS = ABCDEFGHIJKLMNOPRSTUVWXYZabcdefghijklmnoprstuvwxyz
MAX_KEY_LENGTH = 3
AUTO_COMMIT = TRUE
AUTO_SELECT = TRUE
DEF_FULL_WIDTH_PUNCT = FALSE
DEF_FULL_WIDTH_LETTER = FALSE
USER_CAN_DEFINE_PHRASE = FALSE
DYNAMIC_ADJUST = FALSE

END_DEFINITION

BEGIN_TABLE
# Kluge to make ibus work as expected with the prefixed x:
X	X	0
x	x	0
xx	xx	0
XX	XX	0
Xx	Xx	0
# ē, name "Long I", magic-E extension of E (= straight extension of i).
xE	𐐀	1
XE	𐐀	1
Xe	𐐀	1
xe	𐐨	1
# ō, name "Long O", magic-E extension of O.
xO	𐐄	1
XO	𐐄	1
Xo	𐐄	1
xo	𐐬	1
A	𐐈	1
a	𐐰	1
# ī, name "Ay", magic-E extension of I (as in, aye-aye).
xI	𐐌	1
XI	𐐌	1
Xi	𐐌	1
xi	𐐴	1
# H, name "H"
H	𐐐	1
h	𐐸	1
# D, name "Dee"
D	𐐔	1
d	𐐼	1
# G, name "Gay" (hard G)
G	𐐘	1
g	𐑀	1
# Dh, name "Thee", as in ð (THis and THat)
Dh	𐐜	1
DH	𐐜	1
dh	𐑄	1
# Zh, name "Zhee", as in ž (reviSIon)
Zh	𐐠	1
ZH	𐐠	1
zh	𐑈	1
# N, name "En"
N	𐐤	1
n	𐑌	1
# ā, name "Long E", magic-E extension of A, alternative extension of E.
# Note: there is apparently no letter for a straight lengthening of E. Presumably the
# Mary-Merry-Marry distinction (where it exists) would have to be represented using
# 𐐣𐐯𐐯𐑉𐐨-𐐣𐐯𐑉𐐨-𐐣𐐰𐑉𐐨. One can deduce it was incomplete in the Deseret dialect at the time of
# introduction.
xA	𐐁	1
XA	𐐁	1
Xa	𐐁	1
xa	𐐩	1
# ū, name "Long Oo", magic-E extension of U (also the "oo" in "spoon").
xU	𐐅	1
XU	𐐅	1
Xu	𐐅	1
xu	𐐭	1
# ô, confusingly named "Short Ah", former side of the cot-caught distinction.
# Note the distinction between 𐐉 (𐐿𐐱𐐻=cot, 𐐺𐐱𐑄𐑉=bother), 𐐃 (𐐿𐐫𐐻=caught) and 𐐂 (𐑁𐐪𐑄𐑉=father) may be
# incomplete in some accents.
xxO	𐐉	1
XXO	𐐉	1
Xxo	𐐉	1
xxo	𐐱	1
# Aw, name "Ow", as in OUt, bOUT, cOW (phonemic aw; English non-phonemic "aw" is o).
Aw	𐐍	1
AW	𐐍	1
aw	𐐵	1
# P, name "Pee"
P	𐐑	1
p	𐐹	1
# Kluge to make ibus work as expected with the prefixed c:
C	C	0
c	c	0
# Ch, name "Chee"
Ch	𐐕	1
CH	𐐕	1
ch	𐐽	1
# F, name "Eff"
F	𐐙	1
f	𐑁	1
# S, name "Ess"
S	𐐝	1
s	𐑅	1
# R, name "Er"
R	𐐡	1
r	𐑉	1
# Ng, name "Eng"
Ng	𐐥	1
NG	𐐥	1
ng	𐑍	1
# ä, name "Long A", straight extension of A.
# Note the distinction between 𐐉 (𐐿𐐱𐐻=cot, 𐐺𐐱𐑄𐑉=bother), 𐐃 (𐐿𐐫𐐻=caught) and 𐐂 (𐑁𐐪𐑄𐑉=father) may be
# incomplete in some accents.
xxA	𐐂	1
XXA	𐐂	1
Xxa	𐐂	1
xxa	𐐪	1
# I, name "Short I", as in pit, sit, bid.
I	𐐆	1
i	𐐮	1
# U, potentically confusing name "Short O", as in such, much, dud, son, won.
U	𐐊	1
u	𐐲	1
# W, name "Wu"
W	𐐎	1
w	𐐶	1
# B, name "Bee" (possibly a deliberate allusion to "Deseret"?)
B	𐐒	1
b	𐐺	1
# J, name "Jee"
J	𐐖	1
j	𐐾	1
# V, name "Vee"
# Note: Noto appears to use a rather odd glyph variant in which the stroke does not join itself
# on the left, i.e. it appears as a mirrored ā. This may be an issue, in that some handwritten
# styles apparently draw the ā backward to start with (Noto doesn't).
V	𐐚	1
v	𐑂	1
# Z, name "Zee"
Z	𐐞	1
z	𐑆	1
# L, name "El"
L	𐐢	1
l	𐑊	1
# Oy, name "Oi", ligature of ô and i. Earlier versions of Deseret used a simple digraph, later
# versions used a ligature in one of two mutually illegible glyph variants (the Unicode chart
# uses the older of the ligatures).
Oy	𐐦	1
OY	𐐦	1
oy	𐑎	1
# O, name "Long Ah", as in å (cAUGHt, pAW).
# Note the distinction between 𐐉 (𐐿𐐱𐐻=cot, 𐐺𐐱𐑄𐑉=bother), 𐐃 (𐐿𐐫𐐻=caught) and 𐐂 (𐑁𐐪𐑄𐑉=father) may be
# incomplete in some accents.
O	𐐃	1
o	𐐫	1
# E, name "Short E", as in bed.
E	𐐇	1
e	𐐯	1
# ü, name "Short Oo", as in put, or the "oo" in cook and look.
xxU	𐐋	1
XXU	𐐋	1
Xxu	𐐋	1
xxu	𐐳	1
# Y, name "Yee"
Y	𐐏	1
y	𐐷	1
# T, name "Tee"
T	𐐓	1
t	𐐻	1
# K, name "Kay"
K	𐐗	1
k	𐐿	1
# Th, name "Eth", as in þ (THink, THankyou).
# (Name not to be confused with the Roman letter eth or edh (ð), which is Dh above.)
Th	𐐛	1
TH	𐐛	1
th	𐑃	1
# Sh, name "Esh".
Sh	𐐟	1
SH	𐐟	1
sh	𐑇	1
# M, name "Em".
M	𐐣	1
m	𐑋	1
# Kluge to make ibus work as expected with the prefixed xy:
xY	xY	0
xy	xy	0
XY	XY	0
Xy	Xy	0
# Yū, name "Ew", ligature of i with ū. Earlier versions of Deseret used a simple iū or yū digraph,
# later versions used a ligature in one of two mutually illegible glyph variants (the Unicode chart
# uses the older of the ligatures).
xYu	𐐧	1
XYu	𐐧	1
xYU	𐐧	1
XYU	𐐧	1
Xyu	𐐧	1
xyu	𐑏	1
# ï, Shelton Schwa, as in centEr. Had limited adoption. Actually now uses Latin Iota. This is the
# only representation here which does not follow Library of Congress romanisation, due to there not
# being one.
xxI	Ɩ	1
XXI	Ɩ	1
Xxi	Ɩ	1
xxi	ɩ	1
END_TABlE