diff options
Diffstat (limited to 'intl/icu/source/data/translit/es_fa.txt')
| -rw-r--r-- | intl/icu/source/data/translit/es_fa.txt | 20 |
1 files changed, 20 insertions, 0 deletions
diff --git a/intl/icu/source/data/translit/es_fa.txt b/intl/icu/source/data/translit/es_fa.txt new file mode 100644 index 0000000000..9df0448102 --- /dev/null +++ b/intl/icu/source/data/translit/es_fa.txt @@ -0,0 +1,20 @@ +# © 2016 and later: Unicode, Inc. and others. +# License & terms of use: http://www.unicode.org/copyright.html +# Generated using tools/cldr/cldr-to-icu/build-icu-data.xml +# +# File: es_fa.txt +# Generated from CLDR +# + +$Boundary = [^[:L:][:M:][:N:]]; +$Vowel = [i e o u a]; +::es-es_FONIPA; +# In Farsi transcription of Spanish, un-stressed [e] should be treated +# like [ə] which gets stripped off. However, we currently do have not +# have a good way of finding stress in Spanish words. In the long term, +# it would be _much_ better to look at stress markers, but for now +# we do this trivial heuristics to find unstressed [e] in the first +# syllable. +$Boundary [^Vowel] {e} [^$Vowel]* $Vowel → ə; +::und_FONIPA-fa; + |