andthus > and thus in Transliteration section

This commit is contained in:
Nathaniel V. KELSO
2015-12-15 22:45:07 -08:00
parent 3e44910664
commit 7ff7027cdb

View File

@@ -133,7 +133,7 @@ Result:
{ {
"house": "le polikarpov" "house": "le polikarpov"
"house_number": "24" "house_number": "24"
"road": "cours honor d'estienne d'orves" "road": "cours honoré d'estienne d'orves"
"postcode": "13001" "postcode": "13001"
"city": "marseille" "city": "marseille"
"country": "france", "country": "france",
@@ -316,7 +316,7 @@ strips accent marks e.g. à => a and/or applies Latin-ASCII transliteration.
[CLDR transforms](http://www.unicode.org/repos/cldr/trunk/common/transforms/), the exact same as used by ICU, [CLDR transforms](http://www.unicode.org/repos/cldr/trunk/common/transforms/), the exact same as used by ICU,
though libpostal doesn't require pulling in all of ICU (might conflict though libpostal doesn't require pulling in all of ICU (might conflict
with your system's version). Note: some languages, particularly Hebrew, Arabic with your system's version). Note: some languages, particularly Hebrew, Arabic
and Thai may not include vowels andthus will not often match a transliteration and Thai may not include vowels and thus will not often match a transliteration
done by a human. It may be possible to implement statistical transliterators done by a human. It may be possible to implement statistical transliterators
for some of these languages. for some of these languages.