interscript-maps 2.4.1 → 2.4.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 48704a370df7b611d83ab6cf0f95a99cb989b109b80264e3d8c215cc400b84ae
-  data.tar.gz: 82fde9ce9df7256ab66b89a32474a2bc235c0ab0c404a3c7d08fab4cce20949d
+  metadata.gz: a497101e667c29fc9cec9896fe1b0f052187835c566803cbff4541a528155149
+  data.tar.gz: 8b58b2423fb8c6b57815841926e2e98f001a71a67776f8c705e6f98906292552
 SHA512:
-  metadata.gz: c2ec1877d3b7daaabb986f163f688e45f8b2fff2f251ab4d43c472600c8ab2741cfcfac4bdf6d17c7d4ce93667b8b077c8eaac509c0b914f9bda8ae192391ea3
-  data.tar.gz: 5d6efecc1e0cc5c3fd224bcd24ec4f4ad2370b12ad6d38c3c17f7d0cb3dfeb66a2cff797018e0639ff0b0a589d1a43af228520ffc4688d2f2b8cb37cdbe27b1c
+  metadata.gz: bbf2de534e2ef493dc7e8a081d008ebd66808f5b42bc75297dd91b9a9d7b4effd39dfc7b3c138701510949a88086aa400d1015908e1c1019593d797e9197d907
+  data.tar.gz: 87834f177ef2700a7527b40d3f162a78ca0150adfced50ddd5a2545a3ed8fddd5789e0fedff59948672db991a51166d9fbb0f7ad30bc7e845a0803e859a86fef

data/interscript-maps.gemspec

@@ -1,4 +1,4 @@
-INTERSCRIPT_MAPS_VERSION="2.4.1"
+INTERSCRIPT_MAPS_VERSION="2.4.3"
 
 Gem::Specification.new do |spec|
   spec.name          = "interscript-maps"

data/interscript-maps.yaml

@@ -19,14 +19,14 @@ secryst-models:
 # Here we describe Rababa configs for the maps that are required
 rababa-configs:
   "200":
-    model: https://github.com/secryst/rababa-models/releases/download/0.1/diacritization_model_max_len_200.onnx
+    model: https://github.com/secryst/rababa-models/releases/download/0.1/diacritization_model_arabic.onnx
     config:
       session_name: base
       text_encoder: ArabicEncoderWithStartSymbol
       text_cleaner: valid_arabic_cleaners
       max_len: 200
       batch_size: 32
-      
+
 # This instruction denotes that the following maps are incompatible with certain
 # compilers and platforms and shouldn't be tested or compiled in.
 skip:
@@ -167,6 +167,7 @@ aliases:
       alias_to: bgnpcgn-fao-Latn-Latn-1968
     fas_Arab2Latn_ALA_1997:
       description: Persian 1997 ALA-Library of Congress System
+      alias_to: alalc-fas-Arab-Latn-1997
     fas_Arab2Latn_AMMI_1959:
       description: Afghanistan Ministry of Mines and Industries Transliteration System
         for Geographic Names in Afghanistan (Yaghubi)

data/maps/{alalc-per-Arab-Latn-1997.imp → alalc-fas-Arab-Latn-1997.imp}

@@ -1,7 +1,7 @@
 metadata {
   authority_id: alalc
   id: 1997
-  language: iso-639-2:per
+  language: iso-639-2:fas
   source_script: Arab
   destination_script: Latn
   name: Romanization Table -- Persian (1997)

data/maps/bgnpcgn-fas-Arab-Latn-1956.imp

@@ -47,7 +47,7 @@ metadata {
     - Since maddah (آ), which is placed over alif (ا), nearly always occurs in word-initial position, no .)◌َا( as well as for fatḩah alif )آ( confusion results from the use of ā for alif maddah
     - The ligatures لا and لـا represent lām- alif, and should be romanized lā.
 
-  special_rules:
+  implementation_notes:
     # TODO: These are not used
     - Initial definite articles and prepositions should be capitalized and hyphens should not be used to connect parts of names, e.g., Ash Shāriqah and Tall al Laḩm.
     - If any evidence is found for the use of the definite article in a name, the article should be used in the name chosen.

data/maps/elot-ell-Grek-Latn-743-2001-tl.imp

@@ -12,7 +12,7 @@ metadata {
     Reversible transliteration standard, ELOT
 
   notes:
-    - Transliteration standard (reversible): Clause 3.1, Table 1
+    - "Transliteration standard (reversible): Clause 3.1, Table 1"
 }
 
 # This map has been partially converted by the bin/maps_v1_to_v2 script

data/maps/elot-ell-Grek-Latn-743-2001-ts.imp

@@ -12,7 +12,7 @@ metadata {
     Reversible transliteration standard, ELOT
 
   notes:
-    - Transcription standard (reversible): Clause 3.1, Table 2
+    - "Transcription standard (reversible): Clause 3.1, Table 2"
 }
 
 # tests copied from iso-ell-Grek-Latn-843-1997-t1

data/maps/iso-ell-Grek-Latn-843-1997-t1.imp

@@ -15,7 +15,7 @@ metadata {
     or Modern Greek. Replaces ISO/R 843.
 
   notes:
-    - Transliteration of Greek into Latin: Type 1, Clause 3 Table 1
+    - "Transliteration of Greek into Latin: Type 1, Clause 3 Table 1"
     - Equivalent to elot-ell-Grek-Latn-743-2001-ts, the transliteration table of ELOT 743:2001
     - Assuming that ou, au, eu transliterations are only intended for historical diphthongs /u, av, ev/, and that όυ, άυ, έυ are not to be transliterated as ou, au, eu
     - Introduced casing to digamma and lunate sigma. (Casing was late introduction to character sets for those characters)

data/maps/iso-ell-Grek-Latn-843-1997-t2.imp

@@ -15,7 +15,7 @@ metadata {
     or Modern Greek. Replaces ISO/R 843.
 
   notes:
-    - Transliteration of Greek into Latin: Type 2, Clause 3 Table 2
+    - "Transliteration of Greek into Latin: Type 2, Clause 3 Table 2"
     - Introduced casing to digamma, yot, and lunate sigma. (Casing was late introduction to character sets for those characters)
 }
 

data/maps/lshk-yue-Hani-Latn-jyutping-1993.imp

@@ -7,15 +7,15 @@ metadata {
   name: Jyutping Cantonese Romanisation Scheme
   url: https://lshk.org/jyutping
   creation_date: 1993-12
-  description:
-    - The Linguistic Society of Hong Kong Cantonese Romanisation Scheme, or
-      known as Jyutping, was designed and proposed by the Linguistic Society of
-      Hong Kong in 1993. Jyutping is a new Cantonese romanization system which
-      has many advantages. It is multifunctional, systematic, user-friendly,
-      compatible with all possible modern Cantonese sounds, and solely based on
-      alphanumeric characters without any diacritics and strange symbols.
-      Jyutping can also be used as a Chinese computer input method. Its basic
-      principles are simple, easy to learn, and professional.
+  description: |
+    The Linguistic Society of Hong Kong Cantonese Romanisation Scheme, or
+    known as Jyutping, was designed and proposed by the Linguistic Society of
+    Hong Kong in 1993. Jyutping is a new Cantonese romanization system which
+    has many advantages. It is multifunctional, systematic, user-friendly,
+    compatible with all possible modern Cantonese sounds, and solely based on
+    alphanumeric characters without any diacritics and strange symbols.
+    Jyutping can also be used as a Chinese computer input method. Its basic
+    principles are simple, easy to learn, and professional.
 
   notes:
     - One may need to parse the text in order to generate accurate

data/maps/mext-jpn-Hrkt-Latn-1954.imp

@@ -11,34 +11,36 @@ metadata {
   adoption_date: 1954-12-09
   # 昭和二十九年十二月九日
   description:
-    jp: |
-      国語を書き表わす場合に用いるローマ字のつづり方を次のように定める。
+    The spelling method for Roman characters used when writing Japanese language is as follows.
 
-      まえがき
-      １　一般に国語を書き表わす場合は、第1表に掲げたつづり方によるものとする。
-      ２　国際的関係その他従来の慣例をにわかに改めがたい事情にある場合に限り、第2表に掲げたつづり方によつてもさしつかえない。
-      ３　前二項のいずれの場合においても、おおむねそえがきを適用する。
-    en: |
-      The spelling method for Roman characters used when writing Japanese language is as follows.
+    Preface
+    1. In general, when the language is written, the spelling shown in Table 1 shall be used.
+    2. The spelling methods listed in Table 2 can be used only when there is a situation that is difficult to change due to international relations or other conventional practices.
+    3. In either case of the preceding two paragraphs, the general introduction will apply.
 
-      Preface
-      1. In general, when the language is written, the spelling shown in Table 1 shall be used.
-      2. The spelling methods listed in Table 2 can be used only when there is a situation that is difficult to change due to international relations or other conventional practices.
-      3. In either case of the preceding two paragraphs, the general introduction will apply.
+  original_description: |
+    国語を書き表わす場合に用いるローマ字のつづり方を次のように定める。
+
+    まえがき
+    １　一般に国語を書き表わす場合は、第1表に掲げたつづり方によるものとする。
+    ２　国際的関係その他従来の慣例をにわかに改めがたい事情にある場合に限り、第2表に掲げたつづり方によつてもさしつかえない。
+    ３　前二項のいずれの場合においても、おおむねそえがきを適用する。
 
   notes:
-    - jp: はねる音「ン」はすべてnと書く。
-      en: ン / ん is romanized always n in Kunrei-siki
-    - jp: はねる音を表わすnと次にくる母音字またはyとを切り離す必要がある場合には、nの次に’を入れる。
-      en: When it is necessary to separate the sound n from the vowel or y to follow, the apostrophe is added after the n.
-    - jp: つまる音は、最初の子音字を重ねて表わす。
-      en: The clogged sound is represented by overlapping the first consonant characters.
-    - jp: 長音は母音字の上に＾をつけて表わす。なお、大文字の場合は母音字を並べてもよい。
-      en: Long vowels are expressed in Kunrei-siki by placing a circumflex (^) over a vowel. In the case of capital letters, vowel characters may be arranged.
-    - jp: 特殊音の書き表わし方は自由とする。
-      en: The way of writing special sounds is free.
-    - jp: 文の書きはじめ、および固有名詞は語頭を大文字で書く。なお、固有名詞以外の名詞の語頭を大文字で書いてもよい。
-      en: Begin writing sentences and proper nouns with capital letters. Note that the beginning of nouns other than proper nouns may be written in capital letters.
+    - ン / ん is romanized always n in Kunrei-siki
+    - When it is necessary to separate the sound n from the vowel or y to follow, the apostrophe is added after the n.
+    - The clogged sound is represented by overlapping the first consonant characters.
+    - Long vowels are expressed in Kunrei-siki by placing a circumflex (^) over a vowel. In the case of capital letters, vowel characters may be arranged.
+    - The way of writing special sounds is free.
+    - Begin writing sentences and proper nouns with capital letters. Note that the beginning of nouns other than proper nouns may be written in capital letters.
+
+  original_notes:
+    - はねる音「ン」はすべてnと書く。
+    - はねる音を表わすnと次にくる母音字またはyとを切り離す必要がある場合には、nの次に’を入れる。
+    - つまる音は、最初の子音字を重ねて表わす。
+    - 長音は母音字の上に＾をつけて表わす。なお、大文字の場合は母音字を並べてもよい。
+    - 特殊音の書き表わし方は自由とする。
+    - 文の書きはじめ、および固有名詞は語頭を大文字で書く。なお、固有名詞以外の名詞の語頭を大文字で書いてもよい。
 }
 
 tests {

data/maps/un-ell-Grek-Latn-1987-phonetic.imp

@@ -17,12 +17,12 @@ metadata {
 
   notes:
     - Also included in ISO 843:1997, Annex B, Column 5, and ELOT 743:1982, column 5.
-    - Corrected obvious errors, which occur every time the table has reappeared: χ > x, x > ks, oï > oi.
-    - The vowels are taken from the specification, but some are controversial: /ɑ ɛ/ but /o/.
+    - "Corrected obvious errors, which occur every time the table has reappeared: χ > x, x > ks, oï > oi."
+    - "The vowels are taken from the specification, but some are controversial: /ɑ ɛ/ but /o/."
     - Stress is not indicated. (To do so in IPA would require syllabification in preprocessing, since stress is positioned at syllable breaks)
     - Followed specification in treating final μπ as b, but final ντ as nd. That distinction is dubious. (In ELOT 743:1982, both d and nd are erroneously marked as initial, and no final is given.)
-    - τζ is not correctly transcribed as dz: fixed
-    - not reducing geminated consonants: fixed
+    - "τζ is not correctly transcribed as dz: fixed"
+    - "not reducing geminated consonants: fixed"
 }
 
 tests {

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: interscript-maps
 version: !ruby/object:Gem::Version
-  version: 2.4.1
+  version: 2.4.3
 platform: ruby
 authors:
 - Ribose Inc.
@@ -48,6 +48,7 @@ files:
 - maps/alalc-div-Thaa-Latn-2012.imp
 - maps/alalc-ell-Grek-Latn-1997.imp
 - maps/alalc-ell-Grek-Latn-2010.imp
+- maps/alalc-fas-Arab-Latn-1997.imp
 - maps/alalc-guj-Gujr-Latn-1997.imp
 - maps/alalc-guj-Gujr-Latn-2011.imp
 - maps/alalc-hin-Deva-Latn-1997.imp
@@ -70,7 +71,6 @@ files:
 - maps/alalc-ori-Orya-Latn-2011.imp
 - maps/alalc-pan-Guru-Latn-1997.imp
 - maps/alalc-pan-Guru-Latn-2011.imp
-- maps/alalc-per-Arab-Latn-1997.imp
 - maps/alalc-pli-Deva-Latn-2012.imp
 - maps/alalc-pra-Deva-Latn-2012.imp
 - maps/alalc-rus-Cyrl-Latn-1997.imp