Merge pull request #134 from lcnetdev/casei

Fix case-insensitive R2S.

scriptshifter/tables/__init__.py

@@ -16,7 +16,7 @@
     from yaml import Loader
 
 from scriptshifter import DB_PATH
-from scriptshifter.exceptions import BREAK, ConfigError
+from scriptshifter.exceptions import BREAK, ApiError, ConfigError
 
 
 __doc__ = """
@@ -209,6 +209,9 @@ def populate_table(conn, tid, tname):
     if "roman_to_script" in data:
         flags |= FEAT_R2S
 
+    if not data.get("general", {}).get("case_sensitive", True):
+        flags |= FEAT_CASEI
+
     conn.execute(
             "UPDATE tbl_language SET features = ? WHERE id = ?",
             (flags, tid))
@@ -555,6 +558,9 @@ def get_lang_general(conn, lang):
             FROM tbl_language WHERE name = ?""", (lang,))
     lang_data = lang_q.fetchone()
 
+    if not lang_data:
+        raise ApiError(f"No language data found for {lang}", 404)
+
     return {
         "id": lang_data[0],
         "data": {

scriptshifter/tables/data/_chinese_base.yml

@@ -1,13 +1,13 @@
 # This file is derived and kept in sync with Princeton's OCLC Connexion Pinyin
 # converter (https://github.com/pulibrary/oclcpinyin/).
 
-general: # Section names and other keywords are all snake_cased.
+general:  # Section names and other keywords are all snake_cased.
   name: Chinese base (from Princeton)
   parents:
     - _ignore_base
 
 script_to_roman:
-  map: # Mapping section.
+  map:  # Mapping section.
     "\u5DF4\u57FA\u65AF\u5766\u4F0A\u65AF\u862D\u5171\u548C\u570B": "Bajisitan Yisilan Gongheguo "
     "\u5DF4\u57FA\u65AF\u5766\u4F0A\u65AF\u5170\u5171\u548C\u56FD": "Bajisitan Yisilan Gongheguo "
     "\u5DF4\u97F3\u90ED\u695E\u8499\u53E4\u81EA\u6CBB\u5DDE": "Bayinguoleng Menggu Zizhizhou "

scriptshifter/tables/data/arabic.yml

@@ -1,9 +1,11 @@
 # Arabic S2R using the 3rd-party ArabicTransliterator library:
 # https://github.com/MTG/ArabicTransliterator
 
+---
 general:
   name: Arabic
   description: Arabic S2R using a 3rd party library.
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/burmese.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Burmese (Myanmar)
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/chinese.yml

@@ -2,10 +2,13 @@
 #
 # All other Chinese mappings are kept in _chinese_base.yml. This mapping only
 # adds an overlay for parsing numerals and Scriptshifter-specific features.
+
+---
 general:
   name: Chinese
   parents:
     - _chinese_base
+  case_sensitive: false
 
 options:
   - id: marc_field

scriptshifter/tables/data/devanagari.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Devanagari
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/divehi_thaana.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Divehi (Thaana)
+  case_sensitive: false
 
 roman_to_script:
   map:
@@ -50,7 +52,7 @@ roman_to_script:
     "bb": "\u0787\u07B0\u0784"
     "b": "\u0784"
 
-    # THAANA LETTER "L/l" WITH DOT BELOW (0323) 
+    # THAANA LETTER "L/l" WITH DOT BELOW (0323)
     "L\u0323": "\u0785"
     "l\u0323": "\u0785"
     "K": "\u0786"

scriptshifter/tables/data/dogri_devanagari.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Dogri (Devanagari)
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/gujarati.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Gujarati
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/gurmukhi.yml

@@ -1,5 +1,7 @@
+---
 general:
   name: Punjabi (Gurmukhi)
+  case_sensitive: false
 
 script_to_roman:
   hooks:

scriptshifter/tables/data/hebrew.yml

@@ -1,6 +1,8 @@
+---
 general:
   name: Hebrew
   description: Hebrew S2R.
+  case_sensitive: false
 
 options:
   - id: genre
@@ -19,4 +21,3 @@ script_to_roman:
     post_config:
       -
         - hebrew.dicta_api.s2r_post_config
-