change other person ner labels to PERSON

xxyzz · Jul 13, 2021 · 4e9eca9 · 4e9eca9
1 parent 12adf50
commit 4e9eca9
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 3 deletions.
diff --git a/__init__.py b/__init__.py
@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 from calibre.customize import InterfaceActionBase
 
-VERSION = (3, 10, 0)
+VERSION = (3, 10, 1)
 
 
 class WordDumbDumb(InterfaceActionBase):

diff --git a/parse_job.py b/parse_job.py
@@ -118,6 +118,8 @@ def find_named_entity(start, x_ray, doc, is_kfx):
     for ent in doc.ents:
         if ent.label_ not in labels:
             continue
+        if ent.label_ in ['PER', 'persName']:
+            ent.label_ = 'PERSON'
 
         if is_kfx:
             ent_start = start + len(doc.text[:ent.start_char])

diff --git a/x_ray.py b/x_ray.py
@@ -24,7 +24,7 @@ def __init__(self, conn, lang):
         self.terms_counter = Counter()
         self.pending_terms = {}
         self.pending_people = {}
-        self.lang = lang
+        self.wikipedia_api = f'https://{lang}.wikipedia.org/w/api.php'
 
         import requests
         self.s = requests.Session()
@@ -65,7 +65,7 @@ def insert_wiki_intro(self, is_people, title, intro):
                 self.conn, (intro, title, 1, entity['id']))
 
     def search_wikipedia(self, is_people, dic):
-        r = self.s.get(f'https://{self.lang}.wikipedia.org/w/api.php',
+        r = self.s.get(self.wikipedia_api,
                        params={'titles': '|'.join(dic.keys())})
         data = r.json()
         converts = {}