From: Urban Wallasch <urban.wallasch@freenet.de>
Date: Mon, 21 Jun 2021 21:51:07 +0000 (+0200)
Subject: * Correctly handle the case where a tagged field may appear more than once in a kanji... 
X-Git-Tag: v0.1.0~87
X-Git-Url: https://git.packet-gain.de/?a=commitdiff_plain;h=9b072f72e2db48d55200c2b8bf0c653cfab98bc0;p=jiten-pai.git

* Correctly handle the case where a tagged field may appear more than once in a kanjidic entry (e.g. stroke count).
---

diff --git a/kanjidic.py b/kanjidic.py
index b49a034..da7dd2c 100755
--- a/kanjidic.py
+++ b/kanjidic.py
@@ -205,7 +205,9 @@ def _kanjidic_load(dict_fname):
                     line = re_tags.sub('', line, 1)
                 for t in tlist:
                     for k in ktable:
-                        if t[:len(k[0])] == k[0]:
+                        # if a tag appears more than once the first one wins,
+                        # e.g. 'S<num>' (stroke count)
+                        if t[:len(k[0])] == k[0] and not info[k[1]]:
                             info[k[1]] = t[len(k[0]):]
                             break
                 # get readings (i.e. all that's left)