From a152250f1113762223bc56f76f2613d9d896b73d Mon Sep 17 00:00:00 2001
From: frederik <frederik.slos@gmail.com>
Date: Sun, 26 Oct 2025 12:05:42 +0100
Subject: [PATCH] add language helpers

---
 scripts/utils/helpers.py    | 39 ++++++++++++++++++++++++++++++++++++
 scripts/utils/maintainer.py | 40 +++++++++++++++++++++++++++++++++++++
 2 files changed, 79 insertions(+)
 create mode 100644 scripts/utils/maintainer.py

diff --git a/scripts/utils/helpers.py b/scripts/utils/helpers.py
index ae68cf3..59abfa2 100644
--- a/scripts/utils/helpers.py
+++ b/scripts/utils/helpers.py
@@ -1,8 +1,10 @@
 import datetime
 import glob
+import json
 import os
 import re
 import subprocess
+from collections import OrderedDict
 from configparser import ConfigParser
 from itertools import chain
 
@@ -11,6 +13,7 @@ from tzlocal import get_localzone
 _settings = None
 
 DB_PATH = os.path.expanduser('~/BirdNET-Pi/scripts/birds.db')
+MODEL_PATH = os.path.expanduser('~/BirdNET-Pi/model')
 ANALYZING_NOW = os.path.expanduser('~/BirdSongs/StreamData/analyzing_now.txt')
 FONT_DIR = os.path.expanduser('~/BirdNET-Pi/homepage/static')
 
@@ -118,3 +121,39 @@ def get_wav_files():
     open_recs = get_open_files_in_dir(rec_dir)
     files = [file for file in files if file not in open_recs]
     return files
+
+
+def get_language(language=None):
+    if language is None:
+        language = get_settings()['DATABASE_LANG']
+    file_name = os.path.join(MODEL_PATH, f'l18n/labels_{language}.json')
+    with open(file_name) as f:
+        ret = json.loads(f.read())
+    return ret
+
+
+def save_language(labels, language):
+    file_name = os.path.join(MODEL_PATH, f'l18n/labels_{language}.json')
+    with open(file_name, 'w') as f:
+        f.write(json.dumps(OrderedDict(sorted(labels.items())), indent=2, ensure_ascii=False))
+
+
+def get_model_labels(model=None):
+    if model is None:
+        model = get_settings()['MODEL']
+    file_name = os.path.join(MODEL_PATH, f'{model}_Labels.txt')
+    with open(file_name) as f:
+        labels = [line.strip() for line in f.readlines()]
+    if labels and labels[0].count('_') == 1:
+        labels = [re.sub(r'_.+$', '', label) for label in labels]
+    return labels
+
+
+def set_label_file():
+    lang = get_language()
+    labels = [f'{label}_{lang[label]}\n' for label in get_model_labels()]
+    file_name = os.path.join(MODEL_PATH, 'labels.txt')
+    if os.path.islink(file_name):
+        os.remove(file_name)
+    with open(file_name, 'w') as f:
+        f.writelines(labels)
diff --git a/scripts/utils/maintainer.py b/scripts/utils/maintainer.py
new file mode 100644
index 0000000..17d98a8
--- /dev/null
+++ b/scripts/utils/maintainer.py
@@ -0,0 +1,40 @@
+import os
+
+from scripts.utils.helpers import MODEL_PATH, save_language
+
+
+def get_labels(model, language=None):
+    postfix = '' if language is None else f'_{language}'
+    file_name = os.path.join(MODEL_PATH, f'labels_{model}/labels{postfix}.txt')
+    with open(file_name) as f:
+        ret = [line.strip() for line in f.readlines()]
+    return ret
+
+
+def as_dict(labels, den="_", key=0, value=1):
+    return {label.split(den)[key]: label.split(den)[value] for label in labels}
+
+
+def create_language(language):
+    en_l18n = as_dict(get_labels('l18n', 'en'))
+    l18n = as_dict(get_labels('l18n', language))
+    new_language = as_dict(get_labels('nm', language))
+
+    for sci_name, com_name in l18n.items():
+        if sci_name not in new_language or new_language[sci_name] == sci_name:
+            new_language[sci_name] = com_name
+            continue
+
+        # now check if the l18n version is translated
+        if com_name != new_language[sci_name] and new_language[sci_name] == en_l18n[sci_name]:
+            print(f'changing {new_language[sci_name]} -> {com_name}')
+            new_language[sci_name] = com_name
+
+    save_language(new_language, language)
+
+
+def create_all_languages():
+    languages = ['af', 'ar', 'ca', 'cs', 'da', 'de', 'en', 'es', 'et', 'fi', 'fr', 'hr', 'hu', 'id', 'is', 'it', 'ja',
+                 'ko', 'lt', 'lv', 'nl', 'no', 'pl', 'pt', 'ro', 'ru', 'sk', 'sl', 'sr', 'sv', 'th', 'tr', 'uk', 'zh']
+    for language in languages:
+        create_language(language)