From abaf271bf26f95f38d172c3577acad32c74dae61 Mon Sep 17 00:00:00 2001
From: Nitsud Yarg <relay01@gmail.com>
Date: Sun, 3 May 2026 09:29:21 -0700
Subject: [PATCH] some tweaks i dont remember

---
 fugashi_parser.py     | 23 ++++++++++++++++++
 subtitle_processor.py | 54 +++++++++++++++++++++++++++++++++++++++----
 2 files changed, 73 insertions(+), 4 deletions(-)

diff --git a/fugashi_parser.py b/fugashi_parser.py
index a153b69..92f76d4 100644
--- a/fugashi_parser.py
+++ b/fugashi_parser.py
@@ -18,6 +18,7 @@ request_url = "http://127.0.0.1:19633"
 request_timeout = 10
 
 from fugashi import Tagger
+#import unidic
 
 
 # Ensure UTF-8 output
@@ -47,6 +48,28 @@ def has_japanese(text: str) -> bool:
         for ch in text
     )
 
+def anki_fields_term(term : str) -> dict:
+    params = {
+        "text": term,
+        "type": "term",
+        #"markers": ["audio", "cloze-body-kana", "conjugation", "expression", "furigana", "furigana-plain", "glossary", "glossary-brief", "glossary-no-dictionary", "glossary-first", "glossary-first-brief", "glossary-first-no-dictionary", "part-of-speech", "phonetic-transcriptions", "pitch-accents", "pitch-accent-graphs", "pitch-accent-graphs-jj", "pitch-accent-positions", "pitch-accent-categories", "reading", "tags", "clipboard-image", "clipboard-text", "cloze-body", "cloze-prefix", "cloze-suffix", "dictionary", "dictionary-alias", "document-title", "frequencies", "frequency-harmonic-rank", "frequency-harmonic-occurrence", "frequency-average-rank", "frequency-average-occurrence", "screenshot", "search-query", "popup-selection-text", "sentence", "sentence-furigana", "sentence-furigana-plain", "url"],
+        "markers": ["audio", "expression", "cloze-body" , "furigana","furigana-plain", "part-of-speech", "reading", "glossary-first", "glossary-plain-no-dictionary" , "glossary-first-no-dictionary" , "dictionary", "frequencies", "frequency-average-rank", "frequency-average-occurrence", "screenshot", "sentence", "sentence-furigana", "sentence-furigana-plain" ],
+        "maxEntries": 2,
+        "includeMedia": False,
+    }
+    response = requests.post(request_url + "/ankiFields", json = params, timeout = request_timeout)
+    return json.loads(response.text)
+
+def term_entries(term :str) -> dict:
+    #print("Requesting termEntries:")
+    params = {
+        "term": term,
+        "maxEntries": 1,
+        "includeMedia": False,
+    }
+    response = requests.post(request_url + "/termEntries", json = params, timeout = request_timeout)
+    return json.loads(response.text)
+
 
 def is_content_word(pos1: str) -> bool:
     return pos1 in {"名詞", "動詞", "形容詞", "副詞"}
diff --git a/subtitle_processor.py b/subtitle_processor.py
index ae0a3cb..ce88236 100644
--- a/subtitle_processor.py
+++ b/subtitle_processor.py
@@ -1,5 +1,27 @@
 from fugashi_parser import *
 
+import os
+from pathlib import Path
+import shutil
+import hashlib
+
+
+destination_directory_name = '/run/media/deck/YF8SD/Sync/Notebooks/Obsidian/Japanese/Yu-Gi-Oh/AnimeSubs/'
+destination_path = Path(destination_directory_name)
+
+dictionary_name = 'Jitendex.org [2026-04-04]'
+
+#print(annotate_phrase("私は学校へ行きます"))
+
+## Make sure directories exist
+if not destination_path.exists():
+    destination_path.mkdir(parents=True, exist_ok=True)
+if not destination_path.joinpath('_attachments').exists():
+    destination_path.joinpath('_attachments').mkdir(parents=True, exist_ok=True)
+if not destination_path.joinpath('vocab').exists():
+    destination_path.joinpath('vocab').mkdir(parents=True, exist_ok=True)
+
+
 # ---------- CLI ----------
 
 def main():
@@ -18,11 +40,12 @@ def main():
 
     # Example: Modify the 3rd line (index 2)
     #lines[2] = "This is the new subtitle text\n"
-    ## First lien has some unicode bom bs 
-    for i in range(len(lines)):
+    ## First lien has some unicode bom bs
+    i = 0 
+    while i < len(lines):
         if re.fullmatch(r'\d+', lines[i].strip()) or i == 0:
             notes_lines.append('##### ' + lines[i].strip() + " `" + lines[i+1].strip() + "`" + "\n")
-            i+=1
+            i+=2
             continue
         if has_japanese(lines[i]):
             print(lines[i])
@@ -33,7 +56,30 @@ def main():
             if lines[i] != result['kana_reading']:
                 lines[i] = result['kana_reading'] + "\n" # lost the new line
         else:
-           notes_lines.append("`" + lines[i].strip() + "`" + "\n")
+           #notes_lines.append("`" + lines[i].strip() + "`" + "\n")
+           notes_lines.append(lines[i])
+        i+=1
+
+    ## Lookup vocab
+    for word in vocab:
+        #vocab_hash = hashlib.shake_256((word + dictionary_name).encode()).hexdigest(4)
+        if Path(destination_path.joinpath('vocab/' + word + '.md')).exists() == False:
+            definitions = anki_fields_term(word)
+            if definitions.get("fields"):
+                with open(destination_path.as_posix() + '/vocab/' + word + '.md', 'w') as vocab_note:
+                    for entry in definitions.get("fields"):
+                        vocab_note.write("\n")
+                        print(entry['expression'], file=vocab_note)
+                        #print(entry['furigana-plain'], file=vocab_note)
+                        if entry['expression'] != entry['reading']:
+                            print(entry['reading'], file=vocab_note)
+                        print(entry['furigana'], file=vocab_note)
+                        #print(f"Part of Speech: {entry['part-of-speech']}", file=vocab_note)
+                        print(f"Frequency: {entry['frequency-average-rank']}", file=vocab_note)
+                        #print(entry['glossary-plain-no-dictionary'], file=vocab_note)
+                        vocab_note.write(entry['glossary-first'])
+                        print("##### Other Definitions", file=vocab_note)
+                        print(entry['glossary-plain-no-dictionary'], file=vocab_note)
 
     with open("subtitle.srt", "w", encoding="utf-8") as f:
         f.writelines(lines)