Handle symbols/commands at the start of line

src/orig.py

@@ -9,14 +9,37 @@ import parser
 import flow
 import fix
 
-japanese_ranges = [
+def line_should_be_translated(line: str) -> bool:
+    japanese_ranges = [
         (0x4E00, 0x9FFF),  # Kanji
         (0x3040, 0x309F),  # Hiragana
         (0x30A0, 0x30FF),  # Katakana
         (0xFF00, 0xFFEF),  # Full-width Roman characters and symbols
         (0x3000, 0x303F),  # CJK symbols and punctuation (including 「」)
         (8220, 8220), # The character “
-]
+    ]
+
+
+    for start, end in japanese_ranges:
+        if start <= ord(line[0]) <= end:
+            return True
+
+    # if line starts with special commands
+    if line.startswith(('!s', '!w', '!d', '@', '¥')):
+        # ignore line after comment
+        comment_i = line.find(';')
+        if comment_i != -1:
+            line = line[:comment_i]
+
+        # Check if line has japanese chars
+        for c in line:
+            for start, end in japanese_ranges:
+                if start <= ord(c) <= end:
+                    return True
+
+    return False
+
+
 
 debug_current_line = -1
 
@@ -50,8 +73,9 @@ def process_sections():
     origfile.close()
 
 
-def get_symbols(line: str) -> list[str]:
+def get_symbols(line: str) -> (str, list[str]):
     res = []
+    start_symbol = ''
 
     inbetween = 0
     i = 0
@@ -81,14 +105,23 @@ def get_symbols(line: str) -> list[str]:
             continue
 
         # Only reaches this if it's a symbol
-        #print(symbol)
+
+        # Each symbol acts as a separator between dialog texts, if we
+        # have two symbols next to eachother, then we print more dialog
+        # than we should. Concatenate consicutive symbols together to
+        # prevent this
         if inbetween > 0:
             res.append(symbol)
+        else:
+            # Symbols at the start should not have dialog before them.
+            # Treat them as a special "start_symbol"
+            if len(res) == 0:
+                start_symbol = symbol
             else:
                 res[-1] += symbol
         inbetween = 0
 
-    return res
+    return start_symbol, res
 
 
 # Given a set of translation files, the original file and the output file
@@ -103,19 +136,18 @@ def write_translated(outfile, origfile, translation_file_paths):
             debug_current_line += 1
 
             # Check if the current line is a dialogue line or not
-            found = False
-            for start, end in japanese_ranges:
-                if start <= ord(line[0]) <= end:
-                    found = True
 
-            if found:
+            if line_should_be_translated(line):
-                symbols = get_symbols(line)
+                start, symbols = get_symbols(line)
                 print("\n-", debug_current_line, transfilepath, symbols)
 
                 outfile.write('`')
+                outfile.write(start)
 
-                _printed_line_jp = ""
+                print([start])
-                _printed_line_en = ""
+
+                _printed_line_jp = start
+                _printed_line_en = start
                 while True:
                     _printed_line_jp += structure[0].text_jp
                     _printed_line_en += structure[0].text_en