diff --git a/auto_translate.py b/auto_translate.py
new file mode 100644
index 0000000..39f1770
--- /dev/null
+++ b/auto_translate.py
@@ -0,0 +1,109 @@
+import binascii
+import xmltodict
+import xml.etree.ElementTree as ET
+import glob
+import os
+
+from collections import OrderedDict
+from config import *
+from utils import *
+
+class Item():
+    itemid = 0
+    en = ""
+    jp = ""
+
+def dump_table(autotranslate_table):
+    output_filename = 'out/sql/item_translation_table.sql'
+
+    header = '''-- /translate command table
+-- Store as varbinary because of shift-jis shenanigans. English text will be transformed on load.
+DROP TABLE IF EXISTS `item_translation_table`;
+
+/*!40101 SET @saved_cs_client     = @@character_set_client */;
+/*!40101 SET character_set_client = utf8 */;
+CREATE TABLE `item_translation_table` (
+  `itemId` smallint(5) unsigned NOT NULL,
+  `englishText` varbinary(64) NOT NULL DEFAULT 0,
+  `japaneseText` varbinary(64) NOT NULL DEFAULT 0,
+  PRIMARY KEY (`itemId`)
+) ENGINE=Aria TRANSACTIONAL=0 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_general_ci PACK_KEYS=1;
+
+'''
+    with open(output_filename, "w", encoding="utf-8") as file:
+        file.write(header)
+
+        for itemid in sorted(autotranslate_table.keys()):
+            if not 'jp' in autotranslate_table[itemid] or not 'en' in autotranslate_table[itemid]:
+                continue # No key, skip it.
+
+            # Japanese text doesn't seem to use quotes, but will that always be true?
+            en = autotranslate_table[itemid]['en'].replace("'", "\\'").replace("\"","\\\"")
+            jp = autotranslate_table[itemid]['jp'].replace("'", "\\'").replace("\"","\\\"")
+
+            file.write(u"INSERT INTO `item_translation_table` VALUES ({0}, '{1}', '{2}');\n".format(itemid, en, jp))
+
+def auto_translate():
+    # Yuck. polutils doesn't extract generalE2 to the same location.
+    files_en = { "generalE", "../items-general2", "usableE",  "puppetE", "armorE", "armor2E", "weaponE", "slipE", "currencyE", }
+    files_jp = { "generalJ", "general2J", "usableJ", "puppetJ", "armorJ", "armor2J", "weaponJ", "slipJ", "currencyJ"}
+
+    en_items = {}
+    for file in files_en:
+         tree = ET.parse('res/resources/' + file + '.xml')
+         xml = xmltodict.parse(ET.tostring(tree.getroot(), encoding='unicode'))
+         for item in xml["thing-list"]["thing"]:
+            parse_fields(en_items, item)
+
+    jp_items = {}
+    for file in files_jp:
+         tree = ET.parse('res/resources/' + file + '.xml')
+         xml = xmltodict.parse(ET.tostring(tree.getroot(), encoding='unicode'))
+         for item in xml["thing-list"]["thing"]:
+            parse_fields(jp_items, item)
+
+    autotranslate_table = {}
+
+    for item in en_items:
+        if item in autotranslate_table:
+            autotranslate_table[item]['en'] = en_items[item]['name']
+        else:
+            autotranslate_table[item] = {}
+            autotranslate_table[item]['en'] = en_items[item]['name']
+
+    for item in jp_items:
+        if item in autotranslate_table:
+            autotranslate_table[item]['jp'] = jp_items[item]['name']
+        else:
+            autotranslate_table[item] = {}
+            autotranslate_table[item]['jp'] = jp_items[item]['name']
+
+    for at in autotranslate_table:
+        if not 'jp' in autotranslate_table[at]:
+            print("Warning: Item id '{0}' named '{1}' not found in jp table. Please investigate for a real error".format(jp_items[at]['id'], jp_items[at]['name']))
+            continue
+        if not 'en' in autotranslate_table[at]:
+            print("Warning: Item id '{0}' named '{1}' not found in en table. This may not be a real problem -- JP dats can sometimes contain extra data.".format(jp_items[at]['id'], jp_items[at]['name'].encode('utf-8')))
+            continue
+
+    dump_table(autotranslate_table)
+
+def parse_fields(items, item):
+
+    # no clue. sometimes item is "@type" or "field" strings.
+    # check if "item" is a string type, do not parse.
+    if isinstance(item, str):
+        return
+
+    if item["field"][6]["#text"] == ".":
+        return
+
+    items[int(item["field"][0]["#text"])] = {}
+    try:
+        for field in item["field"]:
+            items[int(item["field"][0]["#text"])][field["@name"]] = field["#text"]
+    except:
+        pass
+
+# Enable to test just this file
+#auto_translate()
diff --git a/titles.py b/titles.py
index 6f2b43e..53ad104 100644
--- a/titles.py
+++ b/titles.py
@@ -8,7 +8,7 @@
 # Titles
 ############################
 def titles():
-    print("Generating out/scripts/globals/titles.lua")
+    print("Generating out/scripts/enum/title.lua")
     titles = {}
     titles_xml = xmltodict.parse(ET.tostring(
         ET.parse("res/titles.xml").getroot(), encoding="unicode"))
@@ -35,10 +35,10 @@ def titles():
         #file.write(f"{title_str} = {index_str}, -- {utf8_comment}\n")
     out_string += "}\n"
 
-    with open("out/scripts/globals/titles.lua", "w") as file:
+    with open("out/scripts/enum/title.lua", "w") as file:
         file.write(out_string)
 
-    server_filename = SERVER_DIR + "/scripts/globals/titles.lua"
+    server_filename = SERVER_DIR + "/scripts/enum/title.lua"
     with open(server_filename, 'r+') as server_file:
         raw_server_data = server_file.read()
         # Write to original files (only the text{...} table)
diff --git a/update_extractor.py b/update_extractor.py
index 439e4b0..dd6efcb 100644
--- a/update_extractor.py
+++ b/update_extractor.py
@@ -63,6 +63,7 @@
 from zone_texts import *
 from entity_ids import *
 from misc import *
+from auto_translate import *
 
 ############################
 # Config
@@ -90,6 +91,9 @@
 if not os.path.exists("out/scripts"):
     os.makedirs("out/scripts")
 
+if not os.path.exists("out/scripts/enum"):
+    os.makedirs("out/scripts/enum")
+
 if not os.path.exists("out/scripts/globals"):
     os.makedirs("out/scripts/globals")
 
@@ -119,5 +123,6 @@
 zone_texts()
 #entity_ids()
 #misc()
+#auto_translate()
 
 print("Done")