From 1a97d7db448b513450b650a9b78bf3b4a7671f10 Mon Sep 17 00:00:00 2001
From: DEls <davide.els1@outlook.com>
Date: Sat, 19 Sep 2026 19:36:50 +0200
Subject: [PATCH] fix(migration): preserve quoted index semantics

---
 scripts/pb-add-dn-staged-ingestion.py | 65 +++++++++++++++++++++++++--
 1 file changed, 62 insertions(+), 3 deletions(-)

diff --git a/scripts/pb-add-dn-staged-ingestion.py b/scripts/pb-add-dn-staged-ingestion.py
index d6e497e..c656a76 100644
--- a/scripts/pb-add-dn-staged-ingestion.py
+++ b/scripts/pb-add-dn-staged-ingestion.py
@@ -54,20 +54,70 @@ def field(name, field_type="text", **options):
 
 def index_name(statement):
     """Extract an SQLite/PocketBase index name from a CREATE INDEX statement."""
-    match = re.search(r"\bINDEX\s+`?([^`\s]+)`?", statement, re.IGNORECASE)
+    match = re.search(r"\bINDEX\s+(?:`([^`]+)`|\"([^\"]+)\"|([^\s(]+))", statement, re.IGNORECASE)
     if not match:
         raise ValueError(f"not a CREATE INDEX statement: {statement!r}")
-    return match.group(1).strip().lower()
+    return next(value for value in match.groups() if value is not None).strip().lower()
 
 
 def normalise_index_sql(statement):
     """Compare index SQL by meaning, not PocketBase's formatting choices."""
     value = statement.strip().rstrip(";")
-    value = re.sub(r"`([^`]*)`", lambda match: f"`{match.group(1).lower()}`", value)
+    chunks = []
+    literals = []
+    index = 0
+    while index < len(value):
+        character = value[index]
+        if character == "'":
+            start = index
+            index += 1
+            while index < len(value):
+                if value[index] == "'":
+                    if index + 1 < len(value) and value[index + 1] == "'":
+                        index += 2
+                        continue
+                    index += 1
+                    break
+                index += 1
+            literals.append(value[start:index])
+            chunks.append(f"\x00{len(literals) - 1}\x00")
+            continue
+        if character in '`"':
+            quote = character
+            index += 1
+            identifier = []
+            while index < len(value):
+                if value[index] == quote:
+                    if index + 1 < len(value) and value[index + 1] == quote:
+                        identifier.append(quote)
+                        index += 2
+                        continue
+                    index += 1
+                    break
+                identifier.append(value[index])
+                index += 1
+            chunks.append(f"`{''.join(identifier).lower()}`")
+            continue
+        if character.isalnum() or character in "_$":
+            end = index + 1
+            while end < len(value) and (value[end].isalnum() or value[end] in "_$"):
+                end += 1
+            chunks.append(value[index:end].lower())
+            index = end
+            continue
+        chunks.append(character)
+        index += 1
+
+    value = "".join(chunks)
     value = re.sub(r"\s+", " ", value).strip().lower()
     # SQLite/PocketBase may insert or remove spaces around these tokens while
     # preserving the same columns, uniqueness and WHERE predicate.
     value = re.sub(r"\s*([(),=<>!])\s*", r"\1", value)
+    # Quoted identifiers and safe unquoted identifiers share one canonical
+    # representation; single-quoted literals are placeholders until restore.
+    value = value.replace("`", "")
+    for index, literal in enumerate(literals):
+        value = value.replace(f"\x00{index}\x00", literal)
     return value
 
 
@@ -208,6 +258,15 @@ def self_test():
     ]
     for stored, near_miss in zip(pocketbase_stored, near_misses):
         assert merge_indexes([stored], [near_miss])[1] == [("replace", near_miss)]
+    identifier_styles = (
+        "CREATE UNIQUE INDEX `idx_style` ON `Delivery_Notes` (`Batch_Note`) WHERE `Batch_Note` != ''",
+        'create unique index "IDX_STYLE" on "DELIVERY_NOTES" ("BATCH_NOTE") where "BATCH_NOTE" != \'\'',
+        "CREATE UNIQUE INDEX idx_style ON delivery_notes (batch_note) WHERE batch_note != ''",
+    )
+    assert merge_indexes([identifier_styles[0]], [identifier_styles[1]])[1] == []
+    assert merge_indexes([identifier_styles[0]], [identifier_styles[2]])[1] == []
+    literal_near_miss = identifier_styles[2].replace("!= ''", "!= ' '")
+    assert merge_indexes([identifier_styles[0]], [literal_near_miss])[1] == [("replace", literal_near_miss)]
     print("self-test: index replacement and idempotence passed")
     return 0