chore: update redb for performance

Cargo.lock

@@ -1769,9 +1769,9 @@ dependencies = [
 
 [[package]]
 name = "redb"
-version = "3.0.0"
+version = "3.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "225e8bf881033e020ed87e9f10fc6254cf3ebab8d440e6fbb4c7b34bec2a0543"
+checksum = "3fefa3e5ff4a369819c3d6df4195873d6f9abad109f13c0d505dbe119cfabb10"
 dependencies = [
  "libc",
 ]

Cargo.toml

@@ -18,7 +18,7 @@ serde_json = "1.0.141"
 strum = { version = "0.27.2", features = ["derive"] }
 tokio = { version = "1.47.1", features = ["rt-multi-thread"] }
 compio = { version = "0.15.0", features = ["runtime"] }
-redb = "3.0.0"
+redb = "3.0.1"
 crabtime = { git = "https://github.com/wdanilo/crabtime.git", rev = "2ed856f5" }
 
 parquet = "56.0.0"

utils/rev_hashed_userid.py

@@ -0,0 +1,20 @@
+from sys import argv
+import polars as pl
+import polars_hash as pl_hash
+
+file = argv[1]
+
+user_ids = (
+    pl.DataFrame({"user_id_num": range(11000000, 13000001)})
+    .with_columns(
+        pl.col("user_id_num")
+        .cast(pl.String)
+        .add("Lt2N5xgjJOqRsT5qVt7wWYw6SqOPZDI7")
+        .alias("user_id"),
+    )
+    .with_columns(pl_hash.col("user_id").chash.sha2_256().str.head(16))
+    .join(pl.read_parquet(file), on="user_id", how="inner")["user_id_num"]
+)
+
+with open("id.txt", "w", encoding="utf-8") as f:
+    f.write("\n".join(map(str, user_ids)))