Update this for indradb 4

Cargo.toml

@@ -1,13 +1,14 @@
 [package]
 name = "indradb-sled"
-version = "0.1.0"
+version = "4.0.0-RC1"
 authors = ["Yusuf Simonson <simonson@gmail.com>"]
 description = "A sled-backed datastore for IndraDB"
 homepage = "https://indradb.github.io"
 repository = "https://github.com/indradb/sled"
 keywords = ["graph", "database"]
 categories = ["database", "database-implementations"]
 license = "MPL-2.0"
+edition = "2021"
 
 [lib]
 name = "indradb_sled"
@@ -19,9 +20,11 @@ test-suite = ["indradb-lib/test-suite", "tempfile"]
 bench-suite = ["indradb-lib/bench-suite", "tempfile"]
 
 [dependencies]
-chrono = { version = "0.4.19", features = ["serde"] }
-indradb-lib = "^2.2.0"
-serde_json = "^1.0.57"
-sled = { version = "0.34.6", features = ["compression", "no_metrics"] }
-tempfile = { version = "^3.2.0", optional = true}
-uuid = { version = "~0.8.2", features = ["v1", "serde"] }
+chrono = { version = "0.4", features = ["serde"] }
+indradb-lib = "4.0"
+serde_json = "1.0"
+sled = { version = "0.34", features = ["compression"] }
+tempfile = { version = "3.10", optional = true }
+uuid = { version = "1.9", features = ["v1", "serde"] }
+ecow = { version = "0.2.2" }
+thiserror = { version = "1.0" }

README.md

@@ -2,4 +2,18 @@
 
 This is an implementation of the IndraDB datastore for sled.
 
-The sled datastore is not production-ready yet. sled itself is pre-1.0, and makes no guarantees about on-disk format stability. Upgrading IndraDB may require you to [manually migrate the sled datastore.](https://docs.rs/sled/0.34.6/sled/struct.Db.html#method.export) Additionally, there is a standing issue that prevents the sled datastore from having the same level of safety as the RocksDB datastore.
+## General hints
+
+The sled datastore is not production-ready yet. sled itself is pre-1.0, and makes no guarantees about on-disk format
+stability.
+Upgrading IndraDB may require you
+to [manually migrate the sled datastore.](https://docs.rs/sled/0.34.6/sled/struct.Db.html#method.export)
+Additionally, there is a standing issue that prevents the sled datastore from having the same level of safety as the
+RocksDB datastore.
+Performance of sled in two-hop-queries is better than rocksdb though. We measured about 40% faster query times.
+
+## TODO / Notices:
+
+- This version does index all properties by default (and therefore duplicates the data).
+  This is adding the benefit of faster queries but also increases the storage requirements.
+- Batch operations are not optimized yet,