Incorrect inlined string view comparison after " Add prefix compare for inlined"

[alamb]

Describe the bug

While testing the new arrow-rs branch with DataFusion I found a wrong results bug

• Upgrade arrow/parquet to 56.0.0 datafusion#16690

Here is more detail: apache/datafusion#16690 (comment)

• This commit passes: 674dc17

commit 674dc17b2c423be16d0725a6537b0063ac7b1b58
Author: nathaniel-d-ef <nathaniel@elastiflow.com>
Date:   Sat Jun 28 05:21:45 2025 -0500

This commit fails: aa96097

• Perf: Add prefix compare for inlined compare and change use of inline_value to inline it to a u128  #7748

commit aa960977275f96a42e74569cd4ef833afa38ecf2
Author: Qi Zhu <821684824@qq.com>
Date:   Sun Jun 29 17:02:36 2025 +0800

    Perf: Add prefix compare for inlined compare and change use of inline_value to inline it to a u128  (#7748)

To Reproduce

You can use a local checkout of arrow-rs

 [patch.crates-io]
-arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-array = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-buffer = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-cast = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-data = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-ipc = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-schema = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-select = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-string = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-ord = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-flight = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-array = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-buffer = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-cast = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-data = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-ipc = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-schema = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-select = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-string = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-ord = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+#arrow-flight = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+# parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+
+
+# use local copy in arrow-rs
+
+arrow = { path = "/Users/andrewlamb/Software/arrow-rs/arrow" }
+arrow-array = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-array" }
+arrow-buffer = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-buffer" }
+arrow-cast = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-cast" }
+arrow-data = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-data" }
+arrow-ipc = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-ipc" }
+arrow-schema = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-schema" }
+arrow-select = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-select" }
+arrow-string = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-string" }
+arrow-ord = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-ord" }
+arrow-flight = { path = "/Users/andrewlamb/Software/arrow-rs/arrow-flight" }
+parquet = { path = "/Users/andrewlamb/Software/arrow-rs/parquet" }

And then run apache/datafusion#16690 (comment)

$ cargo test --package datafusion --test parquet_config parquet::row_group_pruning

Expected behavior
The tests should pass

Additional context

[alamb]

One thing I noticed about the tests are they are comparing strings like backend one, backend two, etc

where the first 4 bytes are the same but then differ in their next bytes

[zhuqi-lucas]

take

[zhuqi-lucas]

Thank you @alamb for reporting, i am checking it and try to fix it!

[zhuqi-lucas]

I can reproduce it, it's a corner case, i will submit a fix PR soon!

[zhuqi-lucas]

diff --git a/Cargo.lock b/Cargo.lock
index cae951ef5..cccd81745 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -247,7 +247,7 @@ checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50"
 [[package]]
 name = "arrow"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -270,7 +270,7 @@ dependencies = [
 [[package]]
 name = "arrow-arith"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -283,7 +283,7 @@ dependencies = [
 [[package]]
 name = "arrow-array"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "ahash 0.8.12",
  "arrow-buffer",
@@ -299,7 +299,7 @@ dependencies = [
 [[package]]
 name = "arrow-buffer"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "bytes",
  "half",
@@ -309,7 +309,7 @@ dependencies = [
 [[package]]
 name = "arrow-cast"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -329,7 +329,7 @@ dependencies = [
 [[package]]
 name = "arrow-csv"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-cast",
@@ -343,7 +343,7 @@ dependencies = [
 [[package]]
 name = "arrow-data"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-buffer",
  "arrow-schema",
@@ -354,7 +354,7 @@ dependencies = [
 [[package]]
 name = "arrow-flight"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -380,7 +380,7 @@ dependencies = [
 [[package]]
 name = "arrow-ipc"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -394,7 +394,7 @@ dependencies = [
 [[package]]
 name = "arrow-json"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -415,7 +415,7 @@ dependencies = [
 [[package]]
 name = "arrow-ord"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -427,7 +427,7 @@ dependencies = [
 [[package]]
 name = "arrow-pyarrow"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-data",
@@ -438,7 +438,7 @@ dependencies = [
 [[package]]
 name = "arrow-row"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -450,7 +450,7 @@ dependencies = [
 [[package]]
 name = "arrow-schema"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "bitflags 2.9.1",
  "serde",
@@ -460,7 +460,7 @@ dependencies = [
 [[package]]
 name = "arrow-select"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "ahash 0.8.12",
  "arrow-array",
@@ -473,7 +473,7 @@ dependencies = [
 [[package]]
 name = "arrow-string"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4487,7 +4487,7 @@ dependencies = [
 [[package]]
 name = "parquet"
 version = "55.2.0"
-source = "git+https://github.com/apache/arrow-rs.git?rev=57f96f24a1f69504588ecb2c7c42fb91c0592262#57f96f24a1f69504588ecb2c7c42fb91c0592262"
+source = "git+https://github.com/zhuqi-lucas/arrow-rs?branch=issue_7874#02c1e954815dc1458b03f92c412216ed741e4e6a"
 dependencies = [
  "ahash 0.8.12",
  "arrow-array",
diff --git a/Cargo.toml b/Cargo.toml
index e184d65d6..50eb286bb 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -231,15 +231,15 @@ unused_qualifications = "deny"
 ## Temporary arrow-rs patch until 56.0.0 is released
 
 [patch.crates-io]
-arrow = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-array = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-buffer = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-cast = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-data = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-ipc = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-schema = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-select = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-string = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-ord = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-arrow-flight = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
-parquet = { git = "https://github.com/apache/arrow-rs.git", rev = "57f96f24a1f69504588ecb2c7c42fb91c0592262" }
+arrow = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-array = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-buffer = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-cast = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-data = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-ipc = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-schema = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-select = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-string = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-ord = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+arrow-flight = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }
+parquet = { git = "https://github.com/zhuqi-lucas/arrow-rs", branch = "issue_7874" }

Testing the fixed PR, it works well for datafusion.

[zhuqi-lucas]

After the fixed PR merged, i will also add the datafusion side change to apply it.

[alamb]

Thanks again for your quick resolution of this issue @zhuqi-lucas