feat(stdlib) Add new parse_bytes function (#1198)

* Basic implementation for parse_bytes function

* Support exabyte

* Add base optional argument

* change default base to 2 and update testcase

* allow ambiguous unit

* Add more examples and add VRL test

* Add bench and fix clippy

* Adopt parse-size crate

* update comment

* Add change log

* Bump MSVC to 1.81 required by `parse-size` crate

Cargo.toml

@@ -10,7 +10,7 @@ repository = "https://github.com/vectordotdev/vrl"
 readme = "README.md"
 keywords = ["vector", "datadog", "compiler"]
 categories = ["compilers"]
-rust-version = "1.80" # msrv
+rust-version = "1.81" # msrv
 
 [workspace]
 members = [
@@ -93,6 +93,7 @@ stdlib = [
   "dep:nom",
   "dep:ofb",
   "dep:once_cell",
+  "dep:parse-size",
   "dep:percent-encoding",
   "dep:prost",
   "dep:prost-reflect",
@@ -160,6 +161,7 @@ once_cell = { version = "1", default-features = false, features = ["std"], optio
 ordered-float = { version = "4", default-features = false, optional = true }
 md-5 = { version = "0.10", optional = true }
 paste = { version = "1", default-features = false, optional = true }
+parse-size = { version = "1.1.0",  optional = true }
 peeking_take_while = { version = "1", default-features = false, optional = true }
 percent-encoding = { version = "2", optional = true }
 pest = { version = "2", default-features = false, optional = true, features = ["std"] }

LICENSE-3rdparty.csv

@@ -174,6 +174,7 @@ owo-colors,https://github.com/jam1garner/owo-colors,MIT,jam1garner <8260240+jam1
 pad,https://github.com/ogham/rust-pad,MIT,Ben S <[email protected]>
 parking,https://github.com/smol-rs/parking,Apache-2.0 OR MIT,"Stjepan Glavina <[email protected]>, The Rust Project Developers"
 parking_lot,https://github.com/Amanieu/parking_lot,MIT OR Apache-2.0,Amanieu d'Antras <[email protected]>
+parse-size,https://github.com/kennytm/parse-size,MIT,kennytm <[email protected]>
 paste,https://github.com/dtolnay/paste,MIT OR Apache-2.0,David Tolnay <[email protected]>
 peeking_take_while,https://github.com/fitzgen/peeking_take_while,MIT OR Apache-2.0,Nick Fitzgerald <[email protected]>
 pest,https://github.com/pest-parser/pest,MIT OR Apache-2.0,Dragoș Tiselice <[email protected]>

benches/stdlib.rs

@@ -102,6 +102,7 @@ criterion_group!(
               parse_aws_alb_log,
               parse_aws_cloudwatch_log_subscription_message,
               parse_aws_vpc_flow_log,
+              parse_bytes,
               parse_common_log,
               parse_csv,
               parse_duration,
@@ -1644,6 +1645,15 @@ bench_function! {
     }
 }
 
+bench_function! {
+    parse_bytes => vrl::stdlib::ParseBytes;
+
+    literal {
+        args: func_args![value: "1024KiB", unit: "MiB"],
+        want: Ok(1.0),
+    }
+}
+
 bench_function! {
     parse_common_log => vrl::stdlib::ParseCommonLog;
 

changelog.d/1198.feature.md

@@ -0,0 +1 @@
+Added new `parse_bytes` function to parse given bytes string such as `1MiB` or `1TB` either in binary or decimal base.

lib/tests/tests/functions/parse_bytes.vrl

@@ -0,0 +1,3 @@
+# result: 1024.0
+
+parse_bytes!("1KB", unit: "B")

src/stdlib/mod.rs

@@ -141,6 +141,7 @@ cfg_if::cfg_if! {
         mod parse_aws_alb_log;
         mod parse_aws_cloudwatch_log_subscription_message;
         mod parse_aws_vpc_flow_log;
+        mod parse_bytes;
         mod parse_cef;
         mod parse_cbor;
         mod parse_common_log;
@@ -326,6 +327,7 @@ cfg_if::cfg_if! {
         pub use parse_aws_alb_log::ParseAwsAlbLog;
         pub use parse_aws_cloudwatch_log_subscription_message::ParseAwsCloudWatchLogSubscriptionMessage;
         pub use parse_aws_vpc_flow_log::ParseAwsVpcFlowLog;
+        pub use parse_bytes::ParseBytes;
         pub use parse_cbor::ParseCbor;
         pub use parse_cef::ParseCef;
         pub use parse_common_log::ParseCommonLog;
@@ -517,6 +519,7 @@ pub fn all() -> Vec<Box<dyn Function>> {
         Box::new(ParseAwsAlbLog),
         Box::new(ParseAwsCloudWatchLogSubscriptionMessage),
         Box::new(ParseAwsVpcFlowLog),
+        Box::new(ParseBytes),
         Box::new(ParseCbor),
         Box::new(ParseCef),
         Box::new(ParseCommonLog),