feat: bolero feature for profiling protobuf (#1213)

This was already present, but as a dev dependency. This switches it
to be an optional dependency so that other crates can use bolero with
these types.

Author: morrisonlevi

LICENSE-3rdparty.yml

@@ -12,11 +12,13 @@ license.workspace = true
 bench = false
 
 [features]
+bolero = ["dep:bolero"]
 prost_impls = ["dep:prost"]
 
 [dependencies]
 prost = { version = "0.13", optional = true }
+bolero = { version = "0.13", default-features = false, optional = true }
 
 [dev-dependencies]
 bolero = "0.13"
-datadog-profiling-protobuf = { path = ".", features = ["prost_impls"] }
+datadog-profiling-protobuf = { path = ".", features = ["bolero", "prost_impls"] }

datadog-profiling-protobuf/Cargo.toml

@@ -8,7 +8,7 @@ use std::io::{self, Write};
 /// useful to Datadog right now, so we save the bytes/ops.
 #[repr(C)]
 #[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Function {
     /// Unique nonzero id for the function.
     pub id: Record<u64, 1, NO_OPT_ZERO>,

datadog-profiling-protobuf/src/function.rs

@@ -4,11 +4,11 @@
 use super::{Record, StringOffset, Value, WireType, OPT_ZERO};
 use std::io::{self, Write};
 
-/// Label includes additional context for this sample. It can include things
+/// A label includes additional context for this sample. It can include things
 /// like a thread id, allocation size, etc.
 #[repr(C)]
-#[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[derive(Copy, Clone, Debug, Default, Eq, PartialEq, Hash)]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Label {
     /// An annotation for a sample, e.g. "allocation_size".
     pub key: Record<StringOffset, 1, OPT_ZERO>,

datadog-profiling-protobuf/src/label.rs

@@ -88,9 +88,9 @@ use std::io::{self, Write};
 /// of elements in the array.
 ///
 /// [Condensed Reference Card]: https://protobuf.dev/programming-guides/encoding/#cheat-sheet
-#[derive(Copy, Clone, Default, Eq, PartialEq)]
+#[derive(Copy, Clone, Default, Eq, PartialEq, Hash)]
 #[repr(transparent)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Record<P: Value, const F: u32, const O: bool> {
     /// The value of the record. This is pub because of a quirk in Rust's
     /// orphan rules which prevent implementing `From<Record<P,...> for P`.
@@ -128,7 +128,7 @@ pub unsafe trait Value: Default + Eq {
     ///                size encoded as int32 varint
     /// ```
     ///
-    /// Calculate the number of bytes for `(message |  string | packed)` only.
+    /// Calculate the number of bytes for `(message | string | packed)` only.
     ///
     /// For a varint, returns between 1 and 10 bytes for the number of bytes
     /// used to encode the varint.
@@ -152,7 +152,9 @@ pub const OPT_ZERO: bool = true;
 /// Intended to be provided to a [`Record`] to mean that it shouldn't optimize
 /// for a value of zero. Should be used on fields that should not be zero, such
 /// as `Mapping.id` and for Records which hold arrays, since that would cause
-/// the length of the decoded array to change, which is unexpected.
+/// the length of the decoded array to change, which is unexpected. Things
+/// like sample types shouldn't get optimized away, since they get used
+/// element-wise and this would screw up the pairing.
 pub const NO_OPT_ZERO: bool = false;
 
 impl<P: Value, const F: u32, const O: bool> From<P> for Record<P, F, O> {

datadog-profiling-protobuf/src/lib.rs

@@ -9,7 +9,7 @@ use std::io::{self, Write};
 /// field is not omitted for size/CPU reasons.
 #[repr(C)]
 #[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Location {
     /// Unique nonzero id for the location. A profile could use instruction
     /// addresses or any integer sequence as ids.
@@ -27,10 +27,10 @@ pub struct Location {
     pub line: Record<Line, 4, OPT_ZERO>,
 }
 
-/// Represents function and line number information. Omits column.  
+/// Represents function and line number information. Omits column.
 #[repr(C)]
 #[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Line {
     /// The id of the corresponding profile.Function for this line.
     pub function_id: Record<u64, 1, OPT_ZERO>,

datadog-profiling-protobuf/src/location.rs

@@ -8,7 +8,7 @@ use std::io::{self, Write};
 /// save bytes/CPU since they are unused in Datadog.
 #[repr(C)]
 #[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct Mapping {
     /// Unique nonzero id for the mapping.
     pub id: Record<u64, 1, NO_OPT_ZERO>,

datadog-profiling-protobuf/src/mapping.rs

@@ -1,7 +1,9 @@
 // Copyright 2023-Present Datadog, Inc. https://www.datadoghq.com/
 // SPDX-License-Identifier: Apache-2.0
 
-#[derive(Eq, Hash, PartialEq, ::prost::Message)]
+pub use ::prost::Message;
+
+#[derive(Eq, Hash, PartialEq, Message)]
 pub struct Profile {
     #[prost(message, repeated, tag = "1")]
     pub sample_types: Vec<ValueType>,
@@ -33,7 +35,7 @@ pub struct Profile {
     pub default_sample_type: i64,
 }
 
-#[derive(Clone, Eq, Hash, PartialEq, PartialOrd, Ord, ::prost::Message)]
+#[derive(Clone, Eq, Hash, PartialEq, PartialOrd, Ord, Message)]
 pub struct Sample {
     #[prost(uint64, repeated, tag = "1")]
     pub location_ids: Vec<u64>,
@@ -43,15 +45,15 @@ pub struct Sample {
     pub labels: Vec<Label>,
 }
 
-#[derive(Copy, Clone, Eq, PartialEq, Hash, ::prost::Message)]
+#[derive(Copy, Clone, Eq, PartialEq, Hash, Message)]
 pub struct ValueType {
     #[prost(int64, tag = "1")]
     pub r#type: i64, // Index into string table
     #[prost(int64, tag = "2")]
     pub unit: i64, // Index into string table
 }
 
-#[derive(Clone, Eq, PartialEq, Hash, PartialOrd, Ord, ::prost::Message)]
+#[derive(Clone, Eq, PartialEq, Hash, PartialOrd, Ord, Message)]
 pub struct Label {
     #[prost(int64, tag = "1")]
     pub key: i64, // Index into string table
@@ -63,7 +65,7 @@ pub struct Label {
     pub num_unit: i64,
 }
 
-#[derive(Copy, Clone, Eq, PartialEq, Hash, ::prost::Message)]
+#[derive(Copy, Clone, Eq, PartialEq, Hash, Message)]
 pub struct Mapping {
     #[prost(uint64, tag = "1")]
     pub id: u64,
@@ -87,7 +89,7 @@ pub struct Mapping {
     pub has_inline_frames: bool,
 }
 
-#[derive(Clone, Eq, PartialEq, Hash, ::prost::Message)]
+#[derive(Clone, Eq, PartialEq, Hash, Message)]
 pub struct Location {
     #[prost(uint64, tag = "1")]
     pub id: u64,
@@ -101,15 +103,15 @@ pub struct Location {
     pub is_folded: bool,
 }
 
-#[derive(Copy, Clone, Eq, PartialEq, Hash, ::prost::Message)]
+#[derive(Copy, Clone, Eq, PartialEq, Hash, Message)]
 pub struct Line {
     #[prost(uint64, tag = "1")]
     pub function_id: u64,
     #[prost(int64, tag = "2")]
     pub line: i64,
 }
 
-#[derive(Copy, Clone, Eq, PartialEq, Hash, ::prost::Message)]
+#[derive(Copy, Clone, Eq, PartialEq, Hash, Message)]
 pub struct Function {
     #[prost(uint64, tag = "1")]
     pub id: u64,

datadog-profiling-protobuf/src/prost_impls.rs

@@ -2,6 +2,7 @@
 // SPDX-License-Identifier: Apache-2.0
 
 use crate::{varint, Value, WireType};
+use std::fmt;
 use std::io::{self, Write};
 
 unsafe impl Value for &str {
@@ -16,18 +17,26 @@ unsafe impl Value for &str {
     }
 }
 
+// todo: for OTEL, needs to be i32::MAX rather than u32::MAX.
 /// Represents an offset into the Profile's string table. Note that it cannot
 /// exceed u32 because an entire protobuf message must not be larger than or
 /// equal to 2 GiB. By the time you encode the tag and length prefix for each
 /// string, there's no way to get this many unique-ish strings without first
 /// exceeding the protobuf 2 GiB limit.
 ///
 /// A value of 0 means "no string" or "empty string" (they are synonymous).
+/// cbindgen:field-names=[offset]
 #[repr(C)]
-#[derive(Copy, Clone, Default, Debug, Eq, PartialEq, PartialOrd, Ord, Hash)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[derive(Clone, Copy, Debug, Default, Eq, Hash, Ord, PartialEq, PartialOrd)]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct StringOffset(u32);
 
+impl fmt::Display for StringOffset {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        self.0.fmt(f)
+    }
+}
+
 /// # Safety
 /// The Default implementation will return all zero-representations.
 unsafe impl Value for StringOffset {

datadog-profiling-protobuf/src/string.rs

@@ -6,8 +6,8 @@ use std::io::{self, Write};
 
 /// ValueType describes the semantics and measurement units of a value.
 #[repr(C)]
-#[derive(Copy, Clone, Debug, Default, Eq, PartialEq)]
-#[cfg_attr(test, derive(bolero::generator::TypeGenerator))]
+#[derive(Copy, Clone, Debug, Default, Eq, Hash, PartialEq)]
+#[cfg_attr(feature = "bolero", derive(bolero::generator::TypeGenerator))]
 pub struct ValueType {
     pub r#type: Record<StringOffset, 1, OPT_ZERO>,
     pub unit: Record<StringOffset, 2, OPT_ZERO>,