Improve lookup table pattern use in profiles. (#592)

Based on implementation experience with the draft version of the profiles spec, the use of lookup tables for storing values is space efficient but not particularly user friendly. This change makes improvements by changing field names to clarify their role in the pattern and changing their data types to better accommodate their use in languages limited to 32 bit array indexing e.g. Java.

opentelemetry/proto/profiles/v1development/profiles.proto

@@ -139,7 +139,7 @@ message ScopeProfiles {
   // A list of Profiles that originate from an instrumentation scope.
   repeated Profile profiles = 2;
 
-  // The Schema URL, if known. This is the identifier of the Schema that the metric data
+  // The Schema URL, if known. This is the identifier of the Schema that the profile data
   // is recorded in. To learn more about Schema URL see
   // https://opentelemetry.io/docs/specs/otel/schemas/#schema-url
   // This schema_url applies to all profiles in the "profiles" field.
@@ -194,13 +194,13 @@ message Profile {
   // into that address range.  mapping[0] will be the main binary.
   // If multiple binaries contribute to the Profile and no main
   // binary can be identified, mapping[0] has no special meaning.
-  repeated Mapping mapping = 3;
+  repeated Mapping mapping_table = 3;
   // Locations referenced by samples via location_indices.
-  repeated Location location = 4;
+  repeated Location location_table = 4;
   // Array of locations referenced by samples.
-  repeated int64 location_indices = 15;
+  repeated int32 location_indices = 15;
   // Functions referenced by locations.
-  repeated Function function = 5;
+  repeated Function function_table = 5;
   // Lookup table for attributes.
   repeated opentelemetry.proto.common.v1.KeyValue attribute_table = 16;
   // Represents a mapping between Attribute Keys and Units.
@@ -212,10 +212,10 @@ message Profile {
   repeated string string_table = 6;
   // frames with Function.function_name fully matching the following
   // regexp will be dropped from the samples, along with their successors.
-  int64 drop_frames = 7;   // Index into string table.
+  int32 drop_frames_strindex = 7;   // Index into string table.
   // frames with Function.function_name fully matching the following
   // regexp will be kept, even if it matches drop_frames.
-  int64 keep_frames = 8;  // Index into string table.
+  int32 keep_frames_strindex = 8;  // Index into string table.
 
   // The following fields 9-14 are informational, do not affect
   // interpretation of results.
@@ -234,10 +234,10 @@ message Profile {
   // should not be used to store any machine-readable information, it is only
   // for human-friendly content. The profile must stay functional if this field
   // is cleaned.
-  repeated int64 comment = 13; // Indices into string table.
+  repeated int32 comment_strindices = 13; // Indices into string table.
   // Index into the string table of the type of the preferred sample
   // value. If unset, clients should default to the last sample value.
-  int64 default_sample_type = 14;
+  int32 default_sample_type_strindex = 14;
 
 
   // A globally unique identifier for a profile. The ID is a 16-byte array. An ID with
@@ -280,9 +280,9 @@ message Profile {
 // Represents a mapping between Attribute Keys and Units.
 message AttributeUnit {
   // Index into string table.
-  int64 attribute_key = 1;
+  int32 attribute_key_strindex = 1;
   // Index into string table.
-  int64 unit = 2;
+  int32 unit_strindex = 2;
 }
 
 // A pointer from a profile Sample to a trace Span.
@@ -366,8 +366,8 @@ enum AggregationTemporality {
 
 // ValueType describes the type and units of a value, with an optional aggregation temporality.
 message ValueType {
-  int64 type = 1; // Index into string table.
-  int64 unit = 2; // Index into string table.
+  int32 type_strindex = 1; // Index into string table.
+  int32 unit_strindex = 2; // Index into string table.
 
   AggregationTemporality aggregation_temporality = 3;
 }
@@ -377,15 +377,15 @@ message ValueType {
 // augmented with auxiliary information like the thread-id, some
 // indicator of a higher level request being handled etc.
 message Sample {
-  // The indices recorded here correspond to locations in Profile.location.
+  // References to locations in Profile.location_table.
   // The leaf is at location_index[0]. [deprecated, superseded by locations_start_index / locations_length]
-  repeated uint64 location_index = 1;
-  // locations_start_index along with locations_length refers to to a slice of locations in Profile.location.
+  repeated int32 location_index = 1;
+  // locations_start_index along with locations_length refers to to a slice of locations in Profile.location_indices.
   // Supersedes location_index.
-  uint64 locations_start_index = 7;
-  // locations_length along with locations_start_index refers to a slice of locations in Profile.location.
+  int32 locations_start_index = 7;
+  // locations_length along with locations_start_index refers to a slice of locations in Profile.location_indices.
   // Supersedes location_index.
-  uint64 locations_length = 8;
+  int32 locations_length = 8;
   // The type and unit of each value is defined by the corresponding
   // entry in Profile.sample_type. All samples must have the same
   // number of values, the same as the length of Profile.sample_type.
@@ -394,10 +394,10 @@ message Sample {
   // lists of the originals.
   repeated int64 value = 2;
   // References to attributes in Profile.attribute_table. [optional]
-  repeated uint64 attributes = 10;
+  repeated int32 attribute_indices = 10;
 
   // Reference to link in Profile.link_table. [optional]
-  uint64 link = 12;
+  int32 link_index = 12;
 
   // Timestamps associated with Sample represented in nanoseconds. These timestamps are expected
   // to fall within the Profile's time range. [optional]
@@ -407,10 +407,10 @@ message Sample {
 // Provides additional context for a sample,
 // such as thread ID or allocation size, with optional units. [deprecated]
 message Label {
-  int64 key = 1;   // Index into string table
+  int32 key_strindex = 1;   // Index into string table
 
   // At most one of the following must be present
-  int64 str = 2;   // Index into string table
+  int32 str_strindex = 2;   // Index into string table
   int64 num = 3;
 
   // Should only be present when num is present.
@@ -420,7 +420,7 @@ message Label {
   // Consumers may also  interpret units like "bytes" and "kilobytes" as memory
   // units and units like "seconds" and "nanoseconds" as time units,
   // and apply appropriate unit conversions to these.
-  int64 num_unit = 4;  // Index into string table
+  int32 num_unit_strindex = 4;  // Index into string table
 }
 
 // Describes the mapping of a binary in memory, including its address range,
@@ -437,9 +437,9 @@ message Mapping {
   // The object this entry is loaded from.  This can be a filename on
   // disk for the main binary and shared libraries, or virtual
   // abstractions like "[vdso]".
-  int64 filename = 5;  // Index into string table
+  int32 filename_strindex = 5;  // Index into string table
   // References to attributes in Profile.attribute_table. [optional]
-  repeated uint64 attributes = 12;
+  repeated int32 attribute_indices = 12;
   // The following fields indicate the resolution of symbolic info.
   bool has_functions = 7;
   bool has_filenames = 8;
@@ -452,10 +452,10 @@ message Location {
   // Unique nonzero id for the location.  A profile could use
   // instruction addresses or any integer sequence as ids. [deprecated]
   uint64 id = 1;
-  // The index of the corresponding profile.Mapping for this location.
+  // Reference to mapping in Profile.mapping_table.
   // It can be unset if the mapping is unknown or not applicable for
   // this profile type.
-  uint64 mapping_index = 2;
+  int32 mapping_index = 2;
   // The instruction address for this location, if available.  It
   // should be within [Mapping.memory_start...Mapping.memory_limit]
   // for the corresponding mapping. A non-leaf address may be in the
@@ -478,13 +478,13 @@ message Location {
   bool is_folded = 5;
 
   // References to attributes in Profile.attribute_table. [optional]
-  repeated uint64 attributes = 7;
+  repeated int32 attribute_indices = 7;
 }
 
 // Details a specific line in a source code, linked to a function.
 message Line {
-  // The index of the corresponding profile.Function for this line.
-  uint64 function_index = 1;
+  // Reference to function in Profile.function_table.
+  int32 function_index = 1;
   // Line number in source code.
   int64 line = 2;
   // Column number in source code.
@@ -497,12 +497,12 @@ message Function {
   // Unique nonzero id for the function. [deprecated]
   uint64 id = 1;
   // Name of the function, in human-readable form if available.
-  int64 name = 2; // Index into string table
+  int32 name_strindex = 2; // Index into string table
   // Name of the function, as identified by the system.
   // For instance, it can be a C++ mangled name.
-  int64 system_name = 3; // Index into string table
+  int32 system_name_strindex = 3; // Index into string table
   // Source file containing the function.
-  int64 filename = 4; // Index into string table
+  int32 filename_strindex = 4; // Index into string table
   // Line number in source file.
   int64 start_line = 5;
 }