[pkg/stanza] Consolidate attribute code into single package (open-telemetry#30449)

This change simplifies the reader package by pulling all file resolution
logic into a single package. It also moves related config and attribute
name constants into the same package.

Author: djaglowski

.chloggen/pkg-stanza-reader-attributes-refactor.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: deprecation
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: pkg/stanza
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Deprecate pkg/stanza/attrs package in favor of pkg/stanza/fileconsumer/attrs
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [30449]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [api]

pkg/stanza/attrs/attrs.go

@@ -3,9 +3,22 @@
 
 package attrs // import "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/attrs"
 
+import fca "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/attrs"
+
 const (
-	LogFileName         = "log.file.name"
-	LogFilePath         = "log.file.path"
-	LogFileNameResolved = "log.file.name_resolved"
-	LogFilePathResolved = "log.file.path_resolved"
+	// Deprecated: [v0.93.0] Use pkg/stanza/fileconsumer/attrs.LogFileName instead.
+	// Will be removed in v0.94.0.
+	LogFileName = fca.LogFileName
+
+	// Deprecated: [v0.92.0] Use pkg/stanza/fileconsumer/attrs.LogFilePath instead.
+	// Will be removed in v0.94.0.
+	LogFilePath = fca.LogFilePath
+
+	// Deprecated: [v0.92.0] Use pkg/stanza/fileconsumer/attrs.LogFileNameResolved instead.
+	// Will be removed in v0.94.0.
+	LogFileNameResolved = fca.LogFileNameResolved
+
+	// Deprecated: [v0.92.0] Use pkg/stanza/fileconsumer/attrs.LogFilePathResolved instead.
+	// Will be removed in v0.94.0.
+	LogFilePathResolved = fca.LogFilePathResolved
 )

pkg/stanza/fileconsumer/attrs/attrs.go

@@ -0,0 +1,60 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package attrs // import "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/attrs"
+
+import (
+	"fmt"
+	"path/filepath"
+	"runtime"
+)
+
+const (
+	LogFileName         = "log.file.name"
+	LogFilePath         = "log.file.path"
+	LogFileNameResolved = "log.file.name_resolved"
+	LogFilePathResolved = "log.file.path_resolved"
+)
+
+type Resolver struct {
+	IncludeFileName         bool `mapstructure:"include_file_name,omitempty"`
+	IncludeFilePath         bool `mapstructure:"include_file_path,omitempty"`
+	IncludeFileNameResolved bool `mapstructure:"include_file_name_resolved,omitempty"`
+	IncludeFilePathResolved bool `mapstructure:"include_file_path_resolved,omitempty"`
+}
+
+func (r *Resolver) Resolve(path string) (attributes map[string]any, err error) {
+	// size 2 is sufficient if not resolving symlinks. This optimizes for the most performant cases.
+	attributes = make(map[string]any, 2)
+	if r.IncludeFileName {
+		attributes[LogFileName] = filepath.Base(path)
+	}
+	if r.IncludeFilePath {
+		attributes[LogFilePath] = path
+	}
+	if !r.IncludeFileNameResolved && !r.IncludeFilePathResolved {
+		return attributes, nil
+	}
+
+	resolved := path
+	// Dirty solution, waiting for this permanent fix https://github.com/golang/go/issues/39786
+	// EvalSymlinks on windows is partially working depending on the way you use Symlinks and Junctions
+	if runtime.GOOS != "windows" {
+		resolved, err = filepath.EvalSymlinks(path)
+		if err != nil {
+			return nil, fmt.Errorf("resolve symlinks: %w", err)
+		}
+	}
+	abs, err := filepath.Abs(resolved)
+	if err != nil {
+		return nil, fmt.Errorf("resolve abs: %w", err)
+	}
+
+	if r.IncludeFileNameResolved {
+		attributes[LogFileNameResolved] = filepath.Base(abs)
+	}
+	if r.IncludeFilePathResolved {
+		attributes[LogFilePathResolved] = abs
+	}
+	return attributes, nil
+}

pkg/stanza/fileconsumer/attrs/attrs_test.go

@@ -0,0 +1,73 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package attrs
+
+import (
+	"fmt"
+	"path/filepath"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/filetest"
+)
+
+func TestResolver(t *testing.T) {
+	t.Parallel()
+
+	for i := 0; i < 16; i++ {
+
+		// Create a 4 bit string where each bit represents the value of a config option
+		bitString := fmt.Sprintf("%04b", i)
+
+		// Create a resolver with a config that matches the bit pattern of i
+		r := Resolver{
+			IncludeFileName:         bitString[0] == '1',
+			IncludeFilePath:         bitString[1] == '1',
+			IncludeFileNameResolved: bitString[2] == '1',
+			IncludeFilePathResolved: bitString[3] == '1',
+		}
+
+		t.Run(bitString, func(t *testing.T) {
+			// Create a file
+			tempDir := t.TempDir()
+			temp := filetest.OpenTemp(t, tempDir)
+
+			attributes, err := r.Resolve(temp.Name())
+			assert.NoError(t, err)
+
+			var expectLen int
+			if r.IncludeFileName {
+				expectLen++
+				assert.Equal(t, filepath.Base(temp.Name()), attributes[LogFileName])
+			} else {
+				assert.Empty(t, attributes[LogFileName])
+			}
+			if r.IncludeFilePath {
+				expectLen++
+				assert.Equal(t, temp.Name(), attributes[LogFilePath])
+			} else {
+				assert.Empty(t, attributes[LogFilePath])
+			}
+
+			// We don't have an independent way to resolve the path, so the only meangingful validate
+			// is to ensure that the resolver returns nothing vs something based on the config.
+			if r.IncludeFileNameResolved {
+				expectLen++
+				assert.NotNil(t, attributes[LogFileNameResolved])
+				assert.IsType(t, "", attributes[LogFileNameResolved])
+			} else {
+				assert.Empty(t, attributes[LogFileNameResolved])
+			}
+			if r.IncludeFilePathResolved {
+				expectLen++
+				assert.NotNil(t, attributes[LogFilePathResolved])
+				assert.IsType(t, "", attributes[LogFilePathResolved])
+			} else {
+				assert.Empty(t, attributes[LogFilePathResolved])
+			}
+			assert.Equal(t, expectLen, len(attributes))
+		})
+	}
+}

pkg/stanza/fileconsumer/config.go

@@ -14,6 +14,7 @@ import (
 	"golang.org/x/text/encoding"
 
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/decode"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/attrs"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/emit"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/fingerprint"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/header"
@@ -55,29 +56,28 @@ func NewConfig() *Config {
 		MaxLogSize:         reader.DefaultMaxLogSize,
 		Encoding:           defaultEncoding,
 		FlushPeriod:        reader.DefaultFlushPeriod,
-		IncludeFileName:    true,
+		Resolver: attrs.Resolver{
+			IncludeFileName: true,
+		},
 	}
 }
 
 // Config is the configuration of a file input operator
 type Config struct {
-	matcher.Criteria        `mapstructure:",squash"`
-	PollInterval            time.Duration   `mapstructure:"poll_interval,omitempty"`
-	MaxConcurrentFiles      int             `mapstructure:"max_concurrent_files,omitempty"`
-	MaxBatches              int             `mapstructure:"max_batches,omitempty"`
-	StartAt                 string          `mapstructure:"start_at,omitempty"`
-	FingerprintSize         helper.ByteSize `mapstructure:"fingerprint_size,omitempty"`
-	MaxLogSize              helper.ByteSize `mapstructure:"max_log_size,omitempty"`
-	Encoding                string          `mapstructure:"encoding,omitempty"`
-	SplitConfig             split.Config    `mapstructure:"multiline,omitempty"`
-	TrimConfig              trim.Config     `mapstructure:",squash,omitempty"`
-	FlushPeriod             time.Duration   `mapstructure:"force_flush_period,omitempty"`
-	IncludeFileName         bool            `mapstructure:"include_file_name,omitempty"`
-	IncludeFilePath         bool            `mapstructure:"include_file_path,omitempty"`
-	IncludeFileNameResolved bool            `mapstructure:"include_file_name_resolved,omitempty"`
-	IncludeFilePathResolved bool            `mapstructure:"include_file_path_resolved,omitempty"`
-	Header                  *HeaderConfig   `mapstructure:"header,omitempty"`
-	DeleteAfterRead         bool            `mapstructure:"delete_after_read,omitempty"`
+	matcher.Criteria   `mapstructure:",squash"`
+	attrs.Resolver     `mapstructure:",squash"`
+	PollInterval       time.Duration   `mapstructure:"poll_interval,omitempty"`
+	MaxConcurrentFiles int             `mapstructure:"max_concurrent_files,omitempty"`
+	MaxBatches         int             `mapstructure:"max_batches,omitempty"`
+	StartAt            string          `mapstructure:"start_at,omitempty"`
+	FingerprintSize    helper.ByteSize `mapstructure:"fingerprint_size,omitempty"`
+	MaxLogSize         helper.ByteSize `mapstructure:"max_log_size,omitempty"`
+	Encoding           string          `mapstructure:"encoding,omitempty"`
+	SplitConfig        split.Config    `mapstructure:"multiline,omitempty"`
+	TrimConfig         trim.Config     `mapstructure:",squash,omitempty"`
+	FlushPeriod        time.Duration   `mapstructure:"force_flush_period,omitempty"`
+	Header             *HeaderConfig   `mapstructure:"header,omitempty"`
+	DeleteAfterRead    bool            `mapstructure:"delete_after_read,omitempty"`
 }
 
 type HeaderConfig struct {
@@ -150,21 +150,18 @@ func (c Config) buildManager(logger *zap.SugaredLogger, emit emit.Callback, spli
 	}
 
 	readerFactory := reader.Factory{
-		SugaredLogger:           logger.With("component", "fileconsumer"),
-		FromBeginning:           startAtBeginning,
-		FingerprintSize:         int(c.FingerprintSize),
-		MaxLogSize:              int(c.MaxLogSize),
-		Encoding:                enc,
-		SplitFunc:               splitFunc,
-		TrimFunc:                trimFunc,
-		FlushTimeout:            c.FlushPeriod,
-		EmitFunc:                emit,
-		IncludeFileName:         c.IncludeFileName,
-		IncludeFilePath:         c.IncludeFilePath,
-		IncludeFileNameResolved: c.IncludeFileNameResolved,
-		IncludeFilePathResolved: c.IncludeFilePathResolved,
-		HeaderConfig:            hCfg,
-		DeleteAtEOF:             c.DeleteAfterRead,
+		SugaredLogger:   logger.With("component", "fileconsumer"),
+		FromBeginning:   startAtBeginning,
+		FingerprintSize: int(c.FingerprintSize),
+		MaxLogSize:      int(c.MaxLogSize),
+		Encoding:        enc,
+		SplitFunc:       splitFunc,
+		TrimFunc:        trimFunc,
+		FlushTimeout:    c.FlushPeriod,
+		EmitFunc:        emit,
+		Attributes:      c.Resolver,
+		HeaderConfig:    hCfg,
+		DeleteAtEOF:     c.DeleteAfterRead,
 	}
 
 	return &Manager{

pkg/stanza/fileconsumer/file_test.go

@@ -18,7 +18,7 @@ import (
 	"github.com/stretchr/testify/require"
 	"go.opentelemetry.io/collector/featuregate"
 
-	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/attrs"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/attrs"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/emittest"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/filetest"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/stanza/fileconsumer/internal/reader"