MINOR: add identifier_scope configuration option

Introduce the `identifier_scope` setting to control where code identifiers
are collected from to exclude as allowed words. The available scopes are:
- `diff`: only from the diff content (added lines)
- `files`: from the full content of changed files (new default)
- `all`: from every file in the repository

Previously, identifiers were only collected from the diff content. By
defaulting to the full content of changed files, this improves spell
checking accuracy by ensuring valid code identifiers are properly
ignored even if they are not directly modified in the diff.

Author: oktalz

.aspell.yml

@@ -6,9 +6,15 @@ mode: all
 ## Minimum word length to spell-check (words shorter than this are skipped, default: 3)
 # min_length: 3
 
-## If true, code identifiers found in the diff will NOT be excluded from spell check
+## If true, code identifiers will NOT be excluded from spell check
 # no_ignore_identifiers: false
 
+## Where to collect code identifiers from to exclude as allowed words (default: files)
+##   diff  - only from the diff content (added lines)
+##   files - from the full content of changed files
+##   all   - from every file in the repository
+# identifier_scope: diff | files | all
+
 ## File patterns to skip when mode is "all" (glob syntax)
 ignore_files:
   - '*test.go'

aspell/aspell.go

@@ -4,7 +4,9 @@ import (
 	"bytes"
 	"fmt"
 	"log"
+	"os"
 	"os/exec"
+	"path/filepath"
 	"regexp"
 	"slices"
 	"strings"
@@ -25,9 +27,10 @@ type RemoteFile struct {
 
 type Aspell struct {
 	RemoteFile          RemoteFile       `yaml:"remote_file"`
-	Dictionaries        DictionaryConfig `yaml:"dictionaries"`
 	Mode                mode             `yaml:"mode"`
 	HelpText            string           `yaml:"-"`
+	IdentifierScope     identifierScope  `yaml:"identifier_scope"`
+	Dictionaries        DictionaryConfig `yaml:"dictionaries"`
 	IgnoreFiles         []string         `yaml:"ignore_files"`
 	AllowedWords        []string         `yaml:"allowed"`
 	ExtraDicts          []string         `yaml:"-"` // paths to downloaded .rws files for aspell --extra-dicts
@@ -169,20 +172,63 @@ func (a Aspell) collectIdentifiers(content []map[string]string) []string {
 	if a.NoIgnoreIdentifiers {
 		return nil
 	}
-	var identifierWords []string
 	seen := map[string]struct{}{}
-	for _, file := range content {
-		for name, v := range file {
-			for _, word := range match.GetIdentifiersFromContent(name, v) {
-				if _, ok := seen[word]; !ok {
-					seen[word] = struct{}{}
-					identifierWords = append(identifierWords, word)
+	addWords := func(name, data string) {
+		for _, word := range match.GetIdentifiersFromContent(name, data) {
+			if _, ok := seen[word]; !ok {
+				seen[word] = struct{}{}
+			}
+		}
+	}
+
+	switch a.IdentifierScope {
+	case identifierScopeDiff:
+		for _, file := range content {
+			for name, v := range file {
+				addWords(name, v)
+			}
+		}
+	case identifierScopeFiles, "":
+		// Read full file content for each changed file
+		for _, file := range content {
+			for name := range file {
+				data, err := os.ReadFile(name)
+				if err != nil {
+					log.Printf("aspell: could not read file %s for identifiers, using diff: %v", name, err)
+					addWords(name, file[name])
+					continue
 				}
+				addWords(name, string(data))
 			}
 		}
+	case identifierScopeAll:
+		// Collect from all files in the repo
+		_ = filepath.WalkDir(".", func(path string, d os.DirEntry, err error) error {
+			if err != nil {
+				return nil
+			}
+			if d.IsDir() {
+				base := d.Name()
+				if base == ".git" || base == "vendor" || base == "node_modules" {
+					return filepath.SkipDir
+				}
+				return nil
+			}
+			data, err := os.ReadFile(path)
+			if err != nil {
+				return nil
+			}
+			addWords(path, string(data))
+			return nil
+		})
+	}
+
+	identifierWords := make([]string, 0, len(seen))
+	for w := range seen {
+		identifierWords = append(identifierWords, w)
 	}
 	if len(identifierWords) > 0 {
-		log.Printf("collected %d identifiers from diff content for spell check filtering", len(identifierWords))
+		log.Printf("collected %d identifiers (scope: %s) for spell check filtering", len(identifierWords), a.IdentifierScope)
 	}
 	return identifierWords
 }

aspell/fetch_dictionaries.go

@@ -285,10 +285,10 @@ func fetchSingleFile(url, name, tokenEnv string) (fetchedDictionaries, error) {
 			return result, fmt.Errorf("creating temp file for %s: %w", name, err)
 		}
 		if _, err := tmpFile.Write(data); err != nil {
-			tmpFile.Close()
+			_ = tmpFile.Close()
 			return result, fmt.Errorf("writing temp file for %s: %w", name, err)
 		}
-		tmpFile.Close()
+		_ = tmpFile.Close()
 		result.rwsFiles = append(result.rwsFiles, tmpFile.Name())
 		log.Printf("aspell dictionaries: saved .rws dictionary %s to %s", name, tmpFile.Name())
 	case strings.HasSuffix(lower, ".txt"):

aspell/mode.go

@@ -8,3 +8,11 @@ const (
 	modeCommit   mode = "commit"
 	modeAll      mode = "all"
 )
+
+type identifierScope string
+
+const (
+	identifierScopeDiff  identifierScope = "diff"
+	identifierScopeFiles identifierScope = "files"
+	identifierScopeAll   identifierScope = "all"
+)

aspell/new.go

@@ -66,6 +66,14 @@ func New(filename string) (Aspell, error) {
 		return Aspell{}, fmt.Errorf("invalid mode: %s", aspell.Mode)
 	}
 
+	switch aspell.IdentifierScope {
+	case identifierScopeDiff, identifierScopeFiles, identifierScopeAll:
+	case "":
+		aspell.IdentifierScope = identifierScopeFiles
+	default:
+		return Aspell{}, fmt.Errorf("invalid identifier_scope: %s", aspell.IdentifierScope)
+	}
+
 	log.Printf("aspell mode set to %s", aspell.Mode)
 	if fileExists {
 		aspell.HelpText = `aspell can be configured with .aspell.yml file.