feat: Support merging OCR'ed texts

Useful when characters on the screen split the number into separate
readings.

Author: dcelasun

README.md

@@ -123,6 +123,7 @@ All options can be set via CLI flags or environment variables. Flags take preced
 | `--crop` | `CROP` | *(disabled)* | Crop rectangle as `x0,y0,x1,y1` applied before OCR |
 | `--ocr-match-regex` | `OCR_MATCH_REGEX` | `^000\d+$` | Regex to identify the meter reading from OCR text |
 | `--ocr-fix-regex` | `OCR_FIX_REGEX` | *(disabled)* | Comma-separated list of regex substitutions as `pattern=replacement` applied in order before matching (e.g. `^O=0,^030=000`) |
+| `--ocr-merge-texts` | `OCR_MERGE_TEXTS` | `false` | Concatenate all OCR text results into a single string before applying fix/match regexes (useful when readings are split across multiple detections, e.g. `["00036", "128"]` → `"00036128"`) |
 | `--ocr-mask-regions` | `OCR_MASK_REGIONS` | *(disabled)* | Comma-separated rectangle coordinates to mask before OCR, as `x1,y1,x2,y2[,x3,y3,x4,y4,...]` (applied after crop) |
 | `--ocr-mask-colors` | `OCR_MASK_COLORS` | `000000` | Comma-separated hex colors for mask regions. One color applies to all; otherwise must match the number of regions |
 | `--mqtt-broker` | `MQTT_BROKER` | *(disabled)* | MQTT broker URL, e.g. `tcp://192.168.1.100:1883` |

http.go

@@ -111,7 +111,7 @@ func processOCR(imageData []byte, batLevel, batVoltage int) {
 		return
 	}
 
-	reading := extractReading(ocrOut.Texts, ocrMatchRe, ocrFixRules)
+	reading := extractReading(ocrOut.Texts, ocrMatchRe, ocrFixRules, ocrMergeTexts)
 
 	// Append reading to CSV unconditionally so discarded values are still on disk.
 	storeReading(imagePath, reading)

main.go

@@ -31,6 +31,7 @@ var (
 	mqttDeviceModel        string
 	meterDivisor           float64
 	ocrIncrOnly            bool
+	ocrMergeTexts          bool
 )
 
 func main() {
@@ -77,6 +78,11 @@ func main() {
 				Usage:   "Comma-separated list of regex substitutions applied to OCR text before matching, each as pattern=replacement (e.g. ^O=0,^030=000)",
 				Sources: cli.EnvVars("OCR_FIX_REGEX"),
 			},
+			&cli.BoolFlag{
+				Name:    "ocr-merge-texts",
+				Usage:   "Concatenate all OCR text results into a single string before applying fix/match regexes (useful when readings are split across multiple detections)",
+				Sources: cli.EnvVars("OCR_MERGE_TEXTS"),
+			},
 			&cli.StringFlag{
 				Name:    "ocr-mask-regions",
 				Usage:   "Comma-separated rectangle coordinates to mask before OCR, as x1,y1,x2,y2[,x3,y3,x4,y4,...] (applied after crop)",
@@ -165,6 +171,7 @@ func run(_ context.Context, cmd *cli.Command) error {
 	mqttDeviceModel = cmd.String("mqtt-device-model")
 	meterDivisor = cmd.Float("meter-divisor")
 	ocrIncrOnly = cmd.Bool("ocr-incr-only")
+	ocrMergeTexts = cmd.Bool("ocr-merge-texts")
 
 	if mqttBroker != "" {
 		initMQTT()

ocr.go

@@ -239,7 +239,14 @@ func applyFixRules(s string, rules []ocrFixRule) string {
 	return s
 }
 
-func extractReading(texts []string, matchRe *regexp.Regexp, fixRules []ocrFixRule) string {
+func extractReading(texts []string, matchRe *regexp.Regexp, fixRules []ocrFixRule, mergeTexts bool) string {
+	if mergeTexts {
+		var merged strings.Builder
+		for _, t := range texts {
+			merged.WriteString(strings.TrimSpace(t))
+		}
+		texts = []string{merged.String()}
+	}
 	// First pass: apply fix rules, then match.
 	for _, t := range texts {
 		t = applyFixRules(strings.TrimSpace(t), fixRules)

ocr_test.go

@@ -127,14 +127,76 @@ func TestExtractReading(t *testing.T) {
 
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			got := extractReading(tt.texts, tt.matchRe, tt.fixRules)
+			got := extractReading(tt.texts, tt.matchRe, tt.fixRules, false)
 			if got != tt.want {
 				t.Errorf("extractReading() = %q, want %q", got, tt.want)
 			}
 		})
 	}
 }
 
+func TestExtractReadingMerge(t *testing.T) {
+	defaultMatchRe := regexp.MustCompile(`^000\d+$`)
+
+	tests := []struct {
+		name     string
+		texts    []string
+		matchRe  *regexp.Regexp
+		fixRules []ocrFixRule
+		want     string
+	}{
+		{
+			name:    "merge splits into single reading",
+			texts:   []string{"00036", "128"},
+			matchRe: regexp.MustCompile(`^\d+$`),
+			want:    "00036128",
+		},
+		{
+			name:    "merge with match regex",
+			texts:   []string{"000", "354225"},
+			matchRe: defaultMatchRe,
+			want:    "000354225",
+		},
+		{
+			name:    "merge with fix rules",
+			texts:   []string{"O30", "354225"},
+			matchRe: defaultMatchRe,
+			fixRules: []ocrFixRule{
+				{Pattern: regexp.MustCompile(`^O`), Replacement: "0"},
+				{Pattern: regexp.MustCompile(`^030`), Replacement: "000"},
+			},
+			want: "000354225",
+		},
+		{
+			name:    "merge single element unchanged",
+			texts:   []string{"000354225"},
+			matchRe: defaultMatchRe,
+			want:    "000354225",
+		},
+		{
+			name:    "merge trims whitespace from each part",
+			texts:   []string{" 000 ", " 354225 "},
+			matchRe: defaultMatchRe,
+			want:    "000354225",
+		},
+		{
+			name:    "merge empty input returns empty",
+			texts:   []string{},
+			matchRe: defaultMatchRe,
+			want:    "",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := extractReading(tt.texts, tt.matchRe, tt.fixRules, true)
+			if got != tt.want {
+				t.Errorf("extractReading(merge=true) = %q, want %q", got, tt.want)
+			}
+		})
+	}
+}
+
 func TestParseOCRFixRules(t *testing.T) {
 	tests := []struct {
 		name      string