Merge pull request #89 from speedfl/feature/base64

Feature/base64

ocr_request.go

@@ -1,9 +1,11 @@
 package ocrworker
 
 import "fmt"
+import "encoding/base64"
 
 type OcrRequest struct {
 	ImgUrl            string                 `json:"img_url"`
+	ImgBase64         string                 `json:"img_base64"`
 	EngineType        OcrEngineType          `json:"engine"`
 	ImgBytes          []byte                 `json:"img_bytes"`
 	PreprocessorChain []string               `json:"preprocessors"`
@@ -26,7 +28,25 @@ func (ocrRequest *OcrRequest) nextPreprocessor(processorRoutingKey string) strin
 		ocrRequest.PreprocessorChain = s
 		return x
 	}
+}
+
+func (ocrRequest *OcrRequest) decodeBase64() error {
+
+	bytes, decodeError := base64.StdEncoding.DecodeString(ocrRequest.ImgBase64)
+
+	if decodeError != nil {
+		return decodeError
+	}
+
+	ocrRequest.ImgBytes = bytes
+	ocrRequest.ImgBase64 = ""
+
+	return nil
+}
+
+func (ocrRequest *OcrRequest) hasBase64() bool {
 
+	return ocrRequest.ImgBase64 != ""
 }
 
 func (ocrRequest *OcrRequest) downloadImgUrl() error {

ocr_rpc_client.go

@@ -4,9 +4,8 @@ import (
 	"encoding/json"
 	"fmt"
 	"time"
-
-	"github.com/couchbaselabs/logg"
 	"github.com/nu7hatch/gouuid"
+	"github.com/couchbaselabs/logg"
 	"github.com/streadway/amqp"
 )
 
@@ -88,11 +87,24 @@ func (c *OcrRpcClient) DecodeImage(ocrRequest OcrRequest) (OcrResult, error) {
 	// as open-ocr, it will be expensive in terms of bandwidth
 	// to have image binary in messages
 	if ocrRequest.ImgBytes == nil {
-		// if we already have image bytes, ignore image url
-		err = ocrRequest.downloadImgUrl()
-		if err != nil {
-			logg.LogTo("OCR_CLIENT", "Error downloading img url: %v", err)
-			return OcrResult{}, err
+
+		// if we do not have bytes use base 64 file by converting it to bytes
+		if ocrRequest.hasBase64() {
+
+			logg.LogTo("OCR_CLIENT", "OCR request has base 64 convert it to bytes")
+
+			err = ocrRequest.decodeBase64()
+			if err != nil {
+				logg.LogTo("OCR_CLIENT", "Error decoding base64: %v", err)
+				return OcrResult{}, err
+			}
+		} else {
+			// if we do not have base 64 or bytes download the file
+			err = ocrRequest.downloadImgUrl()
+			if err != nil {
+				logg.LogTo("OCR_CLIENT", "Error downloading img url: %v", err)
+				return OcrResult{}, err
+			}
 		}
 	}
 

tesseract_engine.go

@@ -1,6 +1,7 @@
 package ocrworker
 
 import (
+	"encoding/base64"
 	"fmt"
 	"io/ioutil"
 	"os"
@@ -97,7 +98,9 @@ func (t TesseractEngineArgs) Export() []string {
 func (t TesseractEngine) ProcessRequest(ocrRequest OcrRequest) (OcrResult, error) {
 
 	tmpFileName, err := func() (string, error) {
-		if ocrRequest.ImgUrl != "" {
+		if ocrRequest.ImgBase64 != "" {
+			return t.tmpFileFromImageBase64(ocrRequest.ImgBase64)
+		} else if ocrRequest.ImgUrl != "" {
 			return t.tmpFileFromImageUrl(ocrRequest.ImgUrl)
 		} else {
 			return t.tmpFileFromImageBytes(ocrRequest.ImgBytes)
@@ -126,6 +129,8 @@ func (t TesseractEngine) ProcessRequest(ocrRequest OcrRequest) (OcrResult, error
 
 func (t TesseractEngine) tmpFileFromImageBytes(imgBytes []byte) (string, error) {
 
+	logg.LogTo("OCR_TESSERACT", "Use tesseract with bytes image")
+
 	tmpFileName, err := createTempFileName()
 	if err != nil {
 		return "", err
@@ -142,8 +147,35 @@ func (t TesseractEngine) tmpFileFromImageBytes(imgBytes []byte) (string, error)
 
 }
 
+func (t TesseractEngine) tmpFileFromImageBase64(base64Image string) (string, error) {
+
+	logg.LogTo("OCR_TESSERACT", "Use tesseract with base 64")
+
+	tmpFileName, err := createTempFileName()
+	if err != nil {
+		return "", err
+	}
+
+	// decoding into bytes the base64 string
+	decoded, decodeError := base64.StdEncoding.DecodeString(base64Image)
+
+	if decodeError != nil {
+		return "", err
+	}
+
+	err = saveBytesToFileName(decoded, tmpFileName)
+	if err != nil {
+		return "", err
+	}
+
+	return tmpFileName, nil
+
+}
+
 func (t TesseractEngine) tmpFileFromImageUrl(imgUrl string) (string, error) {
 
+	logg.LogTo("OCR_TESSERACT", "Use tesseract with url")
+
 	tmpFileName, err := createTempFileName()
 	if err != nil {
 		return "", err