@xuda.io/drive_module 1.1.953 → 1.1.954
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/index.js +14 -0
- package/package.json +1 -1
package/index.js
CHANGED
|
@@ -1727,6 +1727,13 @@ const ocr_drive_file = async function (app_id, doc) {
|
|
|
1727
1727
|
|
|
1728
1728
|
switch (doc.mime.split("/")[0]) {
|
|
1729
1729
|
case "image": {
|
|
1730
|
+
const initialText = await tesseract.recognize(
|
|
1731
|
+
target_file,
|
|
1732
|
+
generalConfig
|
|
1733
|
+
);
|
|
1734
|
+
|
|
1735
|
+
// Detect the language from the extracted text
|
|
1736
|
+
const detectedLang = franc(initialText);
|
|
1730
1737
|
ocr = await tesseract.recognize(target_file, generalConfig);
|
|
1731
1738
|
|
|
1732
1739
|
break;
|
|
@@ -1739,6 +1746,13 @@ const ocr_drive_file = async function (app_id, doc) {
|
|
|
1739
1746
|
|
|
1740
1747
|
const images = await convertPDF(target_file);
|
|
1741
1748
|
for await (const imageData of images) {
|
|
1749
|
+
const initialText = await tesseract.recognize(
|
|
1750
|
+
target_file,
|
|
1751
|
+
generalConfig
|
|
1752
|
+
);
|
|
1753
|
+
|
|
1754
|
+
// Detect the language from the extracted text
|
|
1755
|
+
const detectedLang = franc(initialText);
|
|
1742
1756
|
ocr += await tesseract.recognize(imageData.path, generalConfig);
|
|
1743
1757
|
}
|
|
1744
1758
|
break;
|