self-hosted/joplin
1
0
Fork 0
mirror of https://github.com/laurent22/joplin.git synced 2025-11-26 22:41:17 +02:00
Code Issues Releases Activity

Merge branch 'release-2.14' into dev

Browse Source
This commit is contained in:
Laurent Cozic
2024-03-18 10:17:39 +00:00
parent 073df50244 971c4e5e84
commit fd4d7ead43
7 changed files with 105 additions and 66 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
packages/lib/shim-init-node.ts
View File

@@ -755,21 +755,24 @@ function shimInit(options: ShimInitOptions = null) {
shim.pdfExtractEmbeddedText = async (pdfPath: string): Promise<string[]> => {
const loadingTask = pdfJs.getDocument(pdfPath);
const doc = await loadingTask.promise;
const textByPage = [];
for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
const page = await doc.getPage(pageNum);
const textContent = await page.getTextContent();
try {
for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
const page = await doc.getPage(pageNum);
const textContent = await page.getTextContent();
const strings = textContent.items.map(item => {
const text = (item as TextItem).str ?? '';
return text;
}).join('\n');
const strings = textContent.items.map(item => {
const text = (item as TextItem).str ?? '';
return text;
}).join('\n');
// Some PDFs contain unsupported characters that can lead to hard-to-debug issues.
// We remove them here.
textByPage.push(replaceUnsupportedCharacters(strings));
// Some PDFs contain unsupported characters that can lead to hard-to-debug issues.
// We remove them here.
textByPage.push(replaceUnsupportedCharacters(strings));
}
} finally {
await doc.destroy();
}
return textByPage;
@@ -807,23 +810,27 @@ function shimInit(options: ShimInitOptions = null) {
const loadingTask = pdfJs.getDocument(pdfPath);
const doc = await loadingTask.promise;
for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
const page = await doc.getPage(pageNum);
const viewport = page.getViewport({ scale: 2 });
const canvas = createCanvas();
const ctx = canvas.getContext('2d');
try {
for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
const page = await doc.getPage(pageNum);
const viewport = page.getViewport({ scale: 2 });
const canvas = createCanvas();
const ctx = canvas.getContext('2d');
canvas.height = viewport.height;
canvas.width = viewport.width;
canvas.height = viewport.height;
canvas.width = viewport.width;
const renderTask = page.render({ canvasContext: ctx, viewport: viewport });
await renderTask.promise;
const renderTask = page.render({ canvasContext: ctx, viewport: viewport });
await renderTask.promise;
const buffer = await canvasToBuffer(canvas);
const filePath = `${outputDirectoryPath}/${filePrefix}_${pageNum.toString().padStart(4, '0')}.jpg`;
output.push(filePath);
await writeFile(filePath, buffer, 'binary');
if (!(await shim.fsDriver().exists(filePath))) throw new Error(`Could not write to file: ${filePath}`);
const buffer = await canvasToBuffer(canvas);
const filePath = `${outputDirectoryPath}/${filePrefix}_${pageNum.toString().padStart(4, '0')}.jpg`;
output.push(filePath);
await writeFile(filePath, buffer, 'binary');
if (!(await shim.fsDriver().exists(filePath))) throw new Error(`Could not write to file: ${filePath}`);
}
} finally {
await doc.destroy();
}
return output;