Skip to content

A fast, helpful, and open-source document parser

feat: screenshot and config tests

Clelia (Astra) Bertelli3d02db97d34f
diff --git a/src/core/config.test.ts b/src/core/config.test.ts
new file mode 100644
index 0000000..a4ae313
--- /dev/null
+++ b/src/core/config.test.ts
@@ -0,0 +1,22 @@
+import { describe, it, expect } from "vitest";
+import { LiteParseConfig } from "./types";
+import { DEFAULT_CONFIG, mergeConfig } from "./config";
+
+describe("test config", () => {
+ it("test merge config with no partial config", () => {
+ const user_config: Partial<LiteParseConfig> = {};
+ const config = mergeConfig(user_config);
+ expect(config).toStrictEqual(DEFAULT_CONFIG);
+ });
+ it("test merge config overrides defaults", () => {
+ const user_config: Partial<LiteParseConfig> = {
+ ocrLanguage: "fr",
+ ocrServerUrl: "http://localhost:8000",
+ dpi: 200,
+ };
+ const config = mergeConfig(user_config);
+ expect(config.ocrLanguage).toBe(user_config.ocrLanguage);
+ expect(config.ocrServerUrl).toBe(user_config.ocrServerUrl);
+ expect(config.dpi).toBe(user_config.dpi);
+ });
+});
diff --git a/src/core/parser.test.ts b/src/core/parser.test.ts
index 99c9262..56e8f7e 100644
--- a/src/core/parser.test.ts
+++ b/src/core/parser.test.ts
@@ -13,6 +13,7 @@ const {
mockTextItems,
mockParsedPagesWithBoundingBoxes,
mockParsedPagesWithBoundingBoxesOcr,
+ mockScreenshotResults,
} = vi.hoisted(() => {
const mockPdfConvertedResult = {
pdfPath: "/tmp/converted.pdf",
@@ -289,6 +290,40 @@ const {
],
};
+ const mockImageBuffer = Buffer.from(new Uint8Array([1, 2, 3, 4, 5]));
+ const mockScreenshotResults: ScreenshotResult[] = [
+ {
+ pageNum: 1,
+ width: 612,
+ height: 792,
+ imageBuffer: mockImageBuffer,
+ },
+ {
+ pageNum: 2,
+ width: 612,
+ height: 792,
+ imageBuffer: mockImageBuffer,
+ },
+ {
+ pageNum: 3,
+ width: 612,
+ height: 792,
+ imageBuffer: mockImageBuffer,
+ },
+ {
+ pageNum: 4,
+ width: 612,
+ height: 792,
+ imageBuffer: mockImageBuffer,
+ },
+ {
+ pageNum: 5,
+ width: 612,
+ height: 792,
+ imageBuffer: mockImageBuffer,
+ },
+ ];
+
return {
mockPdfConvertedResult,
mockPdfDocument,
@@ -302,11 +337,12 @@ const {
mockTextItems,
mockParsedPagesWithBoundingBoxes,
mockParsedPagesWithBoundingBoxesOcr,
+ mockScreenshotResults,
};
});
import { LiteParse } from "./parser";
-import { LiteParseConfig } from "./types";
+import { LiteParseConfig, ScreenshotResult } from "./types";
vi.mock("../conversion/convertToPdf.js", async () => {
const actual = await vi.importActual<typeof import("../conversion/convertToPdf.js")>(
@@ -332,6 +368,7 @@ vi.mock("../engines/pdf/pdfjs.js", async () => {
loadDocument = vi.fn().mockResolvedValue(mockPdfDocument);
extractAllPages = vi.fn().mockResolvedValue(mockPages);
+ extractPage = vi.fn().mockResolvedValue(mockPages[0]);
renderPageImage = vi.fn(async () => Buffer.from(new Uint8Array([1, 2, 3, 4, 5])));
close = vi.fn(async () => {});
}
@@ -407,6 +444,23 @@ vi.mock("../processing/bbox.js", async () => {
};
});
+vi.mock("../engines/pdf/pdfium-renderer.js", async () => {
+ const actual = await vi.importActual<typeof import("../engines/pdf/pdfium-renderer.js")>(
+ "../engines/pdf/pdfium-renderer.js"
+ );
+ return {
+ ...actual,
+ PdfiumRenderer: vi.fn(
+ class {
+ constructor() {}
+
+ renderPageToBuffer = vi.fn(async () => Buffer.from(new Uint8Array([1, 2, 3, 4, 5])));
+ close = vi.fn(async () => {});
+ }
+ ),
+ };
+});
+
describe("Parse tests", () => {
it("test parse without OCR and text format", async () => {
const config: Partial<LiteParseConfig> = { ocrEnabled: false, outputFormat: "text" };
@@ -539,3 +593,24 @@ describe("Parse tests", () => {
).toBe(0);
});
});
+
+describe("test screenshot", () => {
+ it("test screenshot with all pages", async () => {
+ const config: Partial<LiteParseConfig> = { ocrEnabled: false, outputFormat: "text" };
+ const liteparse = new LiteParse(config);
+ const results = await liteparse.screenshot("/tmp/test.docx");
+ expect(results.length).toBe(5);
+ expect(results).toStrictEqual(mockScreenshotResults);
+ });
+ it("test screenshot with selected pages", async () => {
+ const config: Partial<LiteParseConfig> = { ocrEnabled: false, outputFormat: "text" };
+ const liteparse = new LiteParse(config);
+ const results = await liteparse.screenshot("/tmp/test.docx", [1, 3, 4]);
+ expect(results.length).toBe(3);
+ expect(results).toStrictEqual([
+ mockScreenshotResults[0],
+ mockScreenshotResults[2],
+ mockScreenshotResults[3],
+ ]);
+ });
+});