| |
| import assert from 'node:assert/strict' |
| import { existsSync, mkdtempSync, readFileSync, rmSync } from 'node:fs' |
| import { tmpdir } from 'node:os' |
| import { join } from 'node:path' |
| import { after, test } from 'node:test' |
| import { textPdf } from './fixtures/textPdf.ts' |
| import { findDoi, pdfSeed, readPdf, savePdf, setUploadsDir } from './pdfSeed.ts' |
|
|
| const dir = mkdtempSync(join(tmpdir(), 'pdf-seed-')) |
| setUploadsDir(join(dir, 'uploads')) |
| after(() => rmSync(dir, { recursive: true, force: true })) |
|
|
| test('the first DOI in the text, without the punctuation around it', () => { |
| assert.equal(findDoi('Science 352, 189 (2016). doi: 10.1126/science.aad0501.'), '10.1126/science.aad0501') |
| assert.equal(findDoi('see https://doi.org/10.1038/s41586-021-03704-y, then 10.1/x'), '10.1038/s41586-021-03704-y') |
| assert.equal(findDoi('(DOI 10.1016/S0140-6736(20)30183-5)'), '10.1016/S0140-6736(20)30183-5') |
| assert.equal(findDoi('“10.1158/1078-0432.CCR-21-1694”;'), '10.1158/1078-0432.CCR-21-1694') |
| assert.equal(findDoi('Version 10.2 of the tool, page 10'), null) |
| }) |
|
|
| test('a PDF with a DOI on its first page gives the DOI', async () => { |
| const pdf = textPdf([ |
| { text: 'Science 352, 189-196 (2016)', size: 8 }, |
| { text: 'Dissecting the multicellular ecosystem of metastatic melanoma', size: 16 }, |
| { text: 'DOI: 10.1126/science.aad0501.', size: 8 }, |
| ]) |
| const { pages } = await readPdf(pdf) |
| assert.deepEqual(pages[0].map((l) => [l.text, l.size]), [ |
| ['Science 352, 189-196 (2016)', 8], |
| ['Dissecting the multicellular ecosystem of metastatic melanoma', 16], |
| ['DOI: 10.1126/science.aad0501.', 8], |
| ]) |
| assert.deepEqual(await pdfSeed(pdf), { doi: '10.1126/science.aad0501', title: 'Dissecting the multicellular ecosystem of metastatic melanoma' }) |
| }) |
|
|
| test('without a DOI, the title: the largest text on page 1 with the line it wraps onto, or the metadata title', async () => { |
| const lines = [ |
| { text: 'Author manuscript', size: 9 }, |
| { text: 'Defining T cell states associated with response', size: 18 }, |
| { text: 'to checkpoint immunotherapy in melanoma', size: 18 }, |
| { text: 'Moshe Sade-Feldman, Keren Yizhak', size: 11 }, |
| ] |
| assert.deepEqual(await pdfSeed(textPdf(lines)), { |
| doi: null, |
| title: 'Defining T cell states associated with response to checkpoint immunotherapy in melanoma', |
| }) |
| const titled = textPdf(lines, 'Defining T cell states in melanoma (author manuscript)') |
| assert.deepEqual(await pdfSeed(titled), { doi: null, title: 'Defining T cell states in melanoma (author manuscript)' }) |
| |
| assert.equal((await pdfSeed(textPdf(lines, 'Microsoft Word - draft3.docx'))).title, lines[1].text + ' ' + lines[2].text) |
| assert.deepEqual(await pdfSeed(textPdf([{ text: 'p. 1', size: 9 }])), { doi: null, title: null }) |
| }) |
|
|
| test('something that isn’t a PDF throws', async () => { |
| await assert.rejects(readPdf(new TextEncoder().encode('not a pdf at all'))) |
| }) |
|
|
| test('the PDF is kept with the uploads, once per content, with its type beside it', () => { |
| const pdf = textPdf([{ text: 'A paper', size: 12 }]) |
| const id = savePdf(pdf) |
| assert.match(id, /^pdf-[0-9a-f]{24}$/) |
| assert.deepEqual(new Uint8Array(readFileSync(join(dir, 'uploads', id))), pdf) |
| assert.equal(readFileSync(join(dir, 'uploads', `${id}.type`), 'utf8'), 'application/pdf') |
| assert.equal(savePdf(pdf), id) |
| assert.ok(!existsSync(join(dir, 'uploads', `${id}.part`))) |
| }) |
|
|
