Merge branch 'main' of https://huggingface.co/spaces/XicoC/AIM-Endpoint-Assignment

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/paul-graham-to-kindle/paul_graham_essays.epub filter=lfs diff=lfs merge=lfs -text

data/paul-graham-to-kindle/.gitignore

@@ -0,0 +1,32 @@
+# See https://help.github.com/articles/ignoring-files/ for more about ignoring files.
+
+# dependencies
+/node_modules
+/.pnp
+.pnp.js
+
+# testing
+/coverage
+
+# next.js
+/.next/
+/out/
+
+# production
+/build
+
+# misc
+.DS_Store
+*.pem
+
+# debug
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+.pnpm-debug.log*
+
+# local env files
+.env*.local
+
+# vercel
+.vercel

data/paul-graham-to-kindle/README.md

@@ -0,0 +1,25 @@
+# PaulG to Kindle
+
+Hackers and Painters was amazing, but I wanted to read the rest of Paul's essays on my Kindle. I couldn't find a way to do this, so I wrote a script to convert the essays to epub format.
+
+Make sure you have pandoc installed.
+
+    `brew install pandoc`
+
+Then run the script:
+
+    `npm i && npm run generate`
+
+Either email the generated `paul_graham_essays.epub` file to your Kindle email address as an attachment OR send to your kindle from [amazon.com/sendtokindle](https://www.amazon.com/sendtokindle).
+
+## How to find Kindle email address
+
+1. Go to [Manage Your Content and Devices](https://www.amazon.com/gp/digital/fiona/manage)
+
+2. Click on the device you want to send the email to
+
+3. Click on the `Settings` tab
+
+4. Scroll down to `Send-to-Kindle Email Settings`
+
+5. Copy the email address

data/paul-graham-to-kindle/jsconfig.json

@@ -0,0 +1,7 @@
+{
+  "compilerOptions": {
+    "paths": {
+      "@/*": ["./*"]
+    }
+  }
+}

data/paul-graham-to-kindle/lib/metadata.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<metadata xmlns="http://purl.org/dc/elements/1.1/">
+  <dc:title>Paul Graham Essays</dc:title>
+  <dc:creator>Paul Graham</dc:creator>
+  <dc:language>en</dc:language>
+</metadata>

data/paul-graham-to-kindle/lib/pg.js

@@ -0,0 +1,104 @@
+const axios = require('axios');
+const cheerio = require('cheerio');
+const { spawn } = require('child_process');
+const fs = require('fs');
+const TurndownService = require('turndown');
+
+const turndownService = new TurndownService();
+
+const essaysListUrl = 'http://www.paulgraham.com/articles.html';
+
+async function main() {
+    try {
+        const essays = await getEssayLinks();
+        essays.reverse();
+        const essayTexts = await Promise.all(essays.map((essay) => getEssayText(essay)));
+
+        const combinedText = essayTexts.join('\n\n');
+        console.log('Combined text:', combinedText);
+        fs.writeFileSync('paul_graham_essays.txt', combinedText);
+
+        const outputFile = 'paul_graham_essays.epub';
+        await generateEpub('paul_graham_essays.txt', outputFile);
+        console.log(`EPUB file generated: ${outputFile}`);
+    } catch (error) {
+        console.error('Error:', error);
+    }
+}
+
+function generateEpub(inputFile, outputFile) {
+    return new Promise((resolve, reject) => {
+        const pandoc = spawn('pandoc', [
+            inputFile,
+            '-o',
+            outputFile,
+            '--toc',
+            '--toc-depth=1',
+            '--metadata',
+            'title="Paul Graham Essays"'
+        ]);
+
+        pandoc.stdout.on('data', (data) => {
+            console.log(`stdout: ${data}`);
+        });
+
+        pandoc.stderr.on('data', (data) => {
+            console.error(`stderr: ${data}`);
+        });
+
+        pandoc.on('close', (code) => {
+            if (code === 0) {
+                resolve();
+            } else {
+                reject(new Error(`pandoc exited with code ${code}`));
+            }
+        });
+    });
+}
+const baseUrl = 'http://www.paulgraham.com/';
+
+async function getEssayLinks() {
+    const response = await axios.get(essaysListUrl);
+    const $ = cheerio.load(response.data);
+
+    const links = [];
+
+    $('a').each((_, link) => {
+        const url = $(link).attr('href');
+        if (url && url.endsWith('.html')) {
+            links.push(baseUrl + url);
+        }
+    });
+
+    return links;
+}
+
+async function getEssayText(url) {
+    try {
+        const response = await axios.get(url);
+        const $ = cheerio.load(response.data);
+
+        $('script').remove();
+        $('img').remove();
+
+        const title = $('title').text();
+        const htmlContent = $('body').html();
+        const markdownContent = turndownService.turndown(htmlContent);
+
+        return `\n\n# ${title}\n\n${markdownContent}`;
+    } catch (error) {
+        console.error(`Error fetching essay: ${url}\n${error}`);
+        return '';
+    }
+}
+
+main().then(() => {
+    const outputFile = 'paul_graham_essays.epub';
+    generateEpub('paul_graham_essays.txt', outputFile, (error) => {
+        if (error) {
+            console.error('Error generating EPUB file:', error);
+            return;
+        }
+        console.log(`EPUB file generated: ${outputFile}`);
+    });
+});

data/paul-graham-to-kindle/metadata.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<metadata xmlns="http://purl.org/dc/elements/1.1/">
+  <dc:title>Paul Graham Essays</dc:title>
+  <dc:creator>Paul Graham</dc:creator>
+  <dc:language>en</dc:language>
+</metadata>

data/paul-graham-to-kindle/package.json

@@ -0,0 +1,14 @@
+{
+  "name": "paulg-to-kindle",
+  "version": "0.1.0",
+  "private": true,
+  "scripts": {
+    "generate": "node ./lib/pg.js"
+  },
+  "dependencies": {
+    "axios": "^1.3.5",
+    "cheerio": "^1.0.0-rc.12",
+    "html-to-text": "^9.0.5",
+    "turndown": "^7.1.2"
+  }
+}

data/paul-graham-to-kindle/paul_graham_essays.epub

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:051f282341cab2a14b11bf77cf2f858fdc51700136b617a95577e6b529734c64
+size 1364157