Feature - accept content to render pdf by url

.gitignore

@@ -60,3 +60,12 @@ target/
 
 #Ipython Notebook
 .ipynb_checkpoints
+
+# Editor directories and files
+.idea
+.vscode
+*.iml
+*.suo
+*.ntvs*
+*.njsproj
+*.sln

README.md

@@ -39,4 +39,15 @@ curl -v -X POST -d @test.html -JLO http://127.0.0.1:5001/pdf?filename=result.pdf
 
 This example will use the file `test.html` and return a response with `Content-Type: application/pdf` and `Content-Disposition: inline; filename=result.pdf` headers.  The body of the response will be the PDF rendering of the html document.
 
+`POST` to `/pdf` can be used to generate a PDF from a URL. Use the `type=url` query parameter and a url in a text/plain body, call:
+
+```
+curl --location --request POST 'http://127.0.0.1:5001/pdf?filename=result_from_url.pdf&type=url' -JLO --header 'Content-Type: text/plain' --data-raw 'https://www.google.ca/?client=safari&channel=iphone_bm'
+```
+
+This example will use the url `https://www.google.ca/?client=safari&channel=iphone_bm` and return a response with `Content-Type: application/pdf` and `Content-Disposition: inline; filename=result_from_url.pdf` headers.  The body of the response will be the PDF rendering of the url (renders static html only, not for Single Page Applications).
+
+When `type=url`, additional query parameters can be passed in: `encoding`, `media_type`, `base_url`.  See [Weasyprint HTML api](https://weasyprint.readthedocs.io/en/stable/api.html#weasyprint.HTML).  
+
+
 In addition `/health` is a health check endpoint and a `GET` returns 'ok'.

app.py

@@ -40,16 +40,25 @@ def home():
             <li>POST to <code>/multiple?filename=myfile.pdf</code>. The body
                 should contain a JSON list of html strings. They will each
                 be rendered and combined into a single pdf</li>
+            <li>POST to <code>/pdf?filename=myfile.pdf&type=url</code>. The body 
+                should be plain text and must contain a url</li>
         </ul>
     '''
 
 
 @app.route('/pdf', methods=['POST'])
 def generate():
     name = request.args.get('filename', 'unnamed.pdf')
-    app.logger.info('POST  /pdf?filename=%s' % name)
-    #print ( request.get_data() )
-    html = HTML(string=request.get_data())
+    type = request.args.get('type', 'string')
+    app.logger.info('POST  /pdf?filename=%s&type=%s' % (name, type))
+    html = None
+    if type == 'string':
+        html = HTML(string=request.get_data())
+    elif type == 'url':
+        url_args = {'encoding': request.args.get('encoding',  None),
+                    'media_type': request.args.get('media_type', 'print'),
+                    'base_url': request.args.get('base_url', None)}
+        html = HTML(url=request.get_data().decode('utf-8'), **url_args)
     pdf = html.write_pdf()
     response = make_response(pdf)
     response.headers['Content-Type'] = 'application/pdf'

test.py

@@ -76,5 +76,34 @@ def test_body(self):
         self.assertEqual(len(pages), 2)
 
 
+def request_factory_url(path='/'):
+    url = 'http://127.0.0.1:5001%s' % path
+    headers = {
+        'Content-Type': 'text/plain'
+    }
+    return Request(url, data='https://www.google.ca/?client=safari&channel=iphone_bm'.encode('utf-8'), headers=headers, method='POST')
+
+
+class TestUrlPdf(unittest.TestCase):
+
+    def setUp(self):
+        request = request_factory_url('/pdf?filename=sample_url.pdf&type=url')
+        self.response = urlopen(request)
+
+    def tearDown(self):
+        self.response.close()
+
+    def test_response_code(self):
+        self.assertEqual(self.response.getcode(), 200)
+
+    def test_headers(self):
+        headers = dict(self.response.info())
+        self.assertEqual(headers['Content-Type'], 'application/pdf')
+        self.assertEqual(headers['Content-Disposition'], 'inline;filename=sample_url.pdf')
+
+    def test_body(self):
+        self.assertEqual(self.response.read()[:4], b'%PDF')
+
+
 if __name__ == '__main__':
     unittest.main()