Merge pull request #1210 from Kozea/fix_1126

Support download attribute in <a> for attachment

weasyprint/document.py

@@ -119,7 +119,8 @@ def _gather_links_and_bookmarks(box, bookmarks, links, anchors, matrix):
     has_link = link and not isinstance(box, boxes.TextBox)
     # In case of duplicate IDs, only the first is an anchor.
     has_anchor = anchor_name and anchor_name not in anchors
-    is_attachment = hasattr(box, 'is_attachment') and box.is_attachment
+    is_attachment = getattr(box, 'is_attachment', False)
+    download_name = getattr(box, 'attachment_download', None)
 
     if has_bookmark or has_link or has_anchor:
         pos_x, pos_y, width, height = box.hit_area()
@@ -133,9 +134,11 @@ def _gather_links_and_bookmarks(box, bookmarks, links, anchors, matrix):
             if matrix:
                 link = (
                     link_type, target, rectangle_aabb(
-                        matrix, pos_x, pos_y, width, height))
+                        matrix, pos_x, pos_y, width, height), download_name)
             else:
-                link = (link_type, target, (pos_x, pos_y, width, height))
+                link = (
+                    link_type, target, (pos_x, pos_y, width, height),
+                    download_name)
             links.append(link)
         if matrix and (has_bookmark or has_anchor):
             pos_x, pos_y = matrix.transform_point(pos_x, pos_y)
@@ -487,14 +490,15 @@ def resolve_links(self):
         for page in self.pages:
             page_links = []
             for link in page.links:
-                link_type, anchor_name, rectangle = link
+                link_type, anchor_name, rectangle, _ = link
                 if link_type == 'internal':
                     if anchor_name not in anchors:
                         LOGGER.error(
                             'No anchor #%s for internal URI reference',
                             anchor_name)
                     else:
-                        page_links.append((link_type, anchor_name, rectangle))
+                        page_links.append(
+                            (link_type, anchor_name, rectangle, None))
                 else:
                     # External link
                     page_links.append(link)
@@ -565,7 +569,7 @@ def add_hyperlinks(self, links, anchors, context, scale):
         # defined by cairo when drawing targets. This would give a feeling
         # similiar to what browsers do with links that span multiple lines.
         for link in links:
-            link_type, link_target, rectangle = link
+            link_type, link_target, rectangle, _ = link
             if link_type == 'external':
                 attributes = "rect=[{} {} {} {}] uri='{}'".format(*(
                     [int(round(i * scale)) for i in rectangle] +

weasyprint/html.py

@@ -249,6 +249,7 @@ def handle_td(element, box, _get_image_from_uri, _base_url):
 def handle_a(element, box, _get_image_from_uri, base_url):
     """Handle the ``rel`` attribute."""
     box.is_attachment = element_has_link_type(element, 'attachment')
+    box.attachment_download = element.get('download')
     return [box]
 
 

weasyprint/pdf.py

@@ -26,7 +26,6 @@
 
 import hashlib
 import io
-import mimetypes
 import os
 import re
 import string
@@ -361,7 +360,7 @@ def _write_compressed_file_object(pdf, file):
     return object_number
 
 
-def _get_filename_from_result(url, result):
+def _get_filename_from_url(url):
     """Derive a filename from a fetched resource.
 
     This is either the filename returned by the URL fetcher, the last URL path
@@ -371,12 +370,6 @@ def _get_filename_from_result(url, result):
 
     filename = None
 
-    # A given filename will always take precedence
-    if result:
-        filename = result.get('filename')
-        if filename:
-            return filename
-
     # The URL path likely contains a filename, which is a good second guess
     if url:
         split = urlsplit(url)
@@ -385,32 +378,7 @@ def _get_filename_from_result(url, result):
             if filename == '':
                 filename = None
 
-    if filename is None:
-        # The URL lacks a path altogether. Use a synthetic name.
-
-        # Using guess_extension is a great idea, but sadly the extension is
-        # probably random, depending on the alignment of the stars, which car
-        # you're driving and which software has been installed on your machine.
-        #
-        # Unfortuneatly this isn't even imdepodent on one machine, because the
-        # extension can depend on PYTHONHASHSEED if mimetypes has multiple
-        # extensions to offer
-        extension = None
-        if result:
-            mime_type = result.get('mime_type')
-            if mime_type == 'text/plain':
-                # text/plain has a phletora of extensions - all garbage
-                extension = '.txt'
-            else:
-                extension = mimetypes.guess_extension(mime_type) or '.bin'
-        else:
-            extension = '.bin'
-
-        filename = 'attachment' + extension
-    else:
-        filename = unquote(filename)
-
-    return filename
+    return 'attachment.bin' if filename is None else unquote(filename)
 
 
 def _write_pdf_embedded_files(pdf, attachments, url_fetcher):
@@ -438,7 +406,7 @@ def _write_pdf_embedded_files(pdf, attachments, url_fetcher):
     return pdf.write_new_object(b''.join(content))
 
 
-def _write_pdf_attachment(pdf, attachment, url_fetcher):
+def _write_pdf_attachment(pdf, attachment, url_fetcher, download_name=None):
     """Write an attachment to the PDF stream.
 
     :return:
@@ -466,7 +434,7 @@ def _write_pdf_attachment(pdf, attachment, url_fetcher):
 
     # TODO: Use the result object from a URL fetch operation to provide more
     # details on the possible filename
-    filename = _get_filename_from_result(url, None)
+    filename = download_name or _get_filename_from_url(url)
 
     return pdf.write_new_object(pdf_format(
         '<< /Type /Filespec /F () /UF {0!P} /EF << /F {1} 0 R >> '
@@ -509,11 +477,11 @@ def write_pdf_metadata(fileobj, scale, url_fetcher, attachments,
     # because two links might have the same href, but different titles.
     annot_files = {}
     for page_links in attachment_links:
-        for link_type, target, rectangle in page_links:
+        for link_type, target, rectangle, download_name in page_links:
             if link_type == 'attachment' and target not in annot_files:
                 # TODO: use the title attribute as description
                 annot_files[target] = _write_pdf_attachment(
-                    pdf, (target, None), url_fetcher)
+                    pdf, (target, None), url_fetcher, download_name)
 
     for pdf_page, document_page, page_links in zip(
             pdf.pages, pages, attachment_links):
@@ -555,7 +523,7 @@ def write_pdf_metadata(fileobj, scale, url_fetcher, attachments,
         # would give a feeling similiar to what browsers do with links that
         # span multiple lines.
         annotations = []
-        for link_type, target, rectangle in page_links:
+        for link_type, target, rectangle, _ in page_links:
             if link_type == 'attachment' and annot_files[target] is not None:
                 matrix = cairo.Matrix(
                     xx=scale, yy=-scale, y0=document_page.height * scale)

weasyprint/tests/test_api.py

@@ -121,10 +121,12 @@ def _round_meta(pages):
             anchors[anchor_name] = round(pos_x, 6), round(pos_y, 6)
         links = page.links
         for i, link in enumerate(links):
-            link_type, target, (pos_x, pos_y, width, height) = link
+            link_type, target, (pos_x, pos_y, width, height), download_name = (
+                link)
             link = (
                 link_type, target, (round(pos_x, 6), round(pos_y, 6),
-                                    round(width, 6), round(height, 6)))
+                                    round(width, 6), round(height, 6)),
+                download_name)
             links[i] = link
         bookmarks = page.bookmarks
         for i, (level, label, (pos_x, pos_y), state) in enumerate(bookmarks):
@@ -720,28 +722,28 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
         </p>
     ''', [
         [
-            ('external', 'http://weasyprint.org', (0, 0, 30, 20)),
-            ('external', 'http://weasyprint.org', (0, 0, 30, 30)),
-            ('internal', 'lipsum', (10, 100, 32, 20)),
-            ('internal', 'lipsum', (10, 100, 32, 32))
+            ('external', 'http://weasyprint.org', (0, 0, 30, 20), None),
+            ('external', 'http://weasyprint.org', (0, 0, 30, 30), None),
+            ('internal', 'lipsum', (10, 100, 32, 20), None),
+            ('internal', 'lipsum', (10, 100, 32, 32), None)
         ],
-        [('internal', 'hello', (0, 0, 200, 30))],
+        [('internal', 'hello', (0, 0, 200, 30), None)],
     ], [
         {'hello': (0, 200)},
         {'lipsum': (0, 0)}
     ], [
         (
             [
-                ('external', 'http://weasyprint.org', (0, 0, 30, 20)),
-                ('external', 'http://weasyprint.org', (0, 0, 30, 30)),
-                ('internal', 'lipsum', (10, 100, 32, 20)),
-                ('internal', 'lipsum', (10, 100, 32, 32))
+                ('external', 'http://weasyprint.org', (0, 0, 30, 20), None),
+                ('external', 'http://weasyprint.org', (0, 0, 30, 30), None),
+                ('internal', 'lipsum', (10, 100, 32, 20), None),
+                ('internal', 'lipsum', (10, 100, 32, 32), None)
             ],
             [('hello', 0, 200)],
         ),
         (
             [
-                ('internal', 'hello', (0, 0, 200, 30))
+                ('internal', 'hello', (0, 0, 200, 30), None)
             ],
             [('lipsum', 0, 0)]),
     ])
@@ -751,28 +753,30 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
             <body style="width: 200px">
             <a href="../lipsum/é_%E9" style="display: block; margin: 10px 5px">
         ''', [[('external', 'http://weasyprint.org/foo/lipsum/%C3%A9_%E9',
-                (5, 10, 190, 0))]],
+                (5, 10, 190, 0), None)]],
         [{}], [([('external', 'http://weasyprint.org/foo/lipsum/%C3%A9_%E9',
-                  (5, 10, 190, 0))], [])],
+                  (5, 10, 190, 0), None)], [])],
         base_url='http://weasyprint.org/foo/bar/')
+
     assert_links(
         '''
             <body style="width: 200px">
             <div style="display: block; margin: 10px 5px;
                         -weasy-link: url(../lipsum/é_%E9)">
         ''', [[('external', 'http://weasyprint.org/foo/lipsum/%C3%A9_%E9',
-                (5, 10, 190, 0))]],
+                (5, 10, 190, 0), None)]],
         [{}], [([('external', 'http://weasyprint.org/foo/lipsum/%C3%A9_%E9',
-                  (5, 10, 190, 0))], [])],
+                  (5, 10, 190, 0), None)], [])],
         base_url='http://weasyprint.org/foo/bar/')
 
     # Relative URI reference without a base URI: allowed for links
     assert_links(
         '''
             <body style="width: 200px">
             <a href="../lipsum" style="display: block; margin: 10px 5px">
-        ''', [[('external', '../lipsum', (5, 10, 190, 0))]], [{}],
-        [([('external', '../lipsum', (5, 10, 190, 0))], [])], base_url=None)
+        ''', [[('external', '../lipsum', (5, 10, 190, 0), None)]], [{}],
+        [([('external', '../lipsum', (5, 10, 190, 0), None)], [])],
+        base_url=None)
 
     # Relative URI reference without a base URI: not supported for -weasy-link
     assert_links(
@@ -791,11 +795,11 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
             <a href="#lipsum" id="lipsum"
                 style="display: block; margin: 10px 5px"></a>
             <a href="http://weasyprint.org/" style="display: block"></a>
-        ''', [[('internal', 'lipsum', (5, 10, 190, 0)),
-               ('external', 'http://weasyprint.org/', (0, 10, 200, 0))]],
+        ''', [[('internal', 'lipsum', (5, 10, 190, 0), None),
+               ('external', 'http://weasyprint.org/', (0, 10, 200, 0), None)]],
         [{'lipsum': (5, 10)}],
-        [([('internal', 'lipsum', (5, 10, 190, 0)),
-           ('external', 'http://weasyprint.org/', (0, 10, 200, 0))],
+        [([('internal', 'lipsum', (5, 10, 190, 0), None),
+           ('external', 'http://weasyprint.org/', (0, 10, 200, 0), None)],
           [('lipsum', 5, 10)])],
         base_url=None)
 
@@ -805,9 +809,10 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
             <div style="-weasy-link: url(#lipsum);
                         margin: 10px 5px" id="lipsum">
         ''',
-        [[('internal', 'lipsum', (5, 10, 190, 0))]],
+        [[('internal', 'lipsum', (5, 10, 190, 0), None)]],
         [{'lipsum': (5, 10)}],
-        [([('internal', 'lipsum', (5, 10, 190, 0))], [('lipsum', 5, 10)])],
+        [([('internal', 'lipsum', (5, 10, 190, 0), None)],
+          [('lipsum', 5, 10)])],
         base_url=None)
 
     assert_links(
@@ -817,10 +822,11 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
                 <a href="#lipsum"></a>
                 <a href="#missing" id="lipsum"></a>
         ''',
-        [[('internal', 'lipsum', (0, 0, 200, 15)),
-          ('internal', 'missing', (0, 15, 200, 15))]],
+        [[('internal', 'lipsum', (0, 0, 200, 15), None),
+          ('internal', 'missing', (0, 15, 200, 15), None)]],
         [{'lipsum': (0, 15)}],
-        [([('internal', 'lipsum', (0, 0, 200, 15))], [('lipsum', 0, 15)])],
+        [([('internal', 'lipsum', (0, 0, 200, 15), None)],
+          [('lipsum', 0, 15)])],
         base_url=None,
         warnings=[
             'ERROR: No anchor #missing for internal URI reference'])
@@ -831,11 +837,24 @@ def assert_links(html, expected_links_by_page, expected_anchors_by_page,
             <a href="#lipsum" id="lipsum" style="display: block; height: 20px;
                 transform: rotate(90deg) scale(2)">
         ''',
-        [[('internal', 'lipsum', (30, 10, 40, 200))]],
+        [[('internal', 'lipsum', (30, 10, 40, 200), None)]],
         [{'lipsum': (70, 10)}],
-        [([('internal', 'lipsum', (30, 10, 40, 200))], [('lipsum', 70, 10)])],
+        [([('internal', 'lipsum', (30, 10, 40, 200), None)],
+          [('lipsum', 70, 10)])],
         round=True)
 
+    # Download for attachment
+    assert_links(
+        '''
+            <body style="width: 200px">
+            <a rel=attachment href="pattern.png" download="wow.png"
+                style="display: block; margin: 10px 5px">
+        ''', [[('attachment', 'pattern.png',
+                (5, 10, 190, 0), 'wow.png')]],
+        [{}], [([('attachment', 'pattern.png',
+                  (5, 10, 190, 0), 'wow.png')], [])],
+        base_url=None)
+
 
 # Make relative URL references work with our custom URL scheme.
 uses_relative.append('weasyprint-custom')

weasyprint/tools/navigator.py

@@ -65,7 +65,7 @@ def render_template(url):
         for width, height, data_url, links, anchors in get_pages(html):
             write('<section style="width: {0}px; height: {1}px">\n'
                   '  <img src="{2}">\n'.format(width, height, data_url))
-            for link_type, target, (pos_x, pos_y, width, height) in links:
+            for link_type, target, (pos_x, pos_y, width, height), _ in links:
                 href = ('#' + target if link_type == 'internal'
                         else '/view/' + target)
                 write('  <a style="left: {0}px; top: {1}px; '