Merge remote-tracking branch 'origin/pr/2966'

scripts/importer.py

@@ -28,50 +28,176 @@ Currently only importing bookmarks from Netscape Bookmark files is supported.
 
 import argparse
 
+browser_default_input_format = {
+    'chromium': 'netscape',
+    'ie': 'netscape',
+    'firefox': 'netscape',
+    'seamonkey': 'netscape',
+    'palemoon': 'netscape'
+}
+
 
 def main():
     args = get_args()
-    if args.browser in ['chromium', 'firefox', 'ie']:
+    bookmark_types = []
-        import_netscape_bookmarks(args.bookmarks, args.bookmark_format)
+    output_format = None
+    input_format = args.input_format
+    if args.search_output:
+        bookmark_types = ['search']
+        if args.oldconfig:
+            output_format = 'oldsearch'
+        else:
+            output_format = 'search'
+    else:
+        if args.bookmark_output:
+            output_format = 'bookmark'
+        elif args.quickmark_output:
+            output_format = 'quickmark'
+        if args.import_bookmarks:
+            bookmark_types.append('bookmark')
+        if args.import_keywords:
+            bookmark_types.append('keyword')
+    if not bookmark_types:
+        bookmark_types = ['bookmark', 'keyword']
+    if not output_format:
+        output_format = 'quickmark'
+    if not input_format:
+        if args.browser:
+            input_format = browser_default_input_format[args.browser]
+        else:
+            #default to netscape
+            input_format = 'netscape'
+
+    import_function = {'netscape': import_netscape_bookmarks}
+    import_function[input_format](args.bookmarks, bookmark_types,
+                                  output_format)
 
 
 def get_args():
     """Get the argparse parser."""
     parser = argparse.ArgumentParser(
         epilog="To import bookmarks from Chromium, Firefox or IE, "
-               "export them to HTML in your browsers bookmark manager. "
+        "export them to HTML in your browsers bookmark manager. ")
-               "By default, this script will output in a quickmarks format.")
+    parser.add_argument(
-    parser.add_argument('browser', help="Which browser? (chromium, firefox)",
+        'browser',
-                        choices=['chromium', 'firefox', 'ie'],
+        help="Which browser? {%(choices)s}",
+        choices=browser_default_input_format.keys(),
+        nargs='?',
         metavar='browser')
-    parser.add_argument('-b', help="Output in bookmark format.",
+    parser.add_argument(
-                        dest='bookmark_format', action='store_true',
+        '-i',
-                        default=False, required=False)
+        '--input-format',
+        help='Which input format? (overrides browser default; "netscape" if '
+        'neither given)',
+        choices=set(browser_default_input_format.values()),
+        required=False)
+    parser.add_argument(
+        '-b',
+        '--bookmark-output',
+        help="Output in bookmark format.",
+        action='store_true',
+        default=False,
+        required=False)
+    parser.add_argument(
+        '-q',
+        '--quickmark-output',
+        help="Output in quickmark format (default).",
+        action='store_true',
+        default=False,
+        required=False)
+    parser.add_argument(
+        '-s',
+        '--search-output',
+        help="Output config.py search engine format (negates -B and -K)",
+        action='store_true',
+        default=False,
+        required=False)
+    parser.add_argument(
+        '--oldconfig',
+        help="Output search engine format for old qutebrowser.conf format",
+        default=False,
+        action='store_true',
+        required=False)
+    parser.add_argument(
+        '-B',
+        '--import-bookmarks',
+        help="Import plain bookmarks (can be combiend with -K)",
+        action='store_true',
+        default=False,
+        required=False)
+    parser.add_argument(
+        '-K',
+        '--import-keywords',
+        help="Import keywords (can be combined with -B)",
+        action='store_true',
+        default=False,
+        required=False)
     parser.add_argument('bookmarks', help="Bookmarks file (html format)")
     args = parser.parse_args()
     return args
 
 
-def import_netscape_bookmarks(bookmarks_file, is_bookmark_format):
+def search_escape(url):
+    """Escape URLs such that preexisting { and } are handled properly.
+
+    Will obviously trash a properly-formatted Qutebrowser URL.
+    """
+    return url.replace('{', '{{').replace('}', '}}')
+
+
+def import_netscape_bookmarks(bookmarks_file, bookmark_types, output_format):
     """Import bookmarks from a NETSCAPE-Bookmark-file v1.
 
-    Generated by Chromium, Firefox, IE and possibly more browsers
+    Generated by Chromium, Firefox, IE and possibly more browsers. Not all
+    export all possible bookmark types:
+        - Firefox mostly works with everything
+        - Chrome doesn't support keywords at all; searches are a separate
+          database
     """
     import bs4
     with open(bookmarks_file, encoding='utf-8') as f:
         soup = bs4.BeautifulSoup(f, 'html.parser')
-
+    bookmark_query = {
-    html_tags = soup.findAll('a')
+        'search': lambda tag: (
-    if is_bookmark_format:
+            (tag.name == 'a') and
-        output_template = '{tag[href]} {tag.string}'
+            ('shortcuturl' in tag.attrs) and
-    else:
+            ('%s' in tag['href'])),
-        output_template = '{tag.string} {tag[href]}'
+        'keyword': lambda tag: (
-
+            (tag.name == 'a') and
+            ('shortcuturl' in tag.attrs) and
+            ('%s' not in tag['href'])),
+        'bookmark': lambda tag: (
+            (tag.name == 'a') and
+            ('shortcuturl' not in tag.attrs) and
+            (tag.string)),
+    }
+    output_template = {
+        'search': {
+            'search':
+            "c.url.searchengines['{tag[shortcuturl]}'] = "
+            "'{tag[href]}' #{tag.string}"
+        },
+        'oldsearch': {
+            'search': '{tag[shortcuturl]} = {tag[href]} #{tag.string}',
+        },
+        'bookmark': {
+            'bookmark': '{tag[href]} {tag.string}',
+            'keyword': '{tag[href]} {tag.string}'
+        },
+        'quickmark': {
+            'bookmark': '{tag.string} {tag[href]}',
+            'keyword': '{tag[shortcuturl]} {tag[href]}'
+        }
+    }
     bookmarks = []
-    for tag in html_tags:
+    for typ in bookmark_types:
+        tags = soup.findAll(bookmark_query[typ])
+        for tag in tags:
+            if typ == 'search':
+                tag['href'] = search_escape(tag['href']).replace('%s', '{}')
             if tag['href'] not in bookmarks:
-            bookmarks.append(output_template.format(tag=tag))
+                bookmarks.append(
-
+                    output_template[output_format][typ].format(tag=tag))
     for bookmark in bookmarks:
         print(bookmark)