Issue 5333 - Allow cms to generate relative pages

cms/bin/generate_static_pages.py

Index: cms/bin/generate_static_pages.py
diff --git a/cms/bin/generate_static_pages.py b/cms/bin/generate_static_pages.py
index 242dc0128fbeeacac98ff5897144823285c82bae..f0fccc22055fe0540a772e05671446a44f5f3b03 100644
--- a/cms/bin/generate_static_pages.py
+++ b/cms/bin/generate_static_pages.py
@@ -19,6 +19,8 @@ import errno
 import codecs
 import ConfigParser
 import logging
+import lxml.html
+
 from argparse import ArgumentParser
 
 from cms.utils import get_page_params, process_page
@@ -27,7 +29,7 @@ from cms.sources import create_source
 MIN_TRANSLATED = 0.3
 
 
-def generate_pages(repo, output_dir, revision):
+def generate_pages(repo, output_dir, revision, relative):
     known_files = set()
 
     def write_file(path_parts, contents, binary=False):
@@ -90,15 +92,43 @@ def generate_pages(repo, output_dir, revision):
         source.has_locale = has_locale
         source.resolve_link.cache_clear()
 
+        def rewrite_link(link):

[Vasily Kuznetsov, 2017/11/14 17:30:27]

This seems like it would work but the links would look nicer if we compact the
unnecessary pairs of `../some_path`, where the `some_path` part is the same in
the link and in the current page. Perhaps this is not first priority though and
is more of a cosmetic feature.

+            if link.startswith('/'):
+                if source.version:
+                    link += '?' + source.version
+                if relative:
+                    depth = len(page.split('/'))
+                    link = '/'.join(['..'] * depth) + link
+            return link
+
         # Second pass: actually generate pages this time
         for locale, page in pagelist:
             pagedata = process_page(source, locale, page)
 
-            # Make sure links to static files are versioned
-            pagedata = re.sub(r'(<script\s[^<>]*\bsrc="/[^"<>]+)', r'\1?%s' % source.version, pagedata)
-            pagedata = re.sub(r'(<link\s[^<>]*\bhref="/[^"<>]+)', r'\1?%s' % source.version, pagedata)
-            pagedata = re.sub(r'(<img\s[^<>]*\bsrc="/[^"<>]+)', r'\1?%s' % source.version, pagedata)
-
+            root = lxml.html.fromstring(pagedata).getroottree()
+
+            expr = '//img[@src|@srcset]|//script[@src]|//link[@href]|//a[@href]'
+            for elem in root.xpath(expr):
+                if 'srcset' in elem.attrib:
+                    srcset = []
+                    for src_zoom in elem.attrib['srcset'].split(', '):

[Vasily Kuznetsov, 2017/11/14 17:30:28]

Is the space after the comma mandatory? Otherwise we can't just split by ', ',
we need to split by comma.

+                        try:
+                            src, zoom = src_zoom.split(' ')
+                        except ValueError:
+                            src = src_zoom
+                            zoom = None
+                        res = rewrite_link(src)
+                        if zoom:
+                             res = res + ' ' + zoom
+                        srcset.append(res)
+
+                    elem.attrib['srcset'] = ', '.join(srcset)
+
+                for key in ['src', 'href']:
+                    if key in elem.attrib:
+                        elem.attrib[key] = rewrite_link(elem.attrib[key])
+
+            pagedata = lxml.html.tostring(root, encoding='unicode')
             write_file([locale] + page.split('/'), pagedata)
 
         for filename in source.list_localizable_files():
@@ -129,7 +159,8 @@ if __name__ == '__main__':
                         help=('Specify which revision to generate from. '
                               'See "hg help revisions" for details.'),
                         default='default')
+    parser.add_argument('--relative', action='store_true', default=False)
     parser.add_argument('source', help="Path to website's repository")
     parser.add_argument('output', help='Path to desired output directory')
     args = parser.parse_args()
-    generate_pages(args.source, args.output, args.rev)
+    generate_pages(args.source, args.output, args.rev, args.relative)