Issue 2340 - Don`t generate pages if less than 30% have been translated

cms/bin/generate_static_pages.py

Index: cms/bin/generate_static_pages.py
===================================================================
--- a/cms/bin/generate_static_pages.py
+++ b/cms/bin/generate_static_pages.py
@@ -15,28 +15,32 @@
 # You should have received a copy of the GNU General Public License
 # along with Adblock Plus.  If not, see <http://www.gnu.org/licenses/>.
 
 import sys
 import os
 import re
 import errno
 import codecs
+import ConfigParser
 import logging
 
-from cms.utils import process_page
+from cms.utils import get_page_params, process_page
 from cms.sources import MercurialSource
 
+MIN_TRANSLATED = 0.3
+
 def memoize(func):
   memoized = {}
   def wrapper(*args):
     try:
       return memoized[args]
     except KeyError:
       return memoized.setdefault(args, func(*args))
+  wrapper.clear_cache = memoized.clear
   return wrapper
 
 def generate_pages(repo, output_dir):
   known_files = set()
 
   def write_file(path_parts, contents, binary=False):
     encoding = None if binary else "utf-8"
     outfile = os.path.join(output_dir, *path_parts)
@@ -68,27 +72,54 @@ def generate_pages(repo, output_dir):
     source.read_locale = memoize(source.read_locale)
     source.read_include = memoize(source.read_include)
 
     config = source.read_config()
     defaultlocale = config.get("general", "defaultlocale")
     locales = list(source.list_locales())
     if defaultlocale not in locales:
       locales.append(defaultlocale)
+
+    # First pass: compile the list of pages with given translation level
+    pagelist = set()
+    blacklist = set()
     for page, format in source.list_pages():
       for locale in locales:
-        if locale == defaultlocale or source.has_locale(locale, page):
-          pagedata = process_page(source, locale, page, format)
+        if locale == defaultlocale:
+          pagelist.add((locale, page))
+        elif source.has_locale(locale, page):
+          params = get_page_params(source, locale, page, format)
+          if params["translation_ratio"] >= MIN_TRANSLATED:
+            pagelist.add((locale, page))
+          else:
+            blacklist.add((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)
+    # Override existance check to avoid linking to pages we don't generate
+    orig_has_locale = source.has_locale
+    def has_locale(locale, page):
+      try:
+        page = config.get("locale_overrides", page)
+      except ConfigParser.Error:
+        pass
+      if (locale, page) in blacklist:
+        return False
+      return orig_has_locale(locale, page)
+    source.has_locale = has_locale
+    source.resolve_link.clear_cache()
 
-          write_file([locale] + page.split("/"), pagedata)
+    # 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)
+
+      write_file([locale] + page.split("/"), pagedata)
 
     for filename in source.list_localizable_files():
       for locale in locales:
         if source.has_localizable_file(locale, filename):
           filedata = source.read_localizable_file(locale, filename)
           write_file([locale] + filename.split("/"), filedata, binary=True)
 
     for filename in source.list_static():