Issue 2936 - Support syntax for duplicate translatable strings

cms/converters.py

 # coding: utf-8
 
 # This file is part of the Adblock Plus web scripts,
 # Copyright (C) 2006-2015 Eyeo GmbH
 #
 # Adblock Plus is free software: you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 3 as
 # published by the Free Software Foundation.
 #
 # Adblock Plus is distributed in the hope that it will be useful,
@@ skipping 99 matching lines @@
 
 class Converter:
   whitelist = {"a", "em", "strong", "code", "span"}
   missing_translations = 0
   total_translations = 0
 
   def __init__(self, params, key="pagedata"):
     self._params = params
     self._key = key
     self._attribute_parser = AttributeParser(self.whitelist)
+    self._seen_defaults  = {}
 
     # Read in any parameters specified at the beginning of the file
     lines = params[key].splitlines(True)
     while lines and re.search(r"^\s*[\w\-]+\s*=", lines[0]):
       name, value = lines.pop(0).split("=", 1)
       params[name.strip()] = value.strip()
     params[key] = "".join(lines)
 
   def localize_string(self, page, name, default, comment, localedata, escapes):
     def escape(s):
       return re.sub(r".",
         lambda match: escapes.get(match.group(0), match.group(0)),
         s, flags=re.S)
     def re_escape(s):
       return re.escape(escape(s))
 
-    locale = self._params["locale"]
-    is_default_locale = locale == self._params["defaultlocale"]
-
     # Handle duplicated strings
-    if not default:
+    if default:
+      self._seen_defaults[(page, name)] = (default, comment)
+    else:
       try:
-        default = self._params["localedata"][name]

[Wladimir Palant, 2015/08/24 11:07:17]

This logic is wrong, the default value should never be localized - it's the
English string specified in the page source. As you have it now, placeholders
will be messed up.

[kzar, 2015/08/24 14:29:45]

Acknowledged.

+        default, comment = self._seen_defaults[(page, name)]
       except KeyError:
         raise Exception("Text not yet defined for string %s on page %s" %
-                        (name, self._params["page"]))
-    elif is_default_locale:
-      self._params["localedata"][name] = default

[Wladimir Palant, 2015/08/24 11:07:17]

1) This should be using localedata parameter instead of
self._params["localedata"].

2) I don't think that messing up localization data is a good approach, e.g.
because this localization data will be reused by other files and lead to strange
errors. Also, it would be a mess with placeholders specified differently in the
default value and in localization files.

Why not add some property for string defaults we've seen already? IMHO the logic
here should be:

  if default:
    # Maybe warn if the same default value is specified multiple times?
    self._seen_default[(page, name)] = default
  else:
    default = self._seen_default[(page, name)]  # might throw

That's it, completely independent of which locale we are processing.

[kzar, 2015/08/24 14:29:46]

Done.

+                        (name, page))
 
     # Extract tag attributes from default string
     default, saved_attributes, fixed_strings = self._attribute_parser.parse(default, self._params["page"])
 
     # Get translation
-    if is_default_locale:
+    locale = self._params["locale"]
+    if locale == self._params["defaultlocale"]:
       result = default
     elif name in localedata:
       result = localedata[name].strip()
     else:
       result = default
       self.missing_translations += 1
     self.total_translations += 1
 
-    if default:

[Wladimir Palant, 2015/08/24 11:07:17]

Why add this check? The string is still there, regardless of whether a default
value has been specified.

[kzar, 2015/08/24 14:29:45]

Well as a consumer of this callback I would be surprised to receive multiple
callbacks for the same string. Also we would be missing the string comment after
the first callback.

[Wladimir Palant, 2015/08/24 15:00:21]

That's something one has to count on anyway.
True, we can save the comment as well.

-      # Perform callback with the string if required,
-      # e.g. for the translations script
-      callback = self._params["localized_string_callback"]
-      if callback:
-        callback(page, locale, name, result, comment, fixed_strings)
+    # Perform callback with the string if required, e.g. for the translations script
+    callback = self._params["localized_string_callback"]
+    if callback:
+      callback(page, locale, name, result, comment, fixed_strings)
+
 
     # Insert fixed strings
     for i, fixed_string in enumerate(fixed_strings, 1):
       result = result.replace("{%d}" % i, fixed_string)
 
     # Insert attributes
     result = escape(result)
     def stringify_attribute((name, value)):
       return '%s="%s"' % (
         escape(name),
@@ skipping 241 matching lines @@
         stack.pop()
       stack[-1]["subitems"].append(item)
       stack.append(item)
     return structured
 
 converters = {
   "html": RawConverter,
   "md": MarkdownConverter,
   "tmpl": TemplateConverter,
 }