Supporting more APIs with json_schema_compiler

tools/json_schema_compiler/compiler.py

Index: tools/json_schema_compiler/compiler.py
diff --git a/tools/json_schema_compiler/compiler.py b/tools/json_schema_compiler/compiler.py
index afd499dce29fe816baa9347fbae2bd25c63e18d9..00aefbc021122c2b3ee990abb1a749f3af7e538f 100644
--- a/tools/json_schema_compiler/compiler.py
+++ b/tools/json_schema_compiler/compiler.py
@@ -73,12 +73,13 @@ if __name__ == '__main__':
     if not namespace:
       continue
 
-    out_file = cpp_util.CppName(namespace.name).lower() + filename_suffix
+    out_file = namespace.name + filename_suffix
     type_generator = cpp_type_generator.CppTypeGenerator(root_namespace,
                                                          namespace, out_file)
     for referenced_namespace in api_model.namespaces.values():
       type_generator.AddNamespace(referenced_namespace,
-          cpp_util.CppName(referenced_namespace.name).lower() + filename_suffix)
+          cpp_util.Classname(referenced_namespace.name).lower() +
+          filename_suffix)

[not at google - send to devlin, 2012/02/06 13:14:48]

Note here, and in general, indentation here makes it look like filename_suffix
is an argument to AddNamespace().

The c++ style guide would say to format this call like

type_generator.AddNamespace(
    referenced_namespace,
    cpp_util.ClassName(referenced_namespace.name).lower() + filename_suffix)

and if that doesn't fit then

type_generator.AddNamespace(
    referenced_namespace,
    cpp_util.ClassName(referenced_namespace.name).lower() +
        filename_suffix)

i.e. all arguments/parameters should be on same line, or in line.  That's open a
bit to interpretation I suppose, but I think the same sentiment should translate
across to Python with these things.

^
apply to all line-splitting choices.

     cc_generator = cc_generator.CCGenerator(namespace, type_generator)
     cc_code = cc_generator.Generate().Render()
     h_generator = h_generator.HGenerator(namespace, type_generator)