Gregg Kellogg

Seems to have been fixed in commit eb201e64c0b68aee6d0715d44cf48178204c4133 "Fixed Hash#from_xml with keys that are all caps."

Note that authenticate_or_request_http_digest can take either the clear-text password or the HA1 digest hash of the username and password: MD5(username:realm:pa...

Okay, here's an updated activesupport/test/core_ext/hash_ext_test.rb with new test test_single_record_from_xml_with_camel_case

See if this works.

@@@ruby xml = %(<blah_blah> 1 2 <dasherized_tag> 3 </dasherized_tag> </blah_blah>)

Oops! It seems that it's getting eaten on the submit. I'll modify the appropriate unit test and upload it.

It would probably help if i used the whole test string:

xml = %(<blah_blah> 1 2 <dasherized_tag> 3 </dasherized_tag> </blah_blah>)

Calling Hash.from_xml with CamelCase element and attribute names causes the names to be underscored.

From http://www.bluequartz.us/phpBB2/...

xml = %(<blah_bla...

I think it's important to get these patches in, but there should be a follow up to deal with the potential unavailability of the session_id. This could be done ...

I think using ETag is problematic: We'd have to avoid header calculation and authentication detection until after the ETag's been generated: i.e., in an after_f...

Looking within ActionController, it looks like the session secret is an attribute of the CookieSessionStore, and not included in other session stores (e.g., mem...