In bug 310359 and friends, a new All Addresses search/filter field is being added, that also includes search for BCC. This now works for code that uses the local nsIMsgDBHdr for filtering and search, but does not yet work for online IMAP search. Add the protocol support to IMAP to allow online search to support the BCC header, and hook that into the online search field.

It appears that searches with nsMsgSearchAttrib::AllAddresses fail currently because the switch in nsMsgSearchAdapter::EncodeImapTerm falls through to the default branch and fails the assertion. Is there anything I should be careful of when trying to implement this?

(In reply to comment #1) > > Is there anything I should be careful of when trying to implement this? If by "this" you mean this online BCC bug, let me first say that I have done no work on this yet, even though it is assigned to me (which means I decided to put my name on it after I created it). If you are interested in working on this, please feel free to take it. But, as you have noticed, it really needs doing soon, so only take it if you intend to work on it. As for what to watch for, as I recall my initial looks saw that there was tedious string manipulation in C++, so I made a mental note that it was going to be painful to work on it. But the functionality for BCC, I have been assured by bienvenu, is no different than other headers, so a parallel implementation for BCC should be straightforward. Gecko has their own string classes in C++ that would normally be used for this type of stuff. I can't recall whether the code in question uses the old C library strings, but depending on your reviewer you might be encouraged to implement anything new that you do in the newer Gecko string classes like nsCString.

I am definitely interested in working on this, as I feel it blocks bug 161338 which I've got a patch for that I'm sitting on. I intend to work on it, but don't feel comfortable reassigning the bug to myself until I have a working patch that I believe fixes it - I hope that's okay? I will look into the Gecko string classes - what's currently in nsMsgSearchAdapter::EncodeImapTerm disturbs me. Thanks for your guidance.

Since online search has very few (if any) enthusiasts, this is really not a very important bug, so I'm going to unassign myself from it.