chemistry-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sutter, Peter" <>
Subject Re: Browser binding re-factoring
Date Thu, 01 May 2014 14:53:00 GMT
Hi Gavin,

I just uploaded the current state of my implementation:

Now all methods of the interfaces should be implemented. I have also
tested the browser binding implementation in our app and it works very
nicely as far as I can see.

There is maybe one area where I¹m not really sure because the
CMISHttpUploadRequest class contains atom pub specific coding
(prepareXMLWithCMISProperties) which should probably be moved down to the
respective atom class.
And still there is the JSONSerializer bug with very large/small double
values where we need to decide how we deal with it.

You can start with your part 2 refactoring, no problem. Will only do minor
changes in the next working days, if any.

Best regards,

On 5/1/14, 11:03 AM, "Gavin Cornwell" <>

>Hi Peter,
>Great work on the browser binding so far, it¹s really coming together.
>As you¹ve probably seen by now, I did part 1 of the re-factoring we
>agreed upon previously (updating property names).
>Part 2 is going to be a bit more disruptive as it¹s going to involve
>renaming and moving quite a few classes (ensure all Atom classes are
>named as such). So I thought it would be best to co-ordinate this with
>you to avoid a merge headache for you and I.
>I was planning on doing this tomorrow, do you think you¹ll have your
>latest set of changes committed by then?
>As a side note, I looked into the UriBuilder classes last night and
>found, as you¹d hinted at, they are pretty specific to the AtomPub
>binding so I¹m going to rename them as such and move the constants from
>the generic CMISBindingSession class to the CMISAtomPubConstants class.
>Following this, I¹ll be available to help out with the actual
>implementation, if there are any areas you¹d like me to look at (that are
>not on your critical path) please let me know.

View raw message