From: Marty K. <mrk...@co...> - 2012-03-19 13:25:48
|
On 03/16/2012 10:17 AM, White, Greg wrote: > Marty, David, > > Could you remind me of the status of work on unsigned? I was madly trying to do NTypes and service interfaces > when David propsoal came in. The Feb 15th Meeting minutes say Marty will prepare code based on his proposal, and David will prepare a refactor of that when it's complete so we can compare. > > Is that still the plan? So Marty are you now implementing now what is in [1] or have you finished I've missed the notification email? Implementing unsigned impacts a lot of code. There also seems to be agreement that fieldName should be moved from Field to Structure. Matej and I have also discussed if the C++ implementation should expose all subfield info via shared pointers. Since all impact a lot of of code all should be done at the same time. In my working version of pvDataWWW I have moved unsigned.html to pvDataPVAccessChanges.html. It describes all the proposed changes. I do not want to commit this change until the problem of the two heads is resolved so I have attached it. Another issue is how do Matej and I make these changes without impacting everyone else? So far I don't think any of us have worked on mercurial branches on purpose. So what will the procedures be for braches? Marty > David, can you put your design proposal doc in pvDataWWW/alpha/ so we can keep a copy on the web? > > Cheers > Greg > > > [0] http://sourceforge.net/mailarchive/message.php?msg_id=28839054 > [1] http://epics-pvdata.hg.sourceforge.net/hgweb/epics-pvdata/pvDataWWW/raw-file/tip/mainPage/alpha/unsigned.html > |