On 7/30/2014 8:47 PM, Alexander Bogdashevsky wrote:
On 7/29/2014 8:15 PM, Alexander Bogdashevsky wrote:
Not sure if this is a show-stopper or not, but property_tree API was broken a little. My fault.
https://svn.boost.org/trac/boost/ticket/10272
I am working on a fix right now and should be done today-tomorrow.
What is "broken a little"? Can you help us to assess the seriousness of the regression? Does it break common usage scenarios?
Yes, it does break common usage scenario. See ticket https://svn.boost.org/trac/boost/ticket/10272 for example.
I just pushed a fix for this issue https://github.com/sanderbog/property_tree/commit/4e16d57fa634fc9ddb64cde1c6... Could somebody with good enough privileges merge it?
This fix hasn't even been merged to develop yet, and we're trying to get out a second RC. (The pull request(*) looks very dirty; can you clean it up so it only has the commit you want merged?) I think it's missed the boat. Besides, the test results look good on both develop and master, and the Property_tree maintainer hasn't had a chance to weigh in. This might be an intentional change -- I can't say at this point. Sebastian, can you have a look at the bug report and the patch and say what you think? (Even if it's a good fix, I think it has to wait at this point.) Eric (*) https://github.com/boostorg/property_tree/pull/5