> On Oct. 6, 2015, 2:39 p.m., Bernd Mathiske wrote:
> > Ship It!
> 
> Ben Mahler wrote:
>     I don't think we should introduce this into stout in its current form. I 
> realize you're planning to use this for authentication stuff, but looking at 
> this on its own, it seems like a confusing abstraction. Why would we couple 
> the notion of a Tree with the semantics around properties and property 
> inheritance?
> 
> Till Toenshoff wrote:
>     This code is being used in libprocess. So the options are libprocess or 
> stout for introducing it. I believe it would be a better fit for stout than 
> for libprocess as it is a data structure implementation that has no threading 
> or process specifics. Given that it already is in a reusable state, I think 
> that we should go as proposed by this RR.
> 
> Bernd Mathiske wrote:
>     @bmahler, what would you rather have us do? This structure is quite close 
> to Boost's PropertyTree 
> (http://www.boost.org/doc/libs/1_59_0/doc/html/property_tree.html), and we 
> would have loved to use that one. But it does not feature inheritance, which 
> seems a natural, essential thing to have in at least some trees.
> 
> Ben Mahler wrote:
>     Hm... the "inheritance" stuff that you guys need seems to just be a walk 
> of the tree from the root to the leaf you're interested in, collecting 
> properties of nodes along the way. Any reason you can't layer that 
> functionality on top or, even better, on the side via a free standing 
> function so that you don't need to introduce a new data structure?
> 
> Jie Yu wrote:
>     +1 if we can compose/inherit from boost property tree rather than 
> implementing our own.
> 
> Alexander Rojas wrote:
>     Property tree as of the moment is not distributed with Mesos. However, 
> there is a reason why it wasn't added, and it is that it will require a 
> double parsing of the path (one when asking the ptree, the other manually to 
> verify every subpath).
> 
> Ben Mahler wrote:
>     In other words, you were concerned about performance?

That and that pulling property tree from boost will include the whole boost 
spirit. Still, I will withdrawn this patch and follow your suggestion of puting 
this structure only where it will be used.


- Alexander


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/37996/#review101639
-----------------------------------------------------------


On Oct. 7, 2015, 11:54 a.m., Alexander Rojas wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/37996/
> -----------------------------------------------------------
> 
> (Updated Oct. 7, 2015, 11:54 a.m.)
> 
> 
> Review request for mesos, Adam B, Benjamin Hindman, Bernd Mathiske, and Till 
> Toenshoff.
> 
> 
> Bugs: MESOS-3231
>     https://issues.apache.org/jira/browse/MESOS-3231
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Introduces the `InheritanceTree` class which allows to create a tree where 
> nodes can be tag with some property. This property is then inherited by 
> children nodes.
> 
> Two behaviors are implemented, overriding, i.e. Adding a property to the 
> child node of another node with a property already will result in the 
> ancestor property being lost. The second behavior, accumulating, takes a 
> function and accumulates
> properties of all ancestors.
> 
> 
> Diffs
> -----
> 
>   3rdparty/libprocess/3rdparty/stout/Makefile.am 
> 76e1674e08bbe65a4fdf86731823a61f231d6d12 
>   3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
> 9e9c3119ad18f4cbc70c70095c71dc4fd19553df 
>   3rdparty/libprocess/3rdparty/stout/include/stout/inheritancetree.hpp 
> PRE-CREATION 
>   3rdparty/libprocess/3rdparty/stout/tests/CMakeLists.txt 
> 94292f8a46ec31bbaf6e52f48109322bbe123f70 
>   3rdparty/libprocess/3rdparty/stout/tests/inheritancetree_tests.cpp 
> PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/37996/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Alexander Rojas
> 
>

Reply via email to