This is an automatically generated e-mail. To reply, visit:

Good stuff!
Sorry for the delay in reviewing this: only got halfway through - will finish 
it off on the train.

support/cpplint.py (line 4734)

    function names should be `lower_case` according to 

support/cpplint.py (lines 4741 - 4745)

    can we use a format that Python automated doc tools will understand?
    `@param` / `@type` / `@return`
    or the `:param` equivalent.

support/cpplint.py (line 4763)

    not your code, admittedly, but it would be awesome to drag our Python 
scripts into the 21st century and use `argparse` instead :)

support/cpplint.py (line 4766)

    this does not make any sense at all, IMO
    (do we really have a wrapper script that checks `$?` and echoes the number 
of errors?)

support/mesos-style.py (line 11)

    I am not familiar with `cStringIO` - the name would seem to indicate (a) 
it's not a standard library and (b) that it requires some amount of C 
    This is usually a pain "on some OS" (eg, OSX) - can you please confirm 
whether it's in the standard libraries and, if not, what are the steps involved 
in getting it installed on various OSes?
    This may break building/committing everywhere, right?

support/mesos-style.py (line 63)

    I have a suspicion it's just *one* OS that does not support this, but lets' 
not be picky here :)

support/mesos-style.py (lines 71 - 76)

    same comment as before

support/mesos-style.py (line 97)

    unnecessary parentheses

- Marco Massenzio

On Nov. 16, 2015, 9:25 a.m., Alex Clemmer wrote:
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40340/
> -----------------------------------------------------------
> (Updated Nov. 16, 2015, 9:25 a.m.)
> Review request for mesos, Artem Harutyunyan, Michael Hopcroft, Joris Van 
> Remoortere, Joseph Wu, and Marco Massenzio.
> Repository: mesos
> Description
> -------
> Windows: Added Windows support to `support/post-reviews.py`.
> Early draft of the update to the Windows-compatible `post-reviews.py`. I'd 
> like to get early feedback from people in case this is entirely the wrong 
> direction.
> At the outset, I will say that the "right way" to do this would be to 
> refactor `cppylint.py` and `post-reviews.py` entirely so that they are more 
> modular (for example, make `post-reviews.py` function-oriented, instead of a 
> pile of statements and expressions, and `cpplint.py` should be a module 
> proper). I didn't do this the "right way" because it is blocking for our 
> Windows friends. If it's really important to do this the "right way", we may 
> need to break it into a couple reviews.
> Note that the most questionable part of this is the point at which we 
> redirect `stderr` in `cpplint.py` to a string capture it for use in 
> `post-reviews.py`. This is certainly the quickest way to do it, but it is not 
> the best -- the best way would be to refactor `cpplint.py` to be a module, at 
> least, and to handle error logging in a more pluggable way. We aimed for the 
> shortest diff because there are no tests for any of these scripts, and to be 
> honest, I was afraid of changing them without tests.
> Diffs
> -----
>   .gitignore-template 90b6697d19a5e0a68805b23b587b362731a1df25 
>   support/cpplint.py 6890e27f92603b025e25e4db01decf351c33c9a1 
>   support/mesos-style.py 66b45692c3c04f68358b63d52e4d87934f241bd7 
>   support/post-reviews.py 170be83aa6dca6e8175292169d78e8f7915f7e6e 
> Diff: https://reviews.apache.org/r/40340/diff/
> Testing
> -------
> Thanks,
> Alex Clemmer

Reply via email to