On Thu, 22 Dec 2016 18:35:24 +0100
Pino Toscano <ptoscano(a)redhat.com> wrote:
 On Sunday, 18 December 2016 23:16:31 CET Tomáš Golembiovský wrote:
 > Added two new optional arguments to nsplit:
 >
 > * noempty: if set to false empty elements are not stored in the returned
 >   list. The default is to keep the empty elements
 >
 > * count: specifies how many splits to perform; negative count
 >   (the default) means do as many splits as possible
 >
 > Added tests for nsplit.
 >
 > Signed-off-by: Tomáš Golembiovský <tgolembi(a)redhat.com>
 > ---
 >  mllib/common_utils.ml       | 12 +++++++++---
 >  mllib/common_utils.mli      | 12 ++++++++++--
 >  mllib/common_utils_tests.ml | 26 ++++++++++++++++++++++++++
 >  3 files changed, 45 insertions(+), 5 deletions(-)
 >
 > diff --git a/mllib/common_utils.ml b/mllib/common_utils.ml
 > index f948dce..b6f3046 100644
 > --- a/mllib/common_utils.ml
 > +++ b/mllib/common_utils.ml
 > @@ -130,15 +130,21 @@ module String = struct
 >        done;
 >        if not !r then s else Bytes.to_string b2
 >  
 > -    let rec nsplit sep str =
 > +    let rec nsplit ?(noempty = true) ?(count = -1) sep str =  
 Shouldn't the default value of noempty be false, matching the current
 behaviour, and the documentation added? 
It seems I managed to confuse myself about what noempty actually means
(keep or drop). It's more than obvious from the fact that the
description in commit message does not match the documentation. 
I will rename the argument to keep_empty to make it more clear.
    Tomas
-- 
Tomáš Golembiovský <tgolembi(a)redhat.com>