On 27 February 2016 at 01:04, Dirk Hohndel <[email protected]> wrote:
> On Sat, Feb 27, 2016 at 12:58:29AM +0200, Lubomir I. Ivanov wrote:
>> On 27 February 2016 at 00:24, Dirk Hohndel <[email protected]> wrote:
>> > Both of the options you offer above sound like they have significany 
>> > drawbacks, but I'd really
>> > like to know how this would look from a user's perspective.
>>
>> the refactoring Tomaz did, introduced a minor regression in the
>> Grantlee variable syntax which is exposed to the user.
>> NOTE: the same applies to both weights and cylinders, but i'm going to
>> talk about weights (for the sake of example).
>>
>> latest public release (i think?) has these:
>> - dive.weights used to print all weight details
>> - dive.weight1-N prints a single weight detail
>>
>> master:
>> - dive.weights, should print all weight details again, but it doesn't
>> work for me
>> - dive.weight.1-N (notice the extra dot)
>>
>> the options are simple: fix dive.weights and 1) use the new syntax for
>> 1-N or 2) bring back the old syntax.
>> if we introduce the new syntax it has to be re-documented and it will
>> break at least for the user who originally requested the feature.
>>
>> what is currently is master is much more tidy, code wise and if we
>> bring back the old syntax it will make the code less pretty.
>
> I don't want to break this for the user, so my strong preference would be
> to go with the less pretty code that does what the users may already have
> in their template files.
>

ok, i understand.

Tomaz, for the variables not to break i think we need to introduce a
property and a getter method for every Grantlee variable, e.g.:
dive.weight1, dive.weight2...dive.weightN
dive.cylinder1, dive.cylinder2...dive.cylinderN

lubomir
--
_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Reply via email to