[Gambas-user] C like #include for Gambas

[Jussi Lahtinen]

OK, while that solution might work, it sucks.
Instead see new attachment (IncludeProblem-2.0.tar.gz).

Jussi



On 9 June 2012 16:57, Jussi Lahtinen <jussi.lahtinen at ...626...> wrote:

> Thinking more closely your problem, I think I know what you need.
>
> You try to avoid writing *many* lines like this;
> sParameterNameArray = NEW String[MAXIMUM_CSV_PARAMETERS]
>
> That is caused by broken object design, usually because you need to use
> old code which had different approach than OO.
> I have had same problems.
>
> I attached my suggestion how to implement this
> (IncludeProblem-0.0.1.tar.gz).
> Is it working solution?
>
> Jussi
>
>
>
>
>
>
> On 9 June 2012 16:18, Jussi Lahtinen <jussi.lahtinen at ...626...> wrote:
>
>>
>> C thrives on #include without showing signs of design problems.
>>>
>>
>> Yes, I know, I wasn't talking about C, I was talking about object
>> oriented language, Gambas.
>>
>>
>>
>>
>>> Tens of thousands of external lines of code can be pulled in with
>>> a dozen #include statements, leaving the code completely manageable,
>>> readable and a lot more powerful than any other competing method.
>>>
>>
>> Yes, but you use #include with libraries, not with programs that aren't
>> designed for that.
>> Or you have design problem.
>>
>>
>>
>> For example, in module_a.module you might have
>>>
>>>    ...
>>>
>>>
>>> and in a file called initialise_csv_variables.include it would contain:
>>>
>>>  PRIVATE WordMarker1 AS Integer
>>>  :
>>>  '200 further variable declarations and initialisations
>>>  :
>>>  PRIVATE sParameterNameArray AS NEW String[MAXIMUM_CSV_PARAMETERS]
>>>  PRIVATE sParameterValueArray AS NEW String[MAXIMUM_CSV_PARAMETERS]
>>>
>>> The #include files can contain any amount of well trodden
>>> proven gambas code which is maintained in only one file.
>>>
>>
>> So, basically you would add module/class called initialise_csv_variables,
>> and you use that module/class in modules a and b.
>>
>>
>>
>>
>>> I will argue that adding C like #include functionality saves a ton of
>>> work in mega programs which are already working well, but now
>>> needs to be expanded even further.
>>>
>>
>> I don't understand how.
>> You would still need to write that initialise_csv_variables.include file,
>> you could as well write it as module/class.
>>
>> Jussi
>>
>>
>>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IncludeProblem-2.0.tar.gz
Type: application/x-gzip
Size: 5031 bytes
Desc: not available
URL: <http://lists.gambas-basic.org/pipermail/user/attachments/20120609/55418a18/attachment.bin>