Hi Juergen,

I see your point, and it makes sense.  One caviet though, I don't know if
this is the case or not but, I hope that when a )DUMP occurs the functions
and variables are sorted.  )DUMP makes more sense to use in a SCCS.  If I
change one function, I wouldn't want a different sort of the )DUMP file to
make the SCCS think more changed.  Does )DUMP sort its output?

Thanks!

Blake


On Sat, Sep 12, 2015 at 4:55 AM, Juergen Sauermann <
[email protected]> wrote:

> Hi Blake,
>
> yes, the discussion was to change *⎕TZ *on *)LOAD*. The general strategy
> of GNU APL to save
> as much information as possible because some information that looks like
> not being used today
> may become relevant at a later point in time.
>
> Removing *⎕TZ *would not help with source control system, because there
> is another timestamp
> in the workspace file that changes as well when the workspace is *)DUMP*ed
> or *)SAV**E*d. Also,
> the order of objects in a workspace file may change when a workspace is*
> )LOAD*ed and then
> *)SAVE*d again.
>
> Therefore a source control system would detect a change in the file even
> if we would not store *⎕TZ*.
> And I believe that is good because the state of the workspace has in fact
> changed, even though the
> variables or functions inside the workspace have not.
>
> /// Jürgen
>
>
>
>
> On 09/10/2015 04:48 AM, Blake McBride wrote:
>
> Sorry.  I now remember this is a topic we covered before and a change was
> made so the )LOAD ignored ⎕TZ.  That works.
>
> The thing that brought the problem up is because a garbage/unused ⎕TZ,
> although not loaded, is saved.  This creates problems with source code
> control systems because ⎕TZ may keep changing artificially when two
> developers are working on the same code but in different time zones.
>
> Bottom line - ⎕TZ should not be saved.
>
> Thanks.
>
> Blake
>
>
> On Wed, Sep 9, 2015 at 9:10 PM, Blake McBride <[email protected]> wrote:
>
>> Greetings,
>>
>> ⎕TZ should not be saved with SAVE or DUMP.  I should be able to count on
>> ⎕TZ containing the locally applicable time zone - not the one the developer
>> had!  If, for some reason the developer wants to change ⎕TZ, it should be
>> done explicitly in a function.  A global value should never be saved.
>>
>> Blake
>>
>>
>
>

Reply via email to