Re: DP6 more strict?

[Mark Wieder]

On 10/06/2015 12:07 AM, Mark Waddingham wrote:


My personal feeling (having observed people's interaction with the
language for a long time) is that variant forms of syntax which do the
same thing actually make things harder to learn and understand - it
makes the dictionary larger and increases the vocabulary for no real
benefit.


I believe Churchill said the same thing about Basic English .



Other forms which are no longer allowed are things like:

   repeat with x = 1 to 5 with messages

This has no function at all - the 'with messages' is ignored - indeed I
think a couple of people have found bugs in their scripts as a result.


Yeah. that would be me.
Thank you for fixing that one.
I do like having the compiler tell me when I've done something stupid.



Whilst strictness might reduce 'personal expression' to some extent, I
do think it helps the learning and remembering process. It means
everyone uses the same forms to do the same things - making reading each
others code, and helping each other out easier.


OK - playing Devil's Advocate here... I think one of the strengths of 
the xtalk language is that there may be many different paths to the 
solution of any given problem. I've learned a lot, and continue to do 
so, by seeing how other people approach issues differently from the way 
I would. So this artificial restriction cuts down on exploratory coding 
and places limits on the creative process of algorithm development.


--
 Mark Wieder
 ahsoftw...@gmail.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Bob Sneidar]

He also said, "But in the morning Madam, I will be sober."

Bob S


On Oct 6, 2015, at 08:52 , Mark Wieder 
> wrote:

My personal feeling (having observed people's interaction with the
language for a long time) is that variant forms of syntax which do the
same thing actually make things harder to learn and understand - it
makes the dictionary larger and increases the vocabulary for no real
benefit.

I believe Churchill said the same thing about Basic English .

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Waddingham]

On 2015-10-05 09:58, jameshale wrote:

Except the alternate syntax can no longer be handled.


The alternate syntaxes in many cases don't necessarily do what you 
think.


In your case:

repeat with x = 1 to 5

repeat for x = 1 to 5

Is certainly quit clear - one can immediately see what they should do 
quite unambiguously.


My personal feeling (having observed people's interaction with the 
language for a long time) is that variant forms of syntax which do the 
same thing actually make things harder to learn and understand - it 
makes the dictionary larger and increases the vocabulary for no real 
benefit.


Other forms which are no longer allowed are things like:

  repeat with x = 1 to 5 with messages

This has no function at all - the 'with messages' is ignored - indeed I 
think a couple of people have found bugs in their scripts as a result.


Whilst strictness might reduce 'personal expression' to some extent, I 
do think it helps the learning and remembering process. It means 
everyone uses the same forms to do the same things - making reading each 
others code, and helping each other out easier.


Whatever side of the fence one eventually chooses the docs need to 
match,
and if this stricter mode is kept some flag about it in both the docs 
and

the release notes is warranted.


I will happily hold my hands up and say that this was an element of 
social engineering on my part.


For a long time I've stopped any changes which make syntax stricter from 
being attended to - precisely because of the concern of breaking 
scripts. However, this position has always been based on abstract 
argument rather than direct data thus this time round I decided to let 
this change go through, without any sort of fanfare - just to see if it 
truly was a problem.


If the change is kept then we will make sure it is noted much more 
clearly in a future build.


Warmest Regards,

Mark.

--
Mark Waddingham ~ m...@livecode.com ~ http://www.livecode.com/
LiveCode: Everyone can create apps

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Waddingham]

On 2015-10-06 17:52, Mark Wieder wrote:

OK - playing Devil's Advocate here... I think one of the strengths of
the xtalk language is that there may be many different paths to the
solution of any given problem. I've learned a lot, and continue to do
so, by seeing how other people approach issues differently from the
way I would. So this artificial restriction cuts down on exploratory
coding and places limits on the creative process of algorithm
development.


This might well be true *if* there had been explicit decisions about the 
syntactic forms that were allowed. However, this is not the case - the 
fact the parser is lax is a side-effect of implementation rather than 
constructed intent.


Warmest Regards,

Mark.

--
Mark Waddingham ~ m...@livecode.com ~ http://www.livecode.com/
LiveCode: Everyone can create apps

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Wieder]

jameshale  writes:

> Except the alternate syntax can no longer be handled.

Right. I was simply saying that I think if the previous alternate was working
then maybe it would be best to revert to that state...

> Whatever side of the fence one eventually chooses the docs need to match,
> and if this stricter mode is kept some flag about it in both the docs and
> the release notes is warranted.

...and update the docs to reflect that.

But if not, then having the compiler catch the error is fine.

-- 
 Mark Wieder
 ahsoftw...@gmail.com



___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[jameshale]

Except the alternate syntax can no longer be handled.
Whatever side of the fence one eventually chooses the docs need to match,
and if this stricter mode is kept some flag about it in both the docs and
the release notes is warranted.



--
View this message in context: 
http://runtime-revolution.278305.n4.nabble.com/DP6-more-strict-tp4696868p4696888.html
Sent from the Revolution - User mailing list archive at Nabble.com.

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Wieder]

On 10/04/2015 07:21 AM, Mark Waddingham wrote:


I'm still somewhat on the fence about it but am intending to wait and see 
whether the change is going to cause more trouble than it's worth. We can 
always revert the patch in a subsequent build based on further feedback.


I'm on the fence about this one as well.

Although I'm really tending toward allowing the alternate syntax. If the 
engine's parser can handle it then it seems silly not to change the 
documentation to allow either form. Aren't we supposed to be about 
empowering users rather than dictating and restricting what can be done?


--
 Mark Wieder
 ahsoftw...@gmail.com

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Schonewille]

Hi,

Yes, I understand that the engine has become more strict. I believe it 
is a good thing, in most cases. If you notice that a plug-in doesn't 
work properly, please report it to the developer of the plug-in.


--
Mark Schonewille
http://economy-x-talk.com

Buy the most extensive book on the
LiveCode language:
http://livecodebeginner.economy-x-talk.com

Op 10/4/2015 om 15:31 schreef James Hale:

In running some stacks against DP6 I came across a script error on opening one.
The stack in question was simply a copy of a stack I have run without error 
since lc5.5.
The error was well and truly an error.
I had a repeat loop which I inadvertently entered as...
Repeat for X = 1 to aNumber
  
   End repeat

I guess it is a hang up from other languages I used in the past and I normally 
catch myself and enter the correct form...

Repeat with X = 1 to aNumber
  
   End repeat

However this time I didn't.
I hadn't since I first wrote this script way back then.

The point of this post is that not only did I miss this but Livecode missed it 
through every version until LC 8.

Try it out, put the first, incorrect form into a LC7 script and see if it 
compiles.
Now try the same in LC 8 dp6 or dp5

I can't speak for the other DP's as I have only thought to test this stack 
against the current release.

As has been posted some plugins are interfering with the correct function of the IDE and 
I wonder if this "stricter" (well more correct) behavior of the script engine 
in picking up errors is the cause. Perhaps there are quite a few scripts with errors that 
have slipped through in the past but will now stand out like the proverbial?

As an aside, I suggest you remove any plugins from your initial play with the 
latest DP, at least until you are happy with the behavior of the IDE just in 
case this new diligence on the part of the engine catches you out.

James



___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode



___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[Mark Waddingham]

I must confess this was one 'fix' which I was unsure about.

The LiveCode script parser is very very lax - you can write things which look 
like they should work in some fashion, but don't actually do what you think.

For a long time I have always had the point of view that such things should not 
be fixed until a suitable compatibility mechanism is in place, based on 
presumed impact.

However, this time, we decided to see what impact such a change (to the repeat 
command in this instance) would make (making decisions with actual data is 
generally better than without!).

As it turns out a couple of people have commented that it is a good thing as it 
has uncovered bugs in their scripts. A couple have found things have broken (we 
found a couple in the IDE as well) usually in third party components.

I'm still somewhat on the fence about it but am intending to wait and see 
whether the change is going to cause more trouble than it's worth. We can 
always revert the patch in a subsequent build based on further feedback.

Mark.

Sent from my iPhone

> On 4 Oct 2015, at 14:31, James Hale  wrote:
> 
> In running some stacks against DP6 I came across a script error on opening 
> one.
> The stack in question was simply a copy of a stack I have run without error 
> since lc5.5.
> The error was well and truly an error.
> I had a repeat loop which I inadvertently entered as...
>   Repeat for X = 1 to aNumber
> 
>  End repeat 
> 
> I guess it is a hang up from other languages I used in the past and I 
> normally catch myself and enter the correct form...
> 
> Repeat with X = 1 to aNumber
> 
>  End repeat 
> 
> However this time I didn't.
> I hadn't since I first wrote this script way back then.
> 
> The point of this post is that not only did I miss this but Livecode missed 
> it through every version until LC 8.
> 
> Try it out, put the first, incorrect form into a LC7 script and see if it 
> compiles.
> Now try the same in LC 8 dp6 or dp5
> 
> I can't speak for the other DP's as I have only thought to test this stack 
> against the current release.
> 
> As has been posted some plugins are interfering with the correct function of 
> the IDE and I wonder if this "stricter" (well more correct) behavior of the 
> script engine in picking up errors is the cause. Perhaps there are quite a 
> few scripts with errors that have slipped through in the past but will now 
> stand out like the proverbial?
> 
> As an aside, I suggest you remove any plugins from your initial play with the 
> latest DP, at least until you are happy with the behavior of the IDE just in 
> case this new diligence on the part of the engine catches you out.
> 
> James
> 
> 
> 
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription 
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode

Re: DP6 more strict?

[jameshale]

I agree with MarkS on this, it is a good thing.

I think I would have liked a warning though.

One sort of looks past what has been ok before in trying to see where an
error is.
A bit like the difficulty in proofreading something you have written. You
tend to "see" what you expect rather than what's there.

James



--
View this message in context: 
http://runtime-revolution.278305.n4.nabble.com/DP6-more-strict-tp4696868p4696882.html
Sent from the Revolution - User mailing list archive at Nabble.com.

___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription 
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode