On Tue, Aug 6, 2013 at 7:07 PM, Mark Hahn <m...@reevuit.com> wrote:

> Also, the constructor problem would break immediately, not at some random
> later time.


Can you explain how any or all of these being generator functions:

a()
b()
c()

... will break at some random later time? Assuming that those three calls
represent some user code.


Rick






>
>
> On Tue, Aug 6, 2013 at 4:06 PM, Mikeal Rogers <mikeal.rog...@gmail.com>wrote:
>
>>
>> On Aug 6, 2013, at 3:57PM, Rick Waldron <waldron.r...@gmail.com> wrote:
>>
>>
>>
>>
>> On Tue, Aug 6, 2013 at 6:17 PM, Mark Hahn <m...@reevuit.com> wrote:
>>
>>> > Always assume that you will be suspended any time you have a yield or
>>> yield*
>>>
>>> How do you know any of this you see a function call?  Only the function
>>> definition looks different.  Tell me which is these will yield ..
>>>
>>> a()
>>> b()
>>> c()
>>>
>>
>> I see no reason why it would matter if any or all of these are generator
>> functions. How is this argument any different then calling constructor
>> functions with new or not?
>>
>> Tell me which of these is a constructor that must be called with `new`:
>>
>> a()
>> b()
>> c()
>>
>>
>> None of them, they aren't capitalized. This is a trick question right?
>>
>> :P
>>
>>
>>
>> Rick
>>
>>
>>
>>
>>
>>>
>>> On Tue, Aug 6, 2013 at 2:54 PM, Tim Caswell <t...@creationix.com> wrote:
>>>
>>>>
>>>>
>>>>
>>>> On Tue, Aug 6, 2013 at 4:47 PM, Mikeal Rogers 
>>>> <mikeal.rog...@gmail.com>wrote:
>>>>
>>>>> It is the *exact same* hazard as the code you posted, I don't think it
>>>>> is any *more* of a hazard at the language level. The difference is that 
>>>>> the
>>>>> pattern in the function code below is basically never seen in node.js code
>>>>> and it has been *possible* to write it since node's creation.
>>>>>
>>>>> Could be that nobody will use/clobber closure scope in their
>>>>> generators either, but I'm afraid they will because it will work and
>>>>> operate as expected *most* of the time and, unlike the function example
>>>>> you've posted, it will still fit well when used with what I assume will
>>>>> become a healthy ecosystem of generator consuming libraries.
>>>>>
>>>>> The reason nobody writes this kind of thing in node is that it just
>>>>> won't play well with the rest of the libraries in the ecosystem whereas 
>>>>> the
>>>>> hazardous generator example looks like it would fit just as well as a 
>>>>> "well
>>>>> written" generator until it hits scale and causes visible bugs.
>>>>>
>>>>> All kinds of bad patterns and bad code are possible in any language.
>>>>> The patterns we adopt to create a healthy ecosystem are what will keep
>>>>> people away from the third rails. Depending on people to have "well
>>>>> written" code, by some definition of "well", is not a good alternative to
>>>>> solidifying and encouraging patterns that make it *difficult* to write
>>>>> hazardous code in the first place.
>>>>>
>>>>> No pattern or feature in JavaScript escapes the potential for abuse
>>>>> and bugs, and nothing that I'm worried about w/ respect to generators is
>>>>> worse or better than language level hazards w/ callbacks, it's just that
>>>>> we've created a set of patterns that *discourage* those hazards in the
>>>>> current ecosystem and I'm failing to see how we do that in this particular
>>>>> case with generators.
>>>>>
>>>>
>>>> I propose a simple rule for all generator code:
>>>>
>>>>  - Always assume that you will be suspended any time you have a yield
>>>> or yield* and any shared state may change.
>>>>
>>>> If we educate everyone and focus on that one thing (yield) that will
>>>> cover all hazards around shared state and generators, be it async
>>>> promise/continuable based stuff or sync lazy iterator stuff.
>>>>
>>>> Just like with functions we've learned to look for the "function"
>>>> keyword and guard our state around that point, we need to do the same
>>>> around the "yield" keyword.
>>>>
>>>> Teach people correct principles and trust them to do the right thing.
>>>>  We'll be fine.
>>>>
>>>>
>>>>
>>>>>
>>>>> -Mikeal
>>>>>
>>>>> On Aug 6, 2013, at 2:33PM, Tim Caswell <t...@creationix.com> wrote:
>>>>>
>>>>> My question to people on the list is how is this any different than
>>>>> the same hazard with normal function?
>>>>>
>>>>>     var sum, i;
>>>>>     function reduce(arr) {
>>>>>       sum = 0;
>>>>>       i = 0;
>>>>>       return function () {
>>>>>         if (i < arr.length) {
>>>>>           sum += arr[i++];
>>>>>           return sum;
>>>>>         }
>>>>>       };
>>>>>     }
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> --
>>>>> Job Board: http://jobs.nodejs.org/
>>>>> Posting guidelines:
>>>>> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "nodejs" group.
>>>>> To post to this group, send email to nodejs@googlegroups.com
>>>>> To unsubscribe from this group, send email to
>>>>> nodejs+unsubscr...@googlegroups.com
>>>>> For more options, visit this group at
>>>>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>>>>
>>>>> ---
>>>>> You received this message because you are subscribed to the Google
>>>>> Groups "nodejs" group.
>>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>>> an email to nodejs+unsubscr...@googlegroups.com.
>>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> --
>>>> Job Board: http://jobs.nodejs.org/
>>>> Posting guidelines:
>>>> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
>>>> You received this message because you are subscribed to the Google
>>>> Groups "nodejs" group.
>>>> To post to this group, send email to nodejs@googlegroups.com
>>>> To unsubscribe from this group, send email to
>>>> nodejs+unsubscr...@googlegroups.com
>>>> For more options, visit this group at
>>>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>>>
>>>> ---
>>>> You received this message because you are subscribed to the Google
>>>> Groups "nodejs" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to nodejs+unsubscr...@googlegroups.com.
>>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>>
>>>>
>>>>
>>>
>>>
>>> --
>>> --
>>> Job Board: http://jobs.nodejs.org/
>>> Posting guidelines:
>>> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
>>> You received this message because you are subscribed to the Google
>>> Groups "nodejs" group.
>>> To post to this group, send email to nodejs@googlegroups.com
>>> To unsubscribe from this group, send email to
>>> nodejs+unsubscr...@googlegroups.com
>>> For more options, visit this group at
>>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>>
>>> ---
>>> You received this message because you are subscribed to the Google
>>> Groups "nodejs" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to nodejs+unsubscr...@googlegroups.com.
>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>>
>>>
>>
>>
>> --
>> --
>> Job Board: http://jobs.nodejs.org/
>> Posting guidelines:
>> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
>> You received this message because you are subscribed to the Google
>> Groups "nodejs" group.
>> To post to this group, send email to nodejs@googlegroups.com
>> To unsubscribe from this group, send email to
>> nodejs+unsubscr...@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "nodejs" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to nodejs+unsubscr...@googlegroups.com.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>
>>
>>
>>
>>  --
>> --
>> Job Board: http://jobs.nodejs.org/
>> Posting guidelines:
>> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
>> You received this message because you are subscribed to the Google
>> Groups "nodejs" group.
>> To post to this group, send email to nodejs@googlegroups.com
>> To unsubscribe from this group, send email to
>> nodejs+unsubscr...@googlegroups.com
>> For more options, visit this group at
>> http://groups.google.com/group/nodejs?hl=en?hl=en
>>
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "nodejs" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to nodejs+unsubscr...@googlegroups.com.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>
>>
>>
>
>  --
> --
> Job Board: http://jobs.nodejs.org/
> Posting guidelines:
> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
> You received this message because you are subscribed to the Google
> Groups "nodejs" group.
> To post to this group, send email to nodejs@googlegroups.com
> To unsubscribe from this group, send email to
> nodejs+unsubscr...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/nodejs?hl=en?hl=en
>
> ---
> You received this message because you are subscribed to the Google Groups
> "nodejs" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to nodejs+unsubscr...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

-- 
-- 
Job Board: http://jobs.nodejs.org/
Posting guidelines: 
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to nodejs@googlegroups.com
To unsubscribe from this group, send email to
nodejs+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

--- 
You received this message because you are subscribed to the Google Groups 
"nodejs" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to nodejs+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to