Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:58259
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain punkave.com designates 209.85.210.170 as permitted sender)
Received-SPF: pass (google.com: domain of tom@punkave.com designates 10.50.216.231 as permitted sender) client-ip=10.50.216.231;
MIME-Version: 1.0
In-Reply-To: <CAGSj+QvMFBTqsF1_Oyg7tmvH1d4Vh1W2iVvLngivbhE35J4NCg@mail.gmail.com>
References: <01a901ccf622$0645f230$12d1d690$@alliantinternet.com>
	<CAGSj+QtVWzc50kZX=0q-CRukd73eZ1T0B9D8t7EejXMw_9Q4ZQ@mail.gmail.com>
	<fd6fc5212ceab7a265cb505039373a0d.squirrel@www.l-i-e.com>
	<CAL0xaBHDu3xNwdO7XwCRObXTZd_X192KgxAoFH_ZH2TTsPGZ6Q@mail.gmail.com>
	<CAGSj+QuiRojiYnzbE495EGaS_ZjBZF6y47SVL6tBFOBkFKtp+A@mail.gmail.com>
	<CAORXhGJuhaLMvmu0_6fH7zZTXEJyrgwoMBRnBQi-7XSrtKX-_w@mail.gmail.com>
	<CAGSj+QswHaj+SjJRo3_s7Ns0gH3mUb7_W3Xw2zyNXGbnBPkK4g@mail.gmail.com>
	<CAGSj+QvMFBTqsF1_Oyg7tmvH1d4Vh1W2iVvLngivbhE35J4NCg@mail.gmail.com>
Date: Tue, 28 Feb 2012 12:18:09 -0500
Message-ID: <CAORXhGJTk+X1fvn_2xq3bO_7=E=N9_G3gCyqAPtFYFOosEMo3A@mail.gmail.com>
To: Kiall Mac Innes <kiall@managedit.ie>
Cc: Arvids Godjuks <arvids.godjuks@gmail.com>, Richard Lynch <ceo@l-i-e.com>, internals@lists.php.net
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PHP-DEV] Possibility to add finally to try/catch?
From: tom@punkave.com (Tom Boutell)

OK, consider my point successfully buried in counterexamples (:

On Tue, Feb 28, 2012 at 12:02 PM, Kiall Mac Innes <kiall@managedit.ie> wrot=
e:
> An an equivalent snippet of ruby..
>
> $ irb
> irb(main):001:0> def divide(x, y)
> irb(main):002:1> =A0 =A0 begin
> irb(main):003:2* =A0 =A0 =A0 =A0 result =3D x / y
> irb(main):004:2> =A0 =A0 =A0 =A0 puts "result is", result
> irb(main):005:2> =A0 =A0 rescue ZeroDivisionError
> irb(main):006:2> =A0 =A0 =A0 =A0 puts "division by zero!"
> irb(main):007:2> =A0 =A0 =A0 =A0 raise Exception('division by zero')
> irb(main):008:2> =A0 =A0 ensure
> irb(main):009:2* =A0 =A0 =A0 =A0 puts "executing finally clause"
> irb(main):010:2> =A0 =A0 end
> irb(main):011:1> end
> =3D> nil
> irb(main):012:0>
> irb(main):013:0*
> irb(main):014:0* divide(2, 1)
> result is
> 2
> executing finally clause
> =3D> nil
> irb(main):015:0>
> irb(main):016:0* divide(2, 0)
> division by zero!
> executing finally clause=A0<=3D=3D=3D=3D=3D=3D The interesting part
> NoMethodError: undefined method `Exception' for main:Object
> from (irb):7:in `divide'
> from (irb):16
> from :0
> irb(main):017:0>
> irb(main):018:0* divide("2", "1")
> executing finally clause
> NoMethodError: undefined method `/' for "2":String
> from (irb):3:in `divide'
> from (irb):18
> from :0
>
> Thanks,
> Kiall
>
>
>
> On Tue, Feb 28, 2012 at 4:54 PM, Kiall Mac Innes <kiall@managedit.ie> wro=
te:
>>
>> On Tue, Feb 28, 2012 at 4:48 PM, Tom Boutell <tom@punkave.com> wrote:
>>>
>>> On Tue, Feb 28, 2012 at 11:32 AM, Kiall Mac Innes <kiall@managedit.ie>
>>> wrote:
>>> > Yes, You could abstract the try/catch into a new (and un-needed)
>>> > function
>>> > to try and emulate the behavior of finally.. Unless, for example, you
>>> > re-throw the exception after logging in the catch.
>>>
>>> 'finally' doesn't run for stuff that throws an exception not caught in
>>> the try { } block, or an exception thrown again in the catch { } block
>>> - does it?
>>>
>>> I would hope not, since that means "something this block of code did
>>> not anticipate at all - another sort of exceptional situation
>>> altogether" and really should not run any more local code, nothing
>>> until and unless there is a catch block somewhere further up that does
>>> catch that exception.
>>
>>
>> I would indeed expect the finally to run regardless of what happens in t=
he
>> catch block.
>>
>> $ python
>> >>> def divide(x, y):
>> ... =A0 =A0 =A0try:
>> ... =A0 =A0 =A0 =A0 =A0result =3D x / y
>> ... =A0 =A0 =A0except ZeroDivisionError:
>> ... =A0 =A0 =A0 =A0 =A0print "division by zero!"
>> ... =A0 =A0 =A0 =A0 =A0raise Exception('division by zero')
>> ... =A0 =A0 =A0else:
>> ... =A0 =A0 =A0 =A0 =A0print "result is", result
>> ... =A0 =A0 =A0finally:
>> ... =A0 =A0 =A0 =A0 =A0print "executing finally clause"
>> ...
>> >>> divide(2, 1)
>> result is 2
>> executing finally clause
>> >>>
>> >>> divide(2, 0)
>> division by zero!
>> executing finally clause=A0<=3D=3D=3D=3D=3D=3D The interesting part
>> Traceback (most recent call last):
>> =A0 File "<stdin>", line 1, in <module>
>> =A0 File "<stdin>", line 6, in divide
>> Exception: division by zero
>> >>>
>> >>> divide("2", "1")
>> executing finally clause
>> Traceback (most recent call last):
>> =A0 File "<stdin>", line 1, in <module>
>> =A0 File "<stdin>", line 3, in divide
>> TypeError: unsupported operand type(s) for /: 'str' and 'str'
>
>



--=20
Tom Boutell
P'unk Avenue
215 755 1330
punkave.com
window.punkave.com