Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.php > #5459

Re: OOP versus Procedural/Functional

Path csiph.com!usenet.pasdenom.info!news.albasani.net!fu-berlin.de!uni-berlin.de!not-for-mail
From "M. Strobel" <sorry_no_mail_here@nowhere.dee>
Newsgroups comp.lang.php
Subject Re: OOP versus Procedural/Functional
Date Wed, 11 Apr 2012 19:09:05 +0200
Lines 46
Message-ID <9uls5hFufuU1@mid.uni-berlin.de> (permalink)
References <alpine.LNX.2.00.1204091428210.3668@mobile_yoda.darkside> <jm3p6n02qe@news1.newsguy.com>
Mime-Version 1.0
Content-Type text/plain; charset=UTF-8
Content-Transfer-Encoding 7bit
X-Trace news.uni-berlin.de cIKXXR0rwDsOcALoppg7EQZAGtRiwISg5OwxMw56rCH1MP0Mo=
User-Agent Mozilla/5.0 (X11; U; Linux x86_64; de; rv:1.9.2.28) Gecko/20120306 Thunderbird/3.1.20
In-Reply-To <jm3p6n02qe@news1.newsguy.com>
Xref csiph.com comp.lang.php:5459

Show key headers only | View raw


Am 11.04.2012 13:16, schrieb crankypuss:
> On 04/09/2012 01:35 PM, Mr. B-o-B wrote:
>> I was curious about peoples thoughts in regards to OOP with PHP.
>>
>> Are there pro's/con's of writing OOP code versus procedural/functional
>> coding?
>>
>> Strictly speaking for PHP, I do the bulk of my coding as
>> procedural/functional. I really haven't found any added benefit to going
>> the OOP route.
>>
>> Am I missing something, or is it just a different approach to the same
>> problem?
> 
> OOP is basically a good idea carried to an extreme.
> 
> There is a boatload of value in the concept of inheritance, but that's largely
> negated by the mechanics of how inheritance is done in OOP languages.
> 
> Polymorphism is something that's more in the eye of the programmer than in the
> language, data constructs can be specific or generalized.
> 
> The idea that making something into an object gives it superior robustness is a joke
> if you can step back and look at it.
> 
> I remember writing a Java application and trying to derive an actually useful string
> class from the garbage they had declared "final" and that was it for me and Java.  I
> cannot begin to count the times when I've had to hoke up an object pointer to use
> some simple processing method that was part of a C++ class.
> 
> Of course that's all personal opinion.  I avoid all OOP constructs in PHP wherever I
> can, the more closely I can make it resemble C the easier it will be to convert it
> when the time comes.  There is one exception to that in my code, one of the zip
> functionalities is only supported through some objects... that stuff is encapsulated
> in a subroutine.
> 
> Since I value inheritance I've build myself a layered methodology that lets me
> override various functions by adding or inserting different layers at runtime.
> 
> Perhaps the nicest thing about programming is that it puts the lie to religious
> adherences... as one of my college instructors put it, "if your program works for all
> cases it is by definition correct".

Second phrase could have been "But there are many more aspects of software quality".

/Str.

Back to comp.lang.php | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread


Thread

OOP versus Procedural/Functional "Mr. B-o-B" <mr.chew.baka@gmail.com> - 2012-04-09 14:35 -0500
  Re: OOP versus Procedural/Functional "J.O. Aho" <user@example.net> - 2012-04-09 21:59 +0200
    Re: OOP versus Procedural/Functional The Natural Philosopher <tnp@invalid.invalid> - 2012-04-09 22:46 +0100
    Re: OOP versus Procedural/Functional Michael Vilain <vilain@NOspamcop.net> - 2012-04-09 19:29 -0700
  Re: OOP versus Procedural/Functional Jerry Stuckle <jstucklex@attglobal.net> - 2012-04-09 16:55 -0400
  Re: OOP versus Procedural/Functional "M. Strobel" <sorry_no_mail_here@nowhere.dee> - 2012-04-10 00:34 +0200
  Re: OOP versus Procedural/Functional Goran <goran@nospam.com> - 2012-04-10 10:14 +0200
    Re: OOP versus Procedural/Functional "M. Strobel" <sorry_no_mail_here@nowhere.dee> - 2012-04-10 12:06 +0200
    Re: OOP versus Procedural/Functional The Natural Philosopher <tnp@invalid.invalid> - 2012-04-10 13:04 +0100
      Re: OOP versus Procedural/Functional Tim Streater <timstreater@greenbee.net> - 2012-04-10 16:07 +0100
        Re: OOP versus Procedural/Functional The Natural Philosopher <tnp@invalid.invalid> - 2012-04-10 16:28 +0100
        Re: OOP versus Procedural/Functional Jerry Stuckle <jstucklex@attglobal.net> - 2012-04-10 12:08 -0400
        Re: OOP versus Procedural/Functional "M. Strobel" <sorry_no_mail_here@nowhere.dee> - 2012-04-10 19:14 +0200
          Re: OOP versus Procedural/Functional The Natural Philosopher <tnp@invalid.invalid> - 2012-04-10 18:19 +0100
  Re: OOP versus Procedural/Functional Michael Fesser <netizen@gmx.de> - 2012-04-10 14:49 +0200
  Re: OOP versus Procedural/Functional "Álvaro G. Vicario" <alvaro.NOSPAMTHANX@demogracia.com.invalid> - 2012-04-10 17:29 +0200
    Re: OOP versus Procedural/Functional The Natural Philosopher <tnp@invalid.invalid> - 2012-04-10 16:34 +0100
  Re: OOP versus Procedural/Functional crankypuss <no@email.thanks> - 2012-04-11 05:16 -0600
    Re: OOP versus Procedural/Functional Goran <goran@nospam.com> - 2012-04-11 16:25 +0200
      Re: OOP versus Procedural/Functional Jerry Stuckle <jstucklex@attglobal.net> - 2012-04-11 13:06 -0400
        Re: OOP versus Procedural/Functional Tim Streater <timstreater@greenbee.net> - 2012-04-11 18:18 +0100
          Re: OOP versus Procedural/Functional Jerry Stuckle <jstucklex@attglobal.net> - 2012-04-11 16:24 -0400
            Re: OOP versus Procedural/Functional Tim Streater <timstreater@greenbee.net> - 2012-04-11 22:43 +0100
      Re: OOP versus Procedural/Functional crankypuss <no@email.thanks> - 2012-04-12 05:54 -0600
    Re: OOP versus Procedural/Functional "M. Strobel" <sorry_no_mail_here@nowhere.dee> - 2012-04-11 19:09 +0200
      Re: OOP versus Procedural/Functional crankypuss <no@email.thanks> - 2012-04-12 05:56 -0600
        Re: OOP versus Procedural/Functional "M. Strobel" <sorry_no_mail_here@nowhere.dee> - 2012-04-12 17:33 +0200
          Re: OOP versus Procedural/Functional Jerry Stuckle <jstucklex@attglobal.net> - 2012-04-12 13:18 -0400

csiph-web