[GRLUG] python app speed

[Michael Mol]

On Thu, Jul 16, 2009 at 5:42 PM, Adam Tauno
Williams<awilliam at whitemice.org> wrote:
> On Thu, 2009-07-16 at 16:09 -0400, Michael Mol wrote:
>
> Absolutely;  but when you choose a managed or quasi-managed language
> you're buying the car!  You get what you got under the hood,  bugs and
> all [as well as advantages].  And there are bugs,  especially in the
> case of PHP.  I feel like I could give a "101 ways to crash PHP using
> perfect valid language constructs" presentation - although I wouldn't
> want to.

Actually, an ACID test for the language (or any language; Something
like FATE, but for a particular programming language) wouldn't be a
bad idea, IMO.

My favorite was when I filed a bug report because of inconsistent
syntax behavior.  IIRC, something in the following example set fails
to compile, while syntactically similar constructs do.  In the
following examples, groups 1.1.x and 2.x.x worked, but 1.2.x failed to
compile.  The reason is obvious; The curly braces in groups 2.2.x
which denote code blocks changed the way the PHP interpreter
considered "?> (greeting) <?".  But it's still inconsistent syntax
behavior.

$a = true; // Or false, your pick.

// Group 1.1.1
if($a)
  echo "Hi, there";
else
  echo "Ho, there";

// Group 1.1.2
if($a)
  echo "Hi, there";

// group 1.2.1
if($a)
  ?> Hi, there <?
else
  ?> Ho, there <?

// group 1.2.2
if($a)
  ?> Hi, there <?

// group 2.1.1
if($a)
{
  echo "Hi, there";
}
else
{
  echo "Ho, there";
}

// group 2.1.2
if($a)
{
  echo "Hi, there";
}

// group 2.2.1
if($a)
{
  ?> Hi, there <?
}
else
{
  ?> Ho, there <?
}

// group 2.2.2
if($a)
{
  ?> Hi, there <?
}

-- 
:wq