People, when you will start to learn that such approach is a mess?
First, use ' ' instead of and format your SQL better and you woun't have
any problems:
?php
define ('STATUS_ACTIVE', 1);
define ('NUM_PER_PAGE', 25);
// I beleve here is a lot of code
// ...
//
$q = 'SELECT
ID, Name
Hi,
With namespaces and use we'll be introducing a new kind of discrepancy
between a string reference to a function, and the short use-enhanced name for
the same function. This becomes very painful, when I want to load a function
before I run it.
Today I would do (due to lack of function
On Wed, 29 Oct 2008, Sanjay Mantoor wrote:
While I was working on getdate function, found different behavior with
respect to http://in.php.net/manual/en/function.getdate.php
This is observed only on linux 64bit.
Here is code
?php
$timestamp = -12.3456789000e10;
var_dump(
Hi,
I'm hoping I don't cause yet another flame war about the n-word ;)
I was wondering if namespaces are going to support class visibility, and
if no, would it be worth considering adding that feature? What I mean
exacty is:
In namespace Foo:
public class Foo { }
protected class Bar { }
Derick ,
Yes I will do.
Thanks,
Sanjay M
On Wed, Oct 29, 2008 at 1:12 PM, Derick Rethans [EMAIL PROTECTED] wrote:
On Wed, 29 Oct 2008, Sanjay Mantoor wrote:
While I was working on getdate function, found different behavior with
respect to http://in.php.net/manual/en/function.getdate.php
W liście Arvids Godjuks z dnia środa 29 października 2008:
People, when you will start to learn that such approach is a mess?
First, use ' ' instead of and format your SQL better and you woun't
have any problems:
?php
define ('STATUS_ACTIVE', 1);
define ('NUM_PER_PAGE', 25);
$q = 'SELECT
I only worry it could break BC - people might have used {SOMETEXT} in
strings and not expect it to be interpolated (I've done so myself).
I've done that a lot, and I've seen quite a bit of templating code that
does the same.
My personal opinion is that interpolating constants would just
Dave, how is a variable name any less a bareword than a constant name?
Thats what the backets were for, perhaps combined with a symbol to
make it even less likely, and of course if the constant is not found
in the symbol table, the constant name would be outputted directly.
On Wed, Oct 29, 2008
I think this would have been a useful feature if it had always been there...
like variable names being translated inside double quoted strings.
But to add it now would, as has been pointed out, have a good chance of
breaking a lot of existing code, and for that reason I think it's probably a
I think this has merit especially when you consider the heredoc syntax. Many
of my queries are in heredoc for the sakes of readability and cleansliness.
However, the cases are rare in which I would do so.
Rather, with offsets and other variables in queries it makes more sense for
variables
Just a quick reminder for all of the developers, RC3 (the final RC)
will be tagged tomorrow afternoon (EST) time, please try to get
whatever patches (bug fixes only, please) you still want into it. If
there are reasons to delay, critical patches missing, please let me
know.
Thanks,
Ilia Alshanetsky wrote:
Just a quick reminder for all of the developers, RC3 (the final RC) will
be tagged tomorrow afternoon (EST) time, please try to get whatever
patches (bug fixes only, please) you still want into it. If there are
reasons to delay, critical patches missing, please let me
The problem here is that the domain parameter will eventually make its
way into the file path and result in an overflow of the MAX_PATHLEN
there (I suspect). The problem however (IMHO) lies in gettext that
does not check the file path length before using it. For us to do the
check will
Josh wrote:
Dave, how is a variable name any less a bareword than a constant name?
A bareword (in my opinion) is a word without a sigil (e.g. a leading $),
so the way that you would write a constant normally in PHP. What I was
trying to say is that having bareword-style interpolation
Steph
You dismiss the vocal voices far to quickly, claiming its only a minority
against this is very misleading - The majority of Developers are probably
oblivious to this conversation.
The way this argument is heating up and people are being flamed downed,
Namespaces will be known as the
Jani Taskinen wrote:
http://bugs.php.net/bug.php?id=44938edit=1
Marked critical but propably isn't. Also depatable whether it's a PHP
bug at all..
Probably the library exhausting memory using alloca I'd say.
I whipped together a little patch against HEAD which restricts the
length of a text
Since this affects everybody, it should be opened up to the community, so,
the internals should put together several options and then open a poll for
the community to vote on and then use which ever comes out on top.
Cheers
Kelvin
This has been done so many times... I'm following this
Not a bad idea, although I'd put a limit of the domain length o 4kb. I
can't see someone needing a domain longer then that, heck even 1kb
should be more then enough.
On 29-Oct-08, at 11:42 AM, Christian Schneider wrote:
Jani Taskinen wrote:
http://bugs.php.net/bug.php?id=44938edit=1
I really wish that this list was developer-only (which would mean i
couldn't write this post). The developers made a decision. And as much
as I dislike the feel of the '\' as namespace separator, the
argumentation for it is really solid - and it's a one character,
distinctive separator. Would you
On Wednesday 29 October 2008 16:23:55 you wrote:
Since this affects everybody, it should be opened up to the community, so,
the internals should put together several options and then open a poll for
the community to vote on and then use which ever comes out on top.
And the winner is ...
Stan Vassilev | FM wrote:
I suggest we introduce a new construct which will return a string when passed
any identifier to resolve against the current file's use clauses:
nameof Symbol\Name;
With this identifier, the above example can be normalized to the following
code:
I second this
Christian,
Your patch with slight revisions just went in.
On 29-Oct-08, at 11:42 AM, Christian Schneider wrote:
Jani Taskinen wrote:
http://bugs.php.net/bug.php?id=44938edit=1
Marked critical but propably isn't. Also depatable whether it's a PHP
bug at all..
Probably the library
Hello Ron,
we agreed long ago on a very easy scheme, there shall only be is-a and
public classes.
marcus
Wednesday, October 29, 2008, 9:12:30 AM, you wrote:
Hi,
I'm hoping I don't cause yet another flame war about the n-word ;)
I was wondering if namespaces are going to support class
23 matches
Mail list logo
