Per Jessen wrote:
magoo wrote:
In their oppinion stuff like the short PHP init "<?" or inline echo
"<?= $var ?>" is forbidden.
Do you people code that strict?
Yep. If you're using XHTML you have no other option.
If you're parsing XHTML that is choking on any instances of "<%" then
you've got a problem with your script. The point of including PHP tags
inside of an "XHTML" file is that it isn't really XHTML until PHP has
parsed it for its own instructions. Until then, it isn't XHTML yet and
shouldn't be treated as such. If you've some need to treat the raw files
as XHTML on the server *before* it reaches PHP then you're doing
something wrong.
Greg Donald wrote:
>
> I don't put PHP in my HTML. I use PHP to build my HTML, using heredoc
> syntax while doing so. My scripts usually only have a single <?php
> opening tag and never require any closing ?> tags.
You make it sound as if you've saved yourself having to type 2
characters. If you're using Zend or a template engine and your scripts
do not contain anything other than PHP then you *should not* include the
closing tag. It's not really a choice (well, it is, but you run the risk
of creating difficult-to-track bugs).
It all depends on the context. For example:
Say i turn an existing HTML file into a PHP script and add some
instructions somewhere within the middle of the body. Of course i'm
going to have to use a closing tag somewhere before the next bit of HTML
markup. But i certainly don't care that my script *requires* that i use
that closing tag. I don't think, "Gee, i wish i didn't have to put this
closing tag here."
> Code that contains short open tags or instances of <?= is amatuer in
> my opinion.
Amateur? I agree that short open tags should be avoided but i don't see
anything wrong with using the <?= ?> construct. When a full-blown
templating engine is not available it can be a huge help to break down
pages into seperate portions, each containing instructions for the PHP
engine to fill in. Got a script that displays a form on one request and
something different on another? It's easy to simply include() the
necessary HTML part from another file, complete with <?= ?> instances
to fill in the stuff you want PHP to gather.
Hence, a poor man's template. I open the file and see an HTML form and
can follow it easily. I don't have to wade through a bunch of heredoc
blocks in the main script. Instead, the script where i'm doing all the
heavy lifting is empty of presentational markup that would only get in
the way.
I find it amusing when people go on and on about keeping their PHP
instructions out of the markup yet have no qualms about doing precisely
the opposite. Six of one ...
It's a matter of determining what is the model and what is the view and
how best to create an application structure.
And using echo or heredoc to create HTML usually creates unreadable
markup, IMO. That may not matter to some but for those of us who are
responsible for ensuring pages validate it certainly helps when your
editor of choice (i use jEdit) can highlight the syntax--both PHP and
HTML--properly. Markup stuck in a heredoc (or ~shudder~ a thousand echo
statements) isn't so easy on the eyes.
In these "templates"i use, aside from filling in spots with <?= $my_var
?> i may also have the odd loop to iterate over a database result. Not
so different from XSL, in fact. But i don't have all of the DB queries,
checking of credentials, class definitions (yes, i've seen it: classes
defined in the same 10,000-line script that the browser requests) inside
of that file. It wouldn't be anything remotely similar to a template then.
There's certainly a case to be made for ensuring that scripts follow
certain guidelines but to assert that there is *one* proper way for all
cases seems a bit off to me. Use what works, makes sense, and is *safe*
within *your* workflow.
Just thought i'd add something to get the discussion going beyond, "Me
too! That way is teh suxxor." ;-)
brian
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
