php-general Digest 24 Dec 2006 04:02:48 -0000 Issue 4531
Topics (messages 246171 through 246184):
Re: Clarification: Jump to a record/PHP paging...
246171 by: tedd
246172 by: T.J. Mahaffey
246173 by: Robert Cummings
246174 by: tedd
Script's length, echo, and execution speed
246175 by: Jean-Christophe Roux
246177 by: Satyam
246178 by: Paul Novitski
246179 by: Casey Chu
246180 by: Jürgen Wind
246181 by: Jon Anderson
Re: New User Sign up Notification
246176 by: Casey Chu
246182 by: Casey Chu
246183 by: Ryan Fielding
[Announce] GeryEngine a PHP5 template engine
246184 by: Arjen Brouwer
Administrivia:
To subscribe to the digest, e-mail:
[EMAIL PROTECTED]
To unsubscribe from the digest, e-mail:
[EMAIL PROTECTED]
To post to the list, e-mail:
[email protected]
----------------------------------------------------------------------
--- Begin Message ---
At 9:03 AM -0600 12/23/06, T.J. Mahaffey wrote:
I see now that I did not explain myself adequately.
I think "jump to record" was the wrong way to put it. So, here goes.
I already have excellent paging functionality working well, based on
a nice tutorial at PHPFreaks.
My problem is that when a user performs a search, I need to display
the page on which their search string is found, but still display
ALL records within the paging of the entire database.
I've since discovered the core of what I need to do:
1. I can find the record I need through a simple query.
2. I can easily determine WHICH page this record is on by
counting BACKWARDS from the found record to the FIRST record,
totaling the number of records from record 1 to the found record.
Then, by performing a bit of division, I can determine which page
that record appears on and direct the user to "...page=8" via $_GET.
SO, my question is: how might I have MySQL tell me how many records
came BEFORE the found record?
(FYI: there is currently no auto-incrementing ID on these records,
so that obviously easy solution would be unavailable.)
Thanks in advance for any insight.
--
T.J. Mahaffey
[EMAIL PROTECTED]
tj:
In my opinion, wrong thinking. You ask how many records appear before
and after a search string result. That's like asking "What's the
difference between an orange?" -- it doesn't make sense.
The user enters a search string. All records are searched and the
correct one is found and presented. Now you want the records that
were not correct to be arranged in some manner of before and after
correctness -- it doesn't make sense.
You could use a FULL TEXT search in order of relevance, much like
what search engines do. But those searches come back with correct
searches and then sort them in terms of relevance. They don't show
the correct record and then all the records that were before and
after correctness. Do you see what I mean?
You could put in a field that has a number assigned to it, and show
what numbers came before and after the record selected (like sequence
of entry, auto-incrementing ID ), but that isn't relevant to the
original search criteria.
You need to think about what you specifically want to show the user
-- what would make sense, or of use, to them?
hth's
tedd
--
-------
http://sperling.com http://ancientstones.com http://earthstones.com
--- End Message ---
--- Begin Message ---
I agree, tedd. And your points about what makes sense are taken.
That being said, the end point of all this is for a search to simply
act like a real person turning the pages of a phone book.
All I need to do is take the user to a desired point within the
records in the database. This is by request of the client I'm working
for, not what makes sense. (As you may well know, these two don't
always meet. lol)
At this point, it seems to me that in order to achieve my desired
result, I need to incorporate an auto-incremented ID column in the
table so that this value may be used to calculate the number of
records before the found record.(ie, if the found record's ID is 201,
then there were 200 records before it. Divide 200 by 100 records per
page and the desired result is on page 3.)
Assuming no one else sees a method to achieve the desired result
without such a change.
Thank you, tedd, for your input. It does help to have someone
--
T.J. Mahaffey
[EMAIL PROTECTED]
On Dec 23, 2006, at 10:05 AM, tedd wrote:
tj:
In my opinion, wrong thinking. You ask how many records appear
before and after a search string result. That's like asking "What's
the difference between an orange?" -- it doesn't make sense.
The user enters a search string. All records are searched and the
correct one is found and presented. Now you want the records that
were not correct to be arranged in some manner of before and after
correctness -- it doesn't make sense.
You could use a FULL TEXT search in order of relevance, much like
what search engines do. But those searches come back with correct
searches and then sort them in terms of relevance. They don't show
the correct record and then all the records that were before and
after correctness. Do you see what I mean?
You could put in a field that has a number assigned to it, and show
what numbers came before and after the record selected (like
sequence of entry, auto-incrementing ID ), but that isn't relevant
to the original search criteria.
You need to think about what you specifically want to show the user
-- what would make sense, or of use, to them?
hth's
tedd
--- End Message ---
--- Begin Message ---
On Sat, 2006-12-23 at 11:05 -0500, tedd wrote:
>
> In my opinion, wrong thinking. You ask how many records appear before
> and after a search string result. That's like asking "What's the
> difference between an orange?" -- it doesn't make sense.
It makes perfect sense. He has an order to his universe of items. The
search criteria pull out a subset of his universe and he wants to know
where in the universe any given item normally lies.
This is akin to asking I want all words from the dictionary that have
monster in their definition, then look at the word Yeti and saying ok,
now where does Yeti lie with respect to all other words.
Now to provide the answer he wants... AFAIK you need to select all items
from the table and populate into a temporary table with auto increment.
Then select the item in question and it's temporary ID from the
temporary table.
This question comes up every once in a while, I've yet to need such a
feature, and preferrably will never need it since it's not a convenient
solution to select the entire table... but maybe there's a better
solution.
Cheers,
Rob.
--
.------------------------------------------------------------.
| InterJinn Application Framework - http://www.interjinn.com |
:------------------------------------------------------------:
| An application and templating framework for PHP. Boasting |
| a powerful, scalable system for accessing system services |
| such as forms, properties, sessions, and caches. InterJinn |
| also provides an extremely flexible architecture for |
| creating re-usable components quickly and easily. |
`------------------------------------------------------------'
--- End Message ---
--- Begin Message ---
At 11:24 AM -0500 12/23/06, Robert Cummings wrote:
On Sat, 2006-12-23 at 11:05 -0500, tedd wrote:
In my opinion, wrong thinking. You ask how many records appear before
and after a search string result. That's like asking "What's the
difference between an orange?" -- it doesn't make sense.
It makes perfect sense. He has an order to his universe of items. The
search criteria pull out a subset of his universe and he wants to know
where in the universe any given item normally lies.
Okay, but in what order? Numerically, alphabetically, temporally
(time entered, time last altered), likeness, length, height, width,
weight, density, color, or what? You need to measure the item to
others in some frame of reference.
This is akin to asking I want all words from the dictionary that have
monster in their definition, then look at the word Yeti and saying ok,
now where does Yeti lie with respect to all other words.
Again, in my opinion, that's fuzzy thinking. Sure, Yeti may be in the
monster sub-set, but how much monster it is as compared to Lockness?
You need a means, a yardstick, to compare the two with each other.
Like, is Lockness bigger than the Yeti while monster-mouse is
smaller, or what?
Now to provide the answer he wants... AFAIK you need to select all items
from the table and populate into a temporary table with auto increment.
Then select the item in question and it's temporary ID from the
temporary table.
Sure, if you can decide on what bases to select and sort the other
items to populate the table.
Cheers,
tedd
--
-------
http://sperling.com http://ancientstones.com http://earthstones.com
--- End Message ---
--- Begin Message ---
Hello,
I have this php script of 3,500 lines with a big switch that is such that on
each pass maybe 300 lines of codes are executed at most. The current speed
of the file is ok. I like to keep the file like that because at each pass
there is a check on the whole script and it fails if there is a typo
somewhere. Also, I like to have one file instead of many files. But I am
wondering if speed is not severaly hurt. What are the general guidelines in
terms
of length of script?
Also, I am writing things like that:
echo '<table>';
echo '<tr>';
echo "<td>Content</td>";
echo '</tr>';
echo '</table>';
I like it this way because the script is visually very well aligned, with one
action
per line and that makes things easier for me to read and understand. But, so
many echoes may be considered bad practice; maybe it impacts speed
significantly.
Thanks
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
I would assume it does, since parsing each instruction and calling the
internal functions to execute each echo should take time. I would use
heredoc, which would allow you to make a single call and still keep the
formatting of the string.
You might also be interested in my own plugin to the Php Compiler which does
just that, it compacts consecutive echos:
http://satyam.com.ar/pht/compacting_echos.htm. This is a pre-compiler, it
produces a new PHP source which will be the one interpreted. This new
source is not quite legible as you wish, but it is not meant to be, all the
nice formatting for the echo is lost, but it is not a source to be
mantained, you would still use the originals to edit, the output of the
pre-compiler to execute. So, you would be able to format the echoes in the
source in whichever way you wish, the pre-compiler plugin would take care of
compacting them for speed.
Notice that the pre-compiler itself can use my plugin or any other anybody
might develop and I would encourage people in this list to do so. The
plugin to compact echos was a consequence of the plugin I meant to do, which
you will find at http://satyam.com.ar/pht/, which produced a large number of
echos and so it almost forced me to do something about them. I never
thought about other possible compile-time optimizations, but if anyone does,
the Php Compiler is a wonderful platform to do them.
I am sure you will also get a lot of replies sugesting template engines such
as Smarty.
Satyam
----- Original Message -----
From: "Jean-Christophe Roux" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Saturday, December 23, 2006 7:33 PM
Subject: [PHP] Script's length, echo, and execution speed
Hello,
I have this php script of 3,500 lines with a big switch that is such that on
each pass maybe 300 lines of codes are executed at most. The current speed
of the file is ok. I like to keep the file like that because at each pass
there is a check on the whole script and it fails if there is a typo
somewhere. Also, I like to have one file instead of many files. But I am
wondering if speed is not severaly hurt. What are the general guidelines in
terms
of length of script?
Also, I am writing things like that:
echo '<table>';
echo '<tr>';
echo "<td>Content</td>";
echo '</tr>';
echo '</table>';
I like it this way because the script is visually very well aligned, with
one action
per line and that makes things easier for me to read and understand. But, so
many echoes may be considered bad practice; maybe it impacts speed
significantly.
Thanks
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
At 12/23/2006 10:33 AM, Jean-Christophe Roux wrote:
Hello, I have this php script of 3,500 lines with a big switch that
is such that on each pass maybe 300 lines of codes are executed at
most. The current speed of the file is ok. I like to keep the file
like that because at each pass there is a check on the whole script
and it fails if there is a typo somewhere. Also, I like to have one
file instead of many files. But I am wondering if speed is not
severaly hurt. What are the general guidelines in terms of length of
script? Also, I am writing things like that: echo
'<table>'; echo '<tr>'; echo "<td>Content</td>"; echo
'</tr>'; echo '</table>'; I like it this way because the script
is visually very well aligned, with one action per line and that
makes things easier for me to read and understand. But, so many
echoes may be considered bad practice; maybe it impacts speed significantly.
Are you encountering a speed problem or just pondering this
theoretically? My guess is that ECHO 'TEXT'; is one of the fastest,
most optimized operations the PHP interpreter executes.
But you don't need to guess at it: time it. Check the server logs,
or output the current time at the beginning and end of script
execution. My guess is that the difference between the execution of
your script and one much smaller will lie in a scale of tens or
hundreds of milliseconds. Servers are bloody fast; that's their job.
In general I think it's smarter to spend your time making your script
easy to read and proofread for you, the human slow link in the
development chain, than it is to make it faster to run on a
machine. Machines will keep getting faster; you're not likely to.
That said, there are things you can do that make your script more
efficient to execute AND to maintain. Personally I favor separating
out the HTML into larger, pure chunks that can be proofread more
easily than when tags and attributes are atomized and strewn
throughout the logic.
Regards,
Paul
--- End Message ---
--- Begin Message ---
Yeah, try testing. Maybe something like this:
<?php
$form = "<form action=' ' method='post'><textarea
name='code'></textarea><br /><button
type='submit'>Execute</button></form>";
if ($_POST['code']) {
echo "<div style='border: 1px'>";
$time_start = microtime_float();
eval($_POST['code']);
$time_end = microtime_float();
echo "<br />Loading took: ". $time_end - $time_start. '</div>';
}
echo "<div style='float: left'>$form</div>";
// Note: This script is extremely dangerous and is not tested.
?>
On 12/23/06, Paul Novitski <[EMAIL PROTECTED]> wrote:
At 12/23/2006 10:33 AM, Jean-Christophe Roux wrote:
>Hello, I have this php script of 3,500 lines with a big switch that
>is such that on each pass maybe 300 lines of codes are executed at
>most. The current speed of the file is ok. I like to keep the file
>like that because at each pass there is a check on the whole script
>and it fails if there is a typo somewhere. Also, I like to have one
>file instead of many files. But I am wondering if speed is not
>severaly hurt. What are the general guidelines in terms of length of
>script? Also, I am writing things like that: echo
>'<table>'; echo '<tr>'; echo "<td>Content</td>"; echo
>'</tr>'; echo '</table>'; I like it this way because the script
>is visually very well aligned, with one action per line and that
>makes things easier for me to read and understand. But, so many
>echoes may be considered bad practice; maybe it impacts speed significantly.
Are you encountering a speed problem or just pondering this
theoretically? My guess is that ECHO 'TEXT'; is one of the fastest,
most optimized operations the PHP interpreter executes.
But you don't need to guess at it: time it. Check the server logs,
or output the current time at the beginning and end of script
execution. My guess is that the difference between the execution of
your script and one much smaller will lie in a scale of tens or
hundreds of milliseconds. Servers are bloody fast; that's their job.
In general I think it's smarter to spend your time making your script
easy to read and proofread for you, the human slow link in the
development chain, than it is to make it faster to run on a
machine. Machines will keep getting faster; you're not likely to.
That said, there are things you can do that make your script more
efficient to execute AND to maintain. Personally I favor separating
out the HTML into larger, pure chunks that can be proofread more
easily than when tags and attributes are atomized and strewn
throughout the logic.
Regards,
Paul
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
--- End Message ---
--- Begin Message ---
Casey Chu wrote:
>
> Yeah, try testing. Maybe something like this:
> <?php
> $form = "<form action=' ' method='post'><textarea
> name='code'></textarea><br /><button
> type='submit'>Execute</button></form>";
> if ($_POST['code']) {
> echo "<div style='border: 1px'>";
> $time_start = microtime_float();
> eval($_POST['code']);
> $time_end = microtime_float();
> echo "<br />Loading took: ". $time_end - $time_start. '</div>';
> }
> echo "<div style='float: left'>$form</div>";
> // Note: This script is extremely dangerous and is not tested.
> ?>
>
>
under php5 use can use microtime(1) for timestamps
--
View this message in context:
http://www.nabble.com/Script%27s-length%2C-echo%2C-and-execution-speed-tf2875055.html#a8037132
Sent from the PHP - General mailing list archive at Nabble.com.
--- End Message ---
--- Begin Message ---
Casey Chu wrote:
Yeah, try testing. Maybe something like this:
<?php
$form = "<form action=' ' method='post'><textarea
name='code'></textarea><br /><button
type='submit'>Execute</button></form>";
if ($_POST['code']) {
echo "<div style='border: 1px'>";
$time_start = microtime_float();
eval($_POST['code']);
$time_end = microtime_float();
echo "<br />Loading took: ". $time_end - $time_start. '</div>';
}
echo "<div style='float: left'>$form</div>";
// Note: This script is extremely dangerous and is not tested.
?>
I haven't been following this thread, so I don't know if this has been
suggested...
If you can run a local server, IMO, xdebug is *the* way to go...xdebug2
has an excellent profiler built-in. You can use kcachegrind or
wincachegrind to see some pretty mind-boggling detail about where your
performance goes. It's much easier than inserting timing statements into
existing code.
jon
--- End Message ---
--- Begin Message ---
What do you mean by "now"?
On 12/22/06, JMCS Niagara (Jeff) <[EMAIL PROTECTED]> wrote:
hi there. I'm looking to create a new user signup. So when someone signs up, it
emails admin the Username, Real Name, and IP address.
I have this code called newusersignupnotification.php
<?php
$to = '[EMAIL PROTECTED]'; // The email address you want the notification sent
to
$subject = 'Member has registered on YoS'; // What do you want the subject line
of your notification to be?
$memberSpecs =
"<hr size=2 width=300 align=left>".
"<b>Username:</b> ".$_SESSION['username'].
"<br>".
"<b>Real Name:</b> ".$_SESSION['firstname']. " ".$_SESSION['lastname'].
"<br>".
"<b>Date</b> ".date('l dS \of F Y h:i:s A').
"<br>".
"<b>IP address:</b> ".$_SERVER['REMOTE_ADDR'].
"<br>".
"<b>System Specs:</b> ".$_SERVER['HTTP_USER_AGENT'].
"<br>";
$headers = "Content-type: text/html \nFrom: [EMAIL PROTECTED]";
$body = "<body>
<br>
<table cellspacing=1 cellpadding=2 align=center>
<tr>
<td>
<b><font face=arial size=2>A new member has registered! </font></b>
</td></tr>
<tr>
<td>
<font face=arial size=2> ".$memberSpecs." </font>
</td></tr></table>
</body>";
mail($to,$subject,$body,$headers);
?>
When I place this in the "join.php" file .. When someone goes to join.php it
automatically emails it .. and I get date, time, ip, etc etc everything except for the
username and first name.. Obviously, because the person hasn't typed in anything.
When they filled out the sign up form, There's a button that says Sign Up ..
Basically, when they click this button, NOW is when I want it to email me the
details.
Any ideas?
Thanks guys
--- End Message ---
--- Begin Message ---
I still do not understand. Isn't that what your script already does?
On 12/23/06, Casey Chu <[EMAIL PROTECTED]> wrote:
What do you mean by "now"?
On 12/22/06, JMCS Niagara (Jeff) <[EMAIL PROTECTED]> wrote:
> hi there. I'm looking to create a new user signup. So when someone signs up,
it emails admin the Username, Real Name, and IP address.
>
> I have this code called newusersignupnotification.php
>
> <?php
> $to = '[EMAIL PROTECTED]'; // The email address you want the notification
sent to
> $subject = 'Member has registered on YoS'; // What do you want the subject
line of your notification to be?
> $memberSpecs =
> "<hr size=2 width=300 align=left>".
> "<b>Username:</b> ".$_SESSION['username'].
> "<br>".
> "<b>Real Name:</b> ".$_SESSION['firstname']. " ".$_SESSION['lastname'].
> "<br>".
> "<b>Date</b> ".date('l dS \of F Y h:i:s A').
> "<br>".
> "<b>IP address:</b> ".$_SERVER['REMOTE_ADDR'].
> "<br>".
> "<b>System Specs:</b> ".$_SERVER['HTTP_USER_AGENT'].
> "<br>";
> $headers = "Content-type: text/html \nFrom: [EMAIL PROTECTED]";
> $body = "<body>
> <br>
> <table cellspacing=1 cellpadding=2 align=center>
> <tr>
> <td>
> <b><font face=arial size=2>A new member has registered! </font></b>
>
> </td></tr>
> <tr>
> <td>
> <font face=arial size=2> ".$memberSpecs." </font>
> </td></tr></table>
> </body>";
> mail($to,$subject,$body,$headers);
> ?>
>
> When I place this in the "join.php" file .. When someone goes to join.php it
automatically emails it .. and I get date, time, ip, etc etc everything except for the
username and first name.. Obviously, because the person hasn't typed in anything.
>
> When they filled out the sign up form, There's a button that says Sign Up ..
Basically, when they click this button, NOW is when I want it to email me the
details.
>
> Any ideas?
>
> Thanks guys
>
--- End Message ---
--- Begin Message ---
Casey Chu wrote:
I still do not understand. Isn't that what your script already does?
On 12/23/06, Casey Chu <[EMAIL PROTECTED]> wrote:
What do you mean by "now"?
On 12/22/06, JMCS Niagara (Jeff) <[EMAIL PROTECTED]> wrote:
> hi there. I'm looking to create a new user signup. So when someone
signs up, it emails admin the Username, Real Name, and IP address.
>
> I have this code called newusersignupnotification.php
>
> <?php
> $to = '[EMAIL PROTECTED]'; // The email address you want
the notification sent to
> $subject = 'Member has registered on YoS'; // What do you want the
subject line of your notification to be?
> $memberSpecs =
> "<hr size=2 width=300 align=left>".
> "<b>Username:</b> ".$_SESSION['username'].
> "<br>".
> "<b>Real Name:</b> ".$_SESSION['firstname']. "
".$_SESSION['lastname'].
> "<br>".
> "<b>Date</b> ".date('l dS \of F Y h:i:s A').
> "<br>".
> "<b>IP address:</b> ".$_SERVER['REMOTE_ADDR'].
> "<br>".
> "<b>System Specs:</b> ".$_SERVER['HTTP_USER_AGENT'].
> "<br>";
> $headers = "Content-type: text/html \nFrom:
[EMAIL PROTECTED]";
> $body = "<body>
> <br>
> <table cellspacing=1 cellpadding=2 align=center>
> <tr>
> <td>
> <b><font face=arial size=2>A new member has registered! </font></b>
>
> </td></tr>
> <tr>
> <td>
> <font face=arial size=2> ".$memberSpecs." </font>
> </td></tr></table>
> </body>";
> mail($to,$subject,$body,$headers);
> ?>
>
> When I place this in the "join.php" file .. When someone goes to
join.php it automatically emails it .. and I get date, time, ip, etc
etc everything except for the username and first name.. Obviously,
because the person hasn't typed in anything.
>
> When they filled out the sign up form, There's a button that says
Sign Up .. Basically, when they click this button, NOW is when I want
it to email me the details.
>
> Any ideas?
>
> Thanks guys
>
If the user is clicking a sumbit button, the easiest way would be to put
at the top:
if($_REQUEST["Submit"] == "whatever your button says"){
mail(blah);
// process the signup or redirect to the next page.
}
rest of join.php goes here.
Note: You will have to make the submit button action join.php.
--- End Message ---
--- Begin Message ---
Dear PHP community,
I'd like to announce the first public realease of GeryEngine (GE).
GeryEngine is a powerful PHP5 template engine. GE is a port of the well
known Java template engine Velocity with some extra functionality on board,
like inline expressions. GE uses a plain and powerful syntax that can be
mixed in your HTML code.
Example:
<html>
#set($engine = "GeryEngine 0.2.0")
<b>$engine.reverse.upcase</b>
</html>
More information can be found at:
http://geryengine.instant-publishing.nl/
Regards,
Arjen Brouwer
--- End Message ---
