batch

Perhaps using DATE_SUB(CURRENT_TIMESTAMP + 3600, INTERVAL 60 MINUTE) This first adds one hour to the current time and then checks whether ItemPubDate_t is within one hour of that.

I just wanted to share some more information with you. It turns out mysql_real_escape_string() doesn't escape all special characters. For example the % and _ operators for a LIKE clause are not escaped. To properly sanitize the input and literally interpret the $match query I use addcslashes() after mysql_real_escape_string(). $match = "Ti\m"; $match = mysql_real_escape_string($match); // Escapes \x00, \n, \r, \, ', " and \x1a $match = addcslashes($match, "\\%_"); // Escapes \, % and _ Notice how addcslashes() escapes the backslash character (" \ ") again. This is intended behavior so MySQL will interpret the backslash literally. To show you the flow of events: Without addcslashes() "Ti\m" -> escapes to "Ti\\m" -> MySQL unescapes to "Ti\m", unescapes to "Tim" (since "\m" is also treated as a sequence to be unescaped). With addcslashes() "Ti\m" -> escapes to "Ti\\m" -> addcslashes to "Ti\\\\m" -> MySQL unescapes to "Ti\\\m" -> one literal slash + "\m" -> MySQL unescapes "\m" -> result: "Ti\m"

I found a useful entry in the MySQL manual [ here ]. Thanks PFMaBiSmAd!

Edit: Are there any other characters besides the backslash that need extra escaping after being mysql_real_escape_string()'d?

I'd like to have MySQL read queries that contain a " \ " literally - even after they have been mysql_real_escape_string()'d. For example: $match = "Ti\m"; $match = mysql_real_escape_string($match); SELECT * FROM table WHERE name = '$match' Executing this query returns all rows where it matches the name Tim. However, I want it to only return rows where it exactly matches the string Ti\m. I know I can force this behavior by adding an extra slash (ie. Ti\\m), but why doesn't it work by default? Isn't mysql_real_escape_string() supposed to escape the string BUT RETAIN ITS ORIGINAL MEANING? ???

Although I'm disappointed with the lack of responses, I ended up finding a great article on how to solve this very issue [ here ]. Although the instructions are for sorting a single table, you can easily join a second (different) table in the query like so: SELECT t2a.nickname, t2a.date, m1.myid, m1.hours_played FROM table2 AS t2a, members AS m1, ( SELECT myid, MAX( date ) AS t2b_newestdate FROM table2 GROUP BY myid ) AS t2b WHERE t2b.myid = t2a.myid AND t2b_newestdate = t2a.date AND t2b.myid = m1.myid The last line ensures that we will get the proper record from the members table. The two lines before that match the row that contains the unique MyID/Date combination that was returned for the most recent date. Using that information we can then finally select the specific record that contains the most recent date field for this MyID.

When you perform a query in PHPMyAdmin it will also show you the exact code it used to perform the action. So to reproduce the successful result, in theory, all you'd need to do is copy/paste whatever PHPMyAdmin is doing and use that in your SQL query which you execute via PHP.

To join multiple WHERE conditions you use the AND keyword. Example: SELECT * FROM example WHERE username = '$username' AND email = '$email'

The date field is DATETIME, which should be sufficient. Also, the original query is much longer but I only mentioned the relevant details I already: ORDER BY members.hours_played in the real query. I just need to get the most recent nickname+date used by a member using a single query. So whatever record MAX(date) belongs to, that's the row I want returned!

You could use define to tell when to execute the query and when not. Just before you include the common.php file on a page for which you want to execute the query, set: define("DoQuery", true); // second argument is not used in this case, but required for define() require_once('common.php'); // your include In your common.php file you check whether "DoQuery" is defined and if so, you execute the query. if(defined('DoQuery')) { // We were called from a page that requires us to execute the query, so do it! }

I have two tables: members and table2. Here is an example of the contents: members myid hours_played 111 1.5 222 0.8 333 2.4 table2 myid nickname date 111 peter 1-1-2009 222 john 2-2-2009 222 johnnyboy 3-3-2009 333 joe 4-4-2009 I think the tables are fairly obvious: one member can use different nicknames, but can have only one account for which information is stored in the members table. What I want to do: select the most recent record for every member. Problem: by using MAX(table2.date) I do indeed get the most recent date for this particular member (myid), but the rest of the result returned for this member doesn't correspond to the row from which the MAX(date) was obtained. For example I'd get something like this: 222 john 3-3-2009 <- notice this is the most recent date, but the nickname field belongs to the older record! I am expecting to get this: 222 johnnyboy 3-3-2009 <- notice how the nickname field is part of the same record as the date This is my query: SELECT members.myid, table2.nickname, MAX(table2.date) AS date FROM members, table2 WHERE members.myid = table2.myid GROUP BY members.myid