crazytonyi

I figured out the source of the problem, I think. When I join the two queries together, the subquery is treated by the mysql server as a "DEPENDENT SUBQUERY", which means, among other things, that it checks EACH row for to see if it meets the sub-query criteria when it really only needs to find the user IDs to skip once BEFORE looking for the other WHERE criteria. I learned this at: http://forums.mysql.com/read.php?115,128477,128477 The suggested solution from the above site is basically what I was already doing, which is run two queries and pass the data from the first on to the second. If anyone knows a better technique, I'd really appreciate the advice. a

If anyone needs background on this project, or wants to offer more broad advice, go to this post: http://www.phpfreaks.com/forums/index.php/topic,231571.0.html So, now I've got my queries out of any loops, so I'm feeling a lot better overall. However, I'm still getting problems with sub-queries. For those of you who are keeping up, I got around the looping by having the swap query omit any overlapping shift times all at once. I realized that my script had to pull the user's shifts each time the page loaded in order to display them, so now when it loops through them to echo them out to the user, it also adds the shift to a string like this: $conflict_dates .= "AND shiftstart NOT BETWEEN FROM_UNIXTIME('".(strtotime($shift_row['shiftstart']) + 60)."') AND FROM_UNIXTIME('".(strtotime($shift_row['shiftend']) + 60)."') AND FROM_UNIXTIME('".(strtotime($shift_row['shiftstart']) + 60)."') NOT BETWEEN shiftstart AND shiftend\n"; Having done that, I'm now down to 3 queries: The first gets the users shifts The second one finds any OTHER users working at the same time as the shift the user wants to swap The third finds shifts that are the same length as the swap shift AND don't conflict with any of the user's shift AND are not on the blacklisted users from the second query. Now, the first one MUST happen, because without it, the user won't have a list of shifts to choose from. But the second should be a subquery of the third one. But when I tried doing this, the page took 47 seconds to load versus doing the two queries separately which took under half a second. In either case, the same results are returned, so I know I'm not totally doing this wrong. But obviously I'm missing something, because the query shouldn't take 100 times longer. The mySQL server is 5.0.60 Here's the jist of how it looks: First query gets all user's shifts based on predetermined date range: SELECT * FROM shifts WHERE userid = 'joe_smith' AND shiftstart BETWEEN FROM_UNIXTIME('epoch_of_Monday') AND FROM_UNIXTIME('epoch_of_Friday') As mentioned above, the script outputs the shifts to the user AND makes a string variable for the last query. The next query (after the user chooses a shift) gets the user IDs of anyone who has a shift that overlaps with the shift chosen: SELECT DISTINCT userid FROM shifts WHERE ( shiftstart BETWEEN FROM_UNIXTIME('shift_start_plus_1_minute') AND FROM_UNIXTIME('shift_end_minus_1_minute') OR FROM_UNIXTIME('epoch_of_shift_start_plus_1_minute') BETWEEN shiftstart AND shiftend ) Now, having a blacklist of users (formatted correctly by the script), it queries for any shifts that meet the criteria for swapping: SELECT * FROM shifts WHERE userid NOT IN (list_of_userids) AND shiftstart BETWEEN FROM_UNIXTIME('epoch_of_Monday') AND FROM_UNIXTIME('epoch_of_Friday') AND (TIME_TO_SEC(TIMEDIFF(shiftend,shiftstart)) = 'shift_length') //conflicts string, written out for this post// AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift1_start') AND FROM_UNIXTIME('joes_shift1_end') AND FROM_UNIXTIME('joes_shift1_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift2_start') AND FROM_UNIXTIME('joes_shift2_end') AND FROM_UNIXTIME('joes_shift2_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift3_start') AND FROM_UNIXTIME('joes_shift3_end') AND FROM_UNIXTIME('joes_shift3_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift4_start') AND FROM_UNIXTIME('joes_shift4_end') AND FROM_UNIXTIME('joes_shift4_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift5_start') AND FROM_UNIXTIME('joes_shift5_end') AND FROM_UNIXTIME('joes_shift5_start') NOT BETWEEN shiftstart AND shiftend ORDER BY shiftstart, lastname Now, the way I did the subquery was to place the second query's variable name ($busy_query) into the parenthesis of the third query, replacing the variable tht held the list of bad names. So it would look more like(if it was fully written out): SELECT * FROM shifts WHERE userid NOT IN ( SELECT DISTINCT userid FROM shifts WHERE ( shiftstart BETWEEN FROM_UNIXTIME('shift_start_plus_1_minute') AND FROM_UNIXTIME('shift_end_minus_1_minute') OR FROM_UNIXTIME('epoch_of_shift_start_plus_1_minute') BETWEEN shiftstart AND shiftend ) ) AND shiftstart BETWEEN FROM_UNIXTIME('epoch_of_Monday') AND FROM_UNIXTIME('epoch_of_Friday') AND (TIME_TO_SEC(TIMEDIFF(shiftend,shiftstart)) = 'shift_length') //conflicts string, written out for this post// AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift1_start') AND FROM_UNIXTIME('joes_shift1_end') AND FROM_UNIXTIME('joes_shift1_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift2_start') AND FROM_UNIXTIME('joes_shift2_end') AND FROM_UNIXTIME('joes_shift2_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift3_start') AND FROM_UNIXTIME('joes_shift3_end') AND FROM_UNIXTIME('joes_shift3_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift4_start') AND FROM_UNIXTIME('joes_shift4_end') AND FROM_UNIXTIME('joes_shift4_start') NOT BETWEEN shiftstart AND shiftend AND shiftstart NOT BETWEEN FROM_UNIXTIME('joes_shift5_start') AND FROM_UNIXTIME('joes_shift5_end') AND FROM_UNIXTIME('joes_shift5_start') NOT BETWEEN shiftstart AND shiftend ORDER BY shiftstart, lastname Maybe it's too much for one query? But even when I do it without the conflicts, it takes the same amount of time. Is this a syntax problem or a server issue or what? Thanks! a

Okay, so I may have skimmed the "guidelines" of this board. My sincere apologies. Please read this modified version of my post: MySQL server version 5.0.60 I have three queries, currently. Ideally, I would like to have just one. The main issue is that the third query is actually inside of a while loop, running a query for each result of the second statement. So if there are 50 results, that ends up being 50 queries. I know this can't be the only way. User wants to swap his 7:45 am to 12:15 pm shift on January 5th. The script gets the following: 1231163100_1231179300 It splits the string into two variables and adds one minute to the start time and subtracts one minute from the end time. This keeps the query from eliminating people whose shift "overlap" by starting when his shift ends. Before the modification, it gets the difference of the two to query only shifts of the same length in the second query. First query looks for any other users in the database working at that time so that those users will be left out of the second query. This is not only to avoid showing joe shifts that are at the same time as the shift he wants to swap, but also because he can't swap for Jane's friday shift because Jane can't work Joe's Monday shift and her own Monday shift simeltaneously, so none of her shifts are potentials... SELECT DISTINCT userid FROM shifts AND ( shiftstart BETWEEN FROM_UNIXTIME('1231163160') AND FROM_UNIXTIME('1231179240') OR FROM_UNIXTIME('1231163160') BETWEEN shiftstart AND shiftend ) To be clear, the request above asks for usernames of anyone working a shift where the starttime is between the user's shift AND working a shift where the user's starttime time is between thier shift start and end. It took me a while to figure out that this would cover all overlapping shifts. The second query looks for anyone working a shift of the same lenght that workweek. Sorry for the PHP variable, but I'm not sure how else to show that the query omits any user ids found in the above query: FROM shifts AND userid NOT IN ('$busy_users') AND shiftstart BETWEEN FROM_UNIXTIME('1231113600') AND FROM_UNIXTIME('1231718399') AND (TIME_TO_SEC(TIMEDIFF(shiftend,shiftstart)) = 16200) ORDER BY shiftstart Finally, with a list of potential shifts, the php script runs a loop through the results and queries the database to see if the user is working already during that shift. Again, the script adjusts the time by one minute on each end to keep adjacent shifts from being omitted. For the purpose of this post, let's assume that the above query found 10 shifts. The loop will run ten queries, of course, but here is query number 8, which is a shift on Thursday the 8th, from 1:30pm to 6:00pm (adjusted, of course) : SELECT * FROM shifts WHERE userid = 'joe_smith' AND ( shiftstart BETWEEN FROM_UNIXTIME('1231421460') AND FROM_UNIXTIME('1231394340') OR FROM_UNIXTIME('1231421460') BETWEEN shiftstart AND shiftend ) The query retuned no results, which means that Joe, our swap seeker, is not working that afternoon. The script stores that to echo later. And query 9 shows one result, so it is thrown out and we have some confidence that everything works correctly, just not optimally. So, there you have it. Thanks for reading any and all of this. Again, what the query would ideally do is say "show all shifts of this length, between these dates, where userid "joe smith" is not working at the same time and neither is anyone else who has a shift htat overlaps with these given times (joe's shift he wants to ditch.) Again, thanks for reading. Sorry for the confusion. Thanks to anyone who can help! a

Is this question too hard or too easy? Someone let me know, because if it's too easy I'll try read up more, but if it's too hard, maybe I can find someone who wants to figure it out with me. thanks!

any help?

I'm no expert, but I think the way to do this would be to... have the script that handles saving the form (whatever you have in action="" in the form tag) save the data to the database and also echo the html code that you want the invoice to look like to with the data filled in and with a javascript function in the head that sends the command to the user's printer to print. SO! you have a couple of choice: 1. You can have the script that handles inserting the data query the database for that same data (which is probably bad practice), or simply use the same $_POST variables that you are adding to the database to echo into the printable page. 2. You could make the page "printer-ready" meaning it will print what the user sees on the screen which will be coded to look better on paper OR you can use @media print in your CSS and have it look good on the screen and on the page (changing the margins, getting rid any nav divs, change the font, etc)... There are plenty of sites that go over the different way's of going about the CSS option. And as far as the page auto printing on load, you need to put this javascript somewhere in there: onLoad="window.print()"

Okay, don't everybody answer at once. So, I figured out a couple of things: 1. The idea I was looking for (I think) was a SELECT subquery. But when I tried those out, the output was SUPER slow. Which makes me think multiple queries are not as horrible as some people like to act. OR that I did the subquries all wrong. Either way, I did get a little bit closer to my goal, so I thought I'd add it to the post before you guys break your legs trying to help... Now, it queries who is working at the same time first and I have the script transform the returned user_ids into a list that the next query uses as a black list. This removes the need to check if those people are working during the shift in the loop because those people don't show up in the results. Take a look: $busy_query = " SELECT DISTINCT userid FROM shifts AND ( shiftstart BETWEEN FROM_UNIXTIME('$shift_overlap_start') AND FROM_UNIXTIME('$shift_overlap_end') OR FROM_UNIXTIME('$shift_overlap_start') BETWEEN shiftstart AND shiftend ) "; $busy_result = mysql_query($busy_query); while ($busy_row = mysql_fetch_assoc($busy_result)) { $busy_uids[] = "'".$firstbusy_row['userid']."'"; } $busy_uids = implode(",", $busy_userids); $swap_query = "SELECT * FROM shifts AND userid NOT IN ($busy_uids) AND shiftstart BETWEEN FROM_UNIXTIME('$startday') AND FROM_UNIXTIME('$endday') AND shiftstart NOT BETWEEN FROM_UNIXTIME('$shift_overlap_start') AND FROM_UNIXTIME('$shift_overlap_end') AND (TIME_TO_SEC(TIMEDIFF(shiftend,shiftstart)) = $swap_shift_length) ORDER BY shiftstart "; And then goes back to the loop mentioned in my first post, which is now much shorter: $conflict_query = " SELECT * FROM shifts WHERE userid = '$userid' AND ( shiftstart BETWEEN FROM_UNIXTIME('$startdate_overlap') AND FROM_UNIXTIME('$enddate_overlap') OR FROM_UNIXTIME('$startdate_overlap') BETWEEN shiftstart AND shiftend ) "; $conflicts = mysql_num_rows(mysql_query($conflict_query)); if ($conflicts == 0) //print the shift// So I still don't know how to query in advance and say "Hey if this guy is working at the same time, don't grab that row" Help! a

So the basic idea is this: A user goes to this page because he's working a shift and wants to swap with someone else. The boundaries are: [*]The user can't swap for a shift at the same time (they have a court date). [*]The user can't work two shifts at once (so they can't swap for a shift later in the week if they work during that shift). [*]Other employees can't work two shifts at once (so user can't swap for a shift later in the week if the other guy works during that shift). [*]The user's shift and the other guy's shift have to be the exact same length (the boss refuses to deal with anything more complicated). First query: pulls all of the user's shifts for the week: $shift_query = " SELECT * WHERE userid = '$userid' AND shiftstart >= FROM_UNIXTIME('$start_day') AND shiftend <= FROM_UNIXTIME('$end_day') "; After user chooses a shift, it's returned back to the script as $_POST and the script trys to find a match: First it checks for all shifts that are a)not the user's shifts, b)not overlapping the selected shift, and c)the same length of time as the selected shift: $swap_query = " SELECT * FROM postedshifts AND userid <> '$userid' AND shiftstart BETWEEN FROM_UNIXTIME('$startday') AND FROM_UNIXTIME('$end_day') AND shiftstart NOT BETWEEN FROM_UNIXTIME('$swap_shift_start') AND FROM_UNIXTIME('$swap_shift_end') AND shiftend NOT BETWEEN FROM_UNIXTIME('$swap_shift_start') AND FROM_UNIXTIME('$swap_shift_end') AND (TIME_TO_SEC(TIMEDIFF(shiftend,shiftstart)) = $swap_shift_length) ORDER BY shiftstart "; After all possible shifts are chosen, the script loops through the possible shifts (the results from the query run just before) and runs two queries based on the row: The first one makes sure that the person assigned to the eligble shift doesn't already work during the selected shift. (In other words, I can't swap my shift on Thursday for the user's Monday shift if I'm already working during the Monday shift.) The query run just before prevented any of my shifts at the same time from being chosen, but it doesn't realize that means I can't swap any of my other shifts either. $busy_query = " SELECT * FROM postedshifts WHERE userid = '$swap_userid' AND ( (shiftstart BETWEEN FROM_UNIXTIME('$swap_shift_start') AND FROM_UNIXTIME('$swap_shift_end')) OR (shiftend BETWEEN FROM_UNIXTIME('$swap_shift_start') AND FROM_UNIXTIME('$swap_shift_end')) OR (FROM_UNIXTIME('$swap_shift_start') BETWEEN shiftstart AND shiftend) OR (FROM_UNIXTIME('$swap_shift_end') BETWEEN shiftstart AND shiftend) ) "; The second one makes sure that the user isn't already working during the eligible shift. (Basically, the user wants to swap his Monday shift, I'm working on Thursday, but so is the user, so we can't swap.) $conflict_query = " SELECT * FROM postedshifts WHERE userid = '$userid' AND (('$startdate' BETWEEN shiftstart AND shiftend OR '$enddate' BETWEEN shiftstart AND shiftend) OR (shiftstart BETWEEN '$startdate' AND '$enddate' OR shiftend BETWEEN '$startdate' AND '$enddate')) "; Both queries are run for the eligible shift, and if no results are found that means there is no conflict, the script echo's the result: if ((mysql_num_rows(mysql_query($conflict_query)) == 0) && (mysql_num_rows(mysql_query($busy_query)) == 0)) print $row; So once the loop that checks for busys and conflicts is done, shifts that pass all the tests are output. The user can now email the people on the list, or whatever the script does next. So I was looking through the anti-pattern article, and I'm feeling hopeful, but I'm not really sure how (or if I can) get all the eligible shifts in one query instead of having to run a query for each potentially good result to make sure it doesn't conflict in some way. As I see it, I need some way to say "Hey, for that row you are considering, take the user ID and make sure that it doesn't show up on some other row with starttime/endtime conflicts...and oh, while you're at it, make sure that the user ID of the actual user doesn't show up anywhere with an overlap with the current row you are considering..." Honestly, I'm kind of proud I got this far. I feel like I'm way over my head. Thanks! A PS - I hope my php variables are fairly intuitive, but if there is any confusion, I can elaborate.

Good catch. Here is what it should be: /*nabbed from php.net */ function bool2str($bool) { if ($bool === false) { return 'FALSE'; } else { return 'TRUE'; } } function compareObjects(&$o1, &$o2) { echo $o1.' == '.$o2.' : ' . bool2str($o1 == $o2) . "\n"; echo $o1.' != '.$o2.' : ' . bool2str($o1 != $o2) . "\n"; echo $o1.' === '.$o2.' : ' . bool2str($o1 === $o2) . "\n"; echo $o1.' !== '.$o2.' : ' . bool2str($o1 !== $o2) . "\n"; } /* End of nabbed code */ $xml = simplexml_load_file("somefile.xml"); foreach ($xml->object as $value) { $some_name = $value->name; //function copied directly from: http://us2.php.net/manual/en/language.oop5.object-comparison.php compareObjects($some_name, $prev_name); $prev_name = $some_name; } Here is the output: Joe Smith == Joe Smith : FALSE Joe Smith != Joe Smith : TRUE Joe Smith === Joe Smith : FALSE Joe Smith !== Joe Smith : TRUE Doug Liszt == Joe Smith : FALSE Doug Liszt != Joe Smith : TRUE Doug Liszt === Joe Smith : FALSE Doug Liszt !== Joe Smith : TRUE Doug Liszt == Doug Liszt : FALSE Doug Liszt != Doug Liszt : TRUE Doug Liszt === Doug Liszt : FALSE Doug Liszt !== Doug Liszt : TRUE Obviously what I want is for the same value to return "True" so that I can say something like: if ($some_name !== $prev_name) echo $some_name;

I thought that I just didn't get the syntax, but now I think I don't get objects at all. Please keep in mind, when offering help, that I'm using a function that creates all of my objects because I'm still getting my mind around the very idea of classes and objects.... So imagine I have an xml file that is a series of jobs. Each job is assigned to a person, but instead of grouping the xml by person, the names are put in each job. So I want to output only the first instance of each name in my loop. This is not working at all. Take a look: /*nabbed from php.net */ function bool2str($bool) { if ($bool === false) { return 'FALSE'; } else { return 'TRUE'; } } function compareObjects(&$o1, &$o2) { echo $o1.' == '.$o2.' : ' . bool2str($o1 == $o2) . "\n"; echo $o1.' != '.$o2.' : ' . bool2str($o1 != $o2) . "\n"; echo $o1.' === '.$o2.' : ' . bool2str($o1 === $o2) . "\n"; echo $o1.' !== '.$o2.' : ' . bool2str($o1 !== $o2) . "\n"; } /* End of nabbed code */ $xml = simplexml_load_file("somefile.xml"); foreach ($xml->object as $value) { $some_name = $value->name; //function copied directly from: http://us2.php.net/manual/en/language.oop5.object-comparison.php compareObjects($con_name, $prev_name); $prev_name = $some_name; } The compareObjects function is at php.net and seems to work as it's supposed to. Here is the output: Joe Smith == Joe Smith : FALSE Joe Smith != Joe Smith : TRUE Joe Smith === Joe Smith : FALSE Joe Smith !== Joe Smith : TRUE Doug Liszt == Joe Smith : FALSE Doug Liszt != Joe Smith : TRUE Doug Liszt === Joe Smith : FALSE Doug Liszt !== Joe Smith : TRUE Doug Liszt == Doug Liszt : FALSE Doug Liszt != Doug Liszt : TRUE Doug Liszt === Doug Liszt : FALSE Doug Liszt !== Doug Liszt : TRUE Obviously what I want is for the same value to return "True" so that I can say something like: if ($some_name !== $prev_name) echo $some_name; What am I getting wrong? Obviously I'm missing a fundamental principle. Thanks for helping! A

Thanks for all the great feedback! It got my gears turning. I regret to admit that I had to use a for loop after all. Here's what I came up with: /* Gets the pattern for the main date and top shift */ $findshift = '/<day date=\"(.*)\">\n<shift>(<descr>.*<\/descr>)<starttime>(.*)<\/starttime><endtime>(.*)<\/endtime>/'; /* The replacement code...notice how the day tag is gone */ $fixshift ="<shift>$2<startime>$1 $3</startime><endtime>$1 $4</endtime>"; /*moves the date down to the first shift of that date */ $text = preg_replace($findshift, $fixshift, $text); /* looks for the top shift and the one below it */ $findshift2 = '/(\n<shift>.*time>(\d*\/\d*\/\d*).*<\/shift>\n)(<shift>(<descr>.*<\/descr>)<starttime>(\d{1,2}:\d{2} (AM|PM))<\/starttime><endtime>(.*)<\/endtime>)/'; /*returns the first shift and shift below, now both with dates */ $fixshift2 ="$1<shift>$4<startime>$2 $5</startime><endtime>$2 $6</endtime>"; /* tried regular preg_replace, but only worked on groups of two shifts per day. Tried doing it twice, and only got three dates inserted. Figured the likely hood of anyone doing more than 8 shifts in a day, so I have a loop that does it 8 times (the current record for this weeks data was 5 shifts, so better safe than sorry) */ for ($i = 1; $i <= 8; $i++) { $text = preg_replace($findshift2, $fixshift2, $text); } /* finally, delete all day close tags */ $text=str_replace("</day>\n", "", $text); If anyone knows a more slick way of doing this, feel free to school me. A

That would work for the first shift, but what about any others that follow? I could do another find/replace for the ones below using the first one (maybe?) but I'm not even sure that's an option. My original idea was to grab all of the "day" groups and then just make the date a variable that gets added to a regex, BUT I don't know how to get that array back into the string.

Thanks for replying. Sorry, this is all VERY new for me. I've got a ton more problems, but this one may lead me to a generally better understanding. Okay, so the input is: <schedule> <employee name="Joe Smith"> <day date="10/13/2008"> <shift><descr>Sweeping</descr><starttime>7:45 AM</starttime><endtime>12:15 PM</endtime></shift> <shift><descr>Yodeling</descr><starttime>1:15 PM</starttime><endtime>6:00 PM</endtime></shift> </day> <day date="10/14/2008"> <shift><descr>Mopping</descr><starttime>7:45 AM</starttime><endtime>11:15 AM</endtime></shift> </day> </employee> <consultant name="Butternut McClintock"> <day date="10/13/2008"> <shift><descr>Cliff Diving</descr><starttime>7:45 AM</starttime><endtime>12:15 PM</endtime></shift> </day> <day date="10/14/2008"> <shift><descr>Chewing Food</descr><starttime>7:45 AM</starttime><endtime>11:15 AM</endtime></shift> <shift><descr>Crackin' Wise</descr><starttime>1:15 PM</starttime><endtime>6:00 PM</endtime></shift> </day> </employee> </schedule> The output (or a sampling of it) would be: <schedule> <employee name="Joe Smith"> <shift><descr>Sweeping</descr><starttime>10/13/2008 7:45 AM</starttime><endtime>10/13/2008 12:15 PM</endtime></shift> <shift><descr>Yodeling</descr><starttime>10/13/2008 1:15 PM</starttime><endtime>10/13/2008 6:00 PM</endtime></shift> <shift><descr>Mopping</descr><starttime>10/14/2008 7:45 AM</starttime><endtime>10/14/2008 11:15 AM</endtime></shift> </employee> </schedule>

I've written a script that converts a source of data into XML. Now it follows a pretty standard pattern but does it is not consistent (that sounded weird...) Here's an example: <schedule> <employee name="Joe Smith"> <day date="10/13/2008"> <shift><descr>Sweeping</descr><starttime>7:45 AM</starttime><endtime>12:15 PM</endtime></shift> <shift><descr>Yodeling</descr><starttime>1:15 PM</starttime><endtime>6:00 PM</endtime></shift> </day> <day date="10/14/2008"> <shift><descr>Mopping</descr><starttime>7:45 AM</starttime><endtime>11:15 AM</endtime></shift> </day> </employee> <consultant name="Butternut McClintock"> <day date="10/13/2008"> <shift><descr>Cliff Diving</descr><starttime>7:45 AM</starttime><endtime>12:15 PM</endtime></shift> </day> <day date="10/14/2008"> <shift><descr>Chewing Food</descr><starttime>7:45 AM</starttime><endtime>11:15 AM</endtime></shift> <shift><descr>Crackin' Wise</descr><starttime>1:15 PM</starttime><endtime>6:00 PM</endtime></shift> </day> </employee> </schedule> The problem is that I want to move the date in front of the appropriate times (start and end times) so that I can ditch the day tag and just have a list of shifts per employee. But I don't want to create arrays (yet) and I can't use regex (I don't think) because the number of shifts changes per day, so it's not regular enough. Is there a way, preferably without using loops, to just say "find this regular expression, and while you're at it, find this OTHER regular expression that comes after it, and then stop looking when you get to this end point" ? I thought maybe I could use a * or + to imply that the entire section repeats, but that got really confusing and kept throwing errors. Help! Please! Thanks