>Yeah, but here's the thing. The tables hold data that can be stuck into >"key, title, body" columns, but the display of each column is different, so >each collection will be displayed differently, thus the "custom1" column in >my cfsearch. I want to mix the results for display purposes, showing scores >correctly ranked when mixed, but I want to display the results differently. >My solution was to search each collection, place the results into a >structure of some sort, including the "custom1" column that will determine >how to display the data on the details page, and sort them all by "score" so >that output results are "sorted" by "score" yet their "colum1" information >will determine how the data is displayed on the following page. > >Make sense? > >So *can* you do that? stick each result set into some sort of data structure >for each query returned, then sort that structure, ordering by "score", then >outputting the results? > >For instance, collection 1 returns the following: >Score - Title (custom1) >97 - My New Title (1) >67 - A Second Title (1) >78 - The Fifth Title (1) > >And collection 2 returns the following: >86 - The Title of This Document is This (2) >76 - I'm Titling This Document Thusly (2) >23 - A Whimsical Title To Enjoy (2) > >Then my output is as follows: >97 - My New Title (1) >86 - The Title of This Document is This (2) >78 - The Fifth Title (1) >76 - I'm Titling This Document Thusly (2) >67 - A Second Title (1) >23 - A Whimsical Title To Enjoy (2) > >See? Then, when you clicked the link, you'd know *which* collection passed >you the results...then you could display it appropriately, if it needs to be >displayed differently (which mine does).
Perhaps you could avoid trying to put the two queries into one structure... What you want to do is *exactly* like the merge portion of the merge sort algorithm - the problem is to take two already-sorted data structures and combine them into a single sorted structure. Merge sorts are known to be very fast, so perhaps an efficient solution would be to mimic the merge part of the merge sort algorithm. Here's the basic idea: list A = 1, 3, 4, 6 list B = 2, 3, 5, 7 merge into: 1 2 3 3 4 5 6 7 to merge the two lists: loop while A and B have elements remaining --compare the first elements of A and B --if B is less, append to final list --else, append A's first element instead end loop So to mimic this: Initialize two counters - counterA and counterB - to 1 -- you begin by looking at the first record from each query loop while both counters are LTE their respective recordcount --compare queryA.score[counterA] and queryB.score[counterB] --if the score from B is GT the score from A ----output info for B, iterate counterB --else ----output info for A, iterate counterA end loop One thing to remember - you can go through all the records of one query before the other query is finished so you need to test for that. Something like that. Data structures was my favorite class in college :-) Mike Mertsock Alfred University Web Team ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Archives: http://www.houseoffusion.com/lists.cfm?link=t:4 Subscription: http://www.houseoffusion.com/lists.cfm?link=s:4 Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4 This list and all House of Fusion resources hosted by CFHosting.com. The place for dependable ColdFusion Hosting. http://www.cfhosting.com
