funkybeat

Hi xylex, so you would recommend MongoDB ? The only reason for me to use MySql is because I know how to use it and it's already running on my server. However I'm more than happy to try something new. I just went through their website and it seems I'll need some more time to understand all the details of mapreduce & sharding (never heard this before...) thanks for throwing this in! Michael

Hi Chris & Kicken, thx for your feedback! In this case I'd go for a more array based solution instead of bashing the server with too many SQL requests. The dictionary is around 7MB so it should be doable to put everything into the array and check it with isset(). Cheers any many thanks for your help! Mike

Hi phpFreaks! To improve my php skills I do some private coding for my own. I did some small php scripts in the past, however never worked with the "search/DB" topic. This is where someone may help me a bit to give me a good start for my next project: "A scrabble search engine based on a dictionary with php & mysql" I'll try to do this as most efficient (fast) as possible. Requirements: > "field to enter random 1-14 letters" to search for words in the DB > "Dictionary in Database" or in file but I dont think that may be faster... My plan: > Put every single word into a database table called "words" (EDIT: maybe I should add a second row and save the "word length" of each word as well) > split up the 1-14 letters into a letter array > build a loop which groups the 1-14 letterArray together in all possible combinations: > 1st run: two-party groups -> check with "like" if its in the database for every single entry and put the result in a resultArray (EDIT: only check the words with the "word length" = 2) > 2nd run: three-party groups -> check with "like" if its in the database for every single entry and attach the result to the resultArray (EDIT: only check the words with the "word length" = 3) .....>14th run... > output resultArray I'm not looking for code parts, I'm just looking for a good concept to start with. Does someone have any recommendations on how I could do it better than the above? In my eyes there are too many DB search requests and maybe there is a better way to do it? Cheers Mike