greelmo - Developer IT

How to persist a very abstract data type between sessions: PHP

- by Greelmo

I have an abstract data type that behaves much like stack. It represents a history of "graph objects" made by a particular user. Each "graph object" holds one or more "lines", a date range, keys, and a title. Each "line" holds a sql generator configured for a particular subset of data in my db. I would like for these "histories" to be available to users between their sessions. It will be in the form of a tab that reads something like "most recent graphs". What do you believe to be the best way to persist this type of data between sessions. This application could get rather large, so efficiency is a concern. Thanks in advance.

Read the article

Using php's magic methods outside a class

- by Greelmo

Is it possible to use PHP magic methods (specifically __get()) outside a defined class? I'm wanting to use it in a configuration file for quick loading. The configuration file has a single array, $config, with many keys. Therefore, I'd like to override __get() to return the key in the array.

Read the article

Kohana 3 ORM - grouping where conditions with parentheses

- by Greelmo

I'm trying to run a query through the ORM like this: SELECT * from table where (fname like 'string%' or lname like 'string%') AND (fname like 'string2%' or lname like 'string2%'); Here's what i have so far: $results = ORM::factory('profiles'); foreach ($strings as $string) { $result->where('fname', 'like', "$string%"); $result->or_where('lname', 'like', "$string%"); } But this doesn't account for the parentheses. Any ideas?

Read the article

MySQL database design question

- by Greelmo

I'm trying to weigh the pros and cons of a database design, and would like to get some feedback as to the best approach. Here is the situation: I have users of my system that have only a few required items (username, password). They can then supply a lot of optional information. This optional information continues to grow as the system grows, so I want to do it in such a way that adding new optional information is easy. Currently, I have a separate table for each piece of information. For example, there's a table called 'names' that holds 'user_id', 'first_name', and 'last_name'. There's 'address', 'occupation', etc. You get the drift. In most cases, when I talk to my database, I'm looking only for users with one particular qualifier (name, address, etc.). However, there are instances when I want to see what information a user has set. The 'edit account' page, for example, must run queries for each piece of information it wants. Is this wasteful? Is there a way I can structure my queries or my database to make it so I never have to do one query for each piece of information like that without getting my tables to huge? If i want to add 'marital status', how hard will that be if I don't have a one-table-per-attribute system? Thanks in advance.

Read the article

Getting the names of all files in a directory with PHP

- by Greelmo

For some reason, I keep getting a '1' for the file names with this code: if (is_dir($log_directory)) { if ($handle = opendir($log_directory)) { while($file = readdir($handle) !== FALSE) { $results_array[] = $file; } closedir($handle); } } When I echo each element in $results_array, I get a bunch of '1's, not the name of the file. How do I get the name of the files?

Read the article

Database Design for One to One relationships

- by Greelmo

I'm trying to finalize my design of the data model for my project, and am having difficulty figuring out which way to go with it. I have a table of users, and an undetermined number of attributes that apply to that user. The attributes are in almost every case optional, so null values are allowed. Each of these attributes are one to one for the user. Should I put them on the same table, and keep adding columns when attributes are added (making the user table quite wide), or should I put each attribute on a separate table with a foreign key to the user table. I have decided against using the EAV model. Thanks!

Read the article

How can I get the rank of rows relative to total number of rows based on a field?

- by Arms

I have a scores table that has two fields: user_id score I'm fetching specific rows that match a list of user_id's. How can I determine a rank for each row relative to the total number of rows, based on score? The rows in the result set are not necessarily sequential (the scores will vary widely from one row to the next). I'm not sure if this matters, but user_id is a unique field. Edit @Greelmo I'm already ordering the rows. If I fetch 15 rows, I don't want the rank to be 1-15. I need it to be the position of that row compared against the entire table by the score property. So if I have 200 rows, one row's rank may be 3 and another may be 179 (these are arbitrary #'s for example only). Edit 2 I'm having some luck with this query, but I actually want to avoid ties SELECT s.score , s.created_at , u.name , u.location , u.icon_id , u.photo , (SELECT COUNT(*) + 1 FROM scores WHERE score > s.score) AS rank FROM scores s LEFT JOIN users u ON u.uID = s.user_id ORDER BY s.score DESC , s.created_at DESC LIMIT 15 If two or more rows have the same score, I want the latest one (or earliest - I don't care) to be ranked higher. I tried modifying the subquery with AND id > s.id but that ended up giving me an unexpected result set and different ties.

Developer IT