The Big Programming Thread - Page 625

On April 30 2015 11:36 Neo7 wrote:
I'd throw the references into some template. Use a developer mode flag to determine whether to import the template file or reference the production js file based on what environment you're in.

On April 30 2015 16:02 Isualin wrote:
Microsoft reveals visual studio code It's own site doesn't load for me though. I am using eclipse for java, sublime for everything else right now but i might give it a try

On April 30 2015 18:51 Manit0u wrote:
I need a bit of help with MySQL...

What I have:
A table where rows can have a one-to-one relation with other rows in the same table.

What I need:
Get all the rows that either don't have the relation, or if they do I want to only return one side of it.

Where I'm at now:

/*
 *    relation:
 *        inversedBy: otherSide
 *    otherSide:
 *        mappedBy: relation
 */
SELECT a.id AS id, (a.val * 0.10) AS value
FROM table AS a
LEFT JOIN a.otherSide AS b
WHERE b.relation IS NULL

What this gives me is all the rows without the relation. Now I need to get this other side of the relation (inverse side is the owning side) for those that have it, and it has to be included in all the id, value etc. in the results on equal terms as rows without relation.

Help anyone? I've been churning this around so much that my brain boils and I don't know what I'm talking about any more. And I suck horribly at SQL.

On April 30 2015 20:16 berated- wrote:


Use a union? I think thats what you want..

https://dev.mysql.com/doc/refman/5.0/en/union.html

(SELECT a.id AS id, (a.val * 0.10) AS value
FROM table AS a LEFT JOIN table AS b ON a.relation = b.otherSide
WHERE b.relation IS NULL)
UNION
(SELECT a.id AS id, (a.val * 0.10) AS value
FROM table AS a INNER JOIN table AS b ON a.relation = b.otherSide)

On April 30 2015 20:35 Acrofales wrote:

Yeah. Union should do it. Write a second query to get your other results and UNION them:

(SELECT a.id AS id, (a.val * 0.10) AS value
FROM table AS a LEFT JOIN table AS b ON a.relation = b.otherSide
WHERE b.relation IS NULL)
UNION
(SELECT a.id AS id, (a.val * 0.10) AS value
FROM table AS a INNER JOIN table AS b ON a.relation = b.otherSide)

I also don't really understand what was happening in your original query, so I think I fixed it.

That said, this seems like a pretty bad way of modeling this relationship, because there is no way of enforcing the 1-1.0 relationship. Instead, remove the otherSide (it is unnecessary), and add a UNIQUE constraint on relation. I am not sure MySQL allows a foreign key to the same table. If it does, add that one in as well, and you are guaranteed your 1-1,0 relationship.

On April 30 2015 21:19 berated- wrote:


I think you might have flipped the condition a.other side not b.other side. If b.relation being null means it didn't join, then I would think it would be in the join. either way thats why I didn't elaborate and just left it at use union

SELECT id AS id, (val * 0.10) AS value
FROM table WHERE related IS NULL

On April 30 2015 21:32 Acrofales wrote:


You're probably right. It should be WHERE b.otherSide IS NULL. What the LEFT OUTER JOIN (same as LEFT JOIN) query does is to join every row in a with every row in b that matches the ON criteria. If there is no ON criteria it is simply a Cartesian product (as in the original query given, which I think is wrong for that reason). If there is an ON criteria then, if there is a row in a that has no matching result in b, these are left blank (null).

Now that I think about it, I am not sure why we are joining in the first query at all. Shouldn't:

SELECT id AS id, (val * 0.10) AS value
FROM table WHERE related IS NULL

work just fine?

On April 30 2015 23:26 netherh wrote:
I'm not sure there's any "good" way to read in a file format like that. Can you persuade your friend to use xml or json or lua... some actual data description language? (It'll be easier for him in the long run too).

On April 30 2015 23:33 Artesimo wrote:
Unfortunately he only codes in java since its the only thing he knows. He also does it as a hobby since his work isnt related to IT in any form (even though he still is better than me at coding... for now ). You think I should stick with my regex-approach? What about the vector-stuff, should I search for a better container or does my method of stacking vectors just seem unintuitive to me because I am not used to it? At the moment I fondled a little bit around with "map", since it could let me get rid of the multiple vectors and also give me an ability to store the locations of said items on top of the values that belong to them.

struct Item
{
    std::vector<std::string> tags;
    unsigned int price;
};

struct Location
{
    std::vector<Item> items;
};

On April 30 2015 23:33 Artesimo wrote:


Unfortunately he only codes in java since its the only thing he knows. He also does it as a hobby since his work isnt related to IT in any form (even though he still is better than me at coding... for now ). You think I should stick with my regex-approach? What about the vector-stuff, should I search for a better container or does my method of stacking vectors just seem unintuitive to me because I am not used to it? At the moment I fondled a little bit around with "map", since it could let me get rid of the multiple vectors and also give me an ability to store the locations of said items on top of the values that belong to them.

EDIT: Thanks ropid, I can assure you that I already solved the problem using regex before asking for help. I was just searching for optimization since it felt a little overkill since my iterator has to check on every word "does it has letters? yes? good. does a single space come afterwards and then more letters? must be a multiple-words name/attribute then. does it start with a number? good, get it, grab it all and split it up. oh, it is just spaces? just ignore the findings and put a blank in the names-vector for that line". Sometimes I have problems understanding what is "hard" for a machine and whats not^^.


EDIT2: I figured a linked list is what I propably want and most likely would be the "most professional" thing to store this. I am going to use vector first until someone advices me otherwise. In addition, here is a chart of what I am planing to do:

On April 30 2015 06:51 darkness wrote:
Java is a really good language with really good syntax for beginners. As soon as you want to have more control, Java becomes terrible. No deterministic destructor (finalize is unreliable), no deterministic memory deallocation, no pass by reference unless you pass an object. C# solves that by having 'ref' and 'out' though. In a way, C# seems like better or more advanced Java.