CODEWITHSUNDEEP
databasehtmlindexeddb
JMaylin
JMaylin

Reputation: 1406

How do I think in "IndexedDB"?

Let's say I have 3 entries in my store

{
  category: "Science",
  author: "Charles Darwin",
  content: "Lorem ipsum dolor..."
}

{
  category: "Science",
  author: "Albert Einstein",
  content: "sit amet..."
}

{
  category: "Philosophy",
  author: "Albert Einstein",
  content: "consectetur adipisicing elit..."
}

Upvotes: 3

Views: 132

Answers (1)

dumbmatter
dumbmatter

Reputation: 9683

Is it possible to get all the entries "where category = 'Science'", without creating an index on category?

Yes: just iterate through the entire object store and manually look for objects with that category. Even if you write some convenience function to do this easily, it'll still be pretty slow compared to using an index. So in practice, you would want to use an index there.

Is it possible to get all the entries "where category = 'Science' AND author = 'Albert Einstein'"?

Yes, you can use a "compound index" as is described in this question here. However, the big caveat is that it's not supported in IE10. If that is a problem for your application, then you can only index on individual fields. For any other constraints besides one indexed field, you'll have to iterate through all the results and manually compare. There are various libraries built on top of IndexedDB that aim to make this easier, but I haven't used any of them so I can't help you there. Either way, it's going to be pretty slow if you can't use compound indexes.

Is it possible to get all the entries that contain the word "Lorem" is the content field?

You might be noticing a pattern here... Yes: just iterate through the entire object store and manually look for objects with "Lorem" in the content field. There is no special support built in for full text searching. Theoretically one could imagine a full text search API built on top of IndexedDB (just keep track of all the words), but I'm not sure if anyone has built that yet (and if they have built it, I'm not sure how performance would be).

Or should I use a different database for these kinds of queries

If you need to do a lot of queries like this (or, God forbid, even more complex queries) and you have the option of using a different database, then use a different database. But you might not have that option, depending on what you're trying to accomplish.

Upvotes: 1

Related Questions