What is the most efficient solution for reading properties from a properties file by prefix?

Question

What is the most efficient (in terms of time complexity) solution for reading properties from a properties file by prefix?

For example, if the properties file looks like this -

prefix1.prop1=val1
prefix1.prop2=val2
prefix2.prop3=val3
prefix2.prop4=val4

I am looking a for a method which when called like this - getPropertiesByPrefix("prefix1") should return the following list :

["prefix1.prop1", "prefix1.prop2"]

I can see 2 options -

1. On every getPropertiesByPrefix call, read all the properties one by one and get the props starting with prefix.
2. Read properties once and build a trie and then getPropertiesByPrefix uses the trie.

The 2nd option seems to be the more efficient way to go. Are there any existing implementations/third-party libs for these? Or a third option?

Answer 1

The analysis are simple.

Say you have properties of n number of keys, then a loop that does key.startsWith(prefix) would run in O(m*n) time, m being the size of the prefix.

On the other hand, if you build a Trie , this could be reduced O(m) which is clearly better.

Apache Commons have some implementations, although if n is small here then I really won't bother and avoid adding complexity to my code and go with a simple loop.