list.IsEmpty() vs list.Size() > 0 vs list != null
Generally, your lists should always be initialized. One common misconception is that queries can produce a null list, which is never true.
For example, consider the following code:
Account[] records = [SELECT Id, Name, AccountNumber FROM Account];
if(records!=null && !records.isEmpty()) {
for(Account record: records) {
// Do Something Here
}
}
This is a waste of CPU time, because (a) records will never be null, and (b) iterating over an empty list is acceptable. Therefore, none of the three checks you might use are rarely appropriate. The only exception to this rule is when one query is used to fuel another query, in which case you would check for the presence of values before performing the query:
Account[] accounts = [SELECT Id, Name, AccountNumber FROM Account];
if(accounts.isEmpty()) {
return;
}
Contact[] contacts = [SELECT Id, Name, Email FROM Contact WHERE AccountId IN :accounts];
For lists you create yourself, you should always initialize them before using them. This is true for other types, like sets and maps, as well. There are, however, times when you'll need to check for a specific condition before using some lists. It helps if you experiment with the various types to get a feel for when you should be aware that a list may be null.
For example, consider a variable that binds to a multi-select picklist:
public String[] selectedValues { get; set; }
In this case, selectedValues will either be null, or will not be empty. You only ever need to check for != null
in this case. Part of becoming an efficient developer is recognizing the patterns in a system. If you're still using myVar != null && !myVar.isEmpty()
as a condition, that generally means you're making one or even two checks too many.
There is always one correct condition to use in every situation. If this isn't true, that suggests that values were not properly initialized elsewhere.
It is generally best to avoid null values in code so using null list references to indicate empty lists isn't usually good approach. That is this code:
List<String> strings = new List<String>();
...
for (String item : strings) {
...
}
is clearer and safer than this code:
List<String> strings = null;
...
if (strings != null) {
for (String item : strings) {
...
}
}
So prefer empty lists to null list references.
Personally I find myList.size() > 0
to be clearer than !myList.isEmpty()
because the latter amounts to "not is empty" which takes a moment to understand. But as people will use both forms get used to both.
Personally I prefer myList.isEmpty()
. myList.size()
will throw error if myList is null. myList != null
will pass if the list is initialized and empty, which is not what you want.
So usually a combination condition (myList != null && !myList.isEmpty())
or (myList != null && myList.size()>0)