rspec/rules/S2108/swift/rule.adoc

47 lines
2.0 KiB
Plaintext
Raw Normal View History

== Why is this an issue?
2021-01-27 13:42:22 +01:00
Variables that are never updated will always return their default values and so they should be explicitly declared as constant. A ``++let++``-declaration guarantees and clearly signals to the programmer that its value is supposed to and will never change.
2021-02-02 15:02:10 +01:00
2020-06-30 12:48:07 +02:00
This rule applies to non-constant fields and variables which are not set within the codebase.
ifdef::env-github,rspecator-view[]
'''
== Implementation Specification
(visible only on this page)
=== Message
Declare variable "{0}" with "let".
'''
== Comments And Links
(visible only on this page)
=== on 15 May 2015, 11:57:27 Elena Vilchik wrote:
Consider only "set and read but never updated" variables and not "unused" one.
=== on 8 Oct 2014, 07:20:32 Nicolas Peru wrote:
\[~ann.campbell.2]Why non-private ?
=== on 8 Oct 2014, 12:16:15 Ann Campbell wrote:
\[~nicolas.peru] we already have a separate rule about private fields: RSPEC-1068, and because the advice is different. If they're private and not used, they're dead code. If they're not private and _apparently_ unused, they might still be accessed directly by classes that just aren't currently 'visible'.
=== on 10 Oct 2014, 14:15:18 Freddy Mallet wrote:
@Ann :
* I would use the label "pitfall"
* Moreover for me there is no intersection between this rule and rule RSPEC-1068. In case of RSPEC-1068 we're looking for unused fields whereas here we're looking for fields that ARE really used but never updated and so which should be declared as 'final'
* I would decrease the remediation cost to 2 min and associate this rule to the Maintainability > Understandability characteristic
* This rule would be limited at first to private fields but I would not mention this limitation
=== on 15 May 2015, 09:53:17 Elena Vilchik wrote:
\[~ann.campbell.2] I want to expand the scope of this rule for Swift "Fields *and variables* that are never updated should be constant". WDYT can I do it in subtask for this rule?
=== on 15 May 2015, 11:52:30 Ann Campbell wrote:
Fine with me [~elena.vilchik]
endif::env-github,rspecator-view[]