Home/ Questions/Q 176279
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-11T13:50:53+00:00

I’ve seen different questions on SO about not being able to use parameterless constructors

  • 0

I’ve seen different questions on SO about not being able to use parameterless constructors or not setting field initializers, but I think my question kind of goes a step beyond that.

What I would like to know is, how would I go about setting up the ‘default’ value of a struct I need to define? Say I’m making a TaxID struct, and the default value needs to be 999-99-9999 for whatever reason. This is already done with other structs in .NET, namely the DateTime. Any time you declare a DateTime the value is immediately set to Jan 1 0001 12:00 AM. Why does it get set to this value and not 0/0/0000 00:00:0000 AM? There must be something going on behind the scenes that makes the value actually make ‘sense’ at it’s ‘default’, even given the restrictions put on us by c# with regards to structs.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. In general, this should be avoided. If you need to require a good, default case which would require construction, you should consider changing to a class.

    One of the core design guidelines for structs is: Do ensure that a state where all instance data is set to zero, false, or null (as appropriate) is valid.

    For details, see the design guidelines.

    I just looked this section up in the design guidelines 2nd edition, and they have an example in detail there using properties and non-conventional overrides to work around this, as well. The basic concept was to save the value privately in a way that 0 is the ‘good default’, and do some form of transform in every property and method override (including ToString). In their case, they used a positive integer as an example, and always save curVal-1 in the private member, so the default value (0) is treated like a value of 1. They added a constructor with an int val, and save value-1 internally, etc.

    This seems like a lot of hidden, unexpected overhead, though – so I’d personally still recommend (and use) a class in this case.

    — Edit in response to comments —

    DateTime, as your example, is the way it is because 0 == 1/1/0001 at Midnight. DateTime uses a single, ulong to represent ticks from 1/1/0001. That is why (from here):

    ‘The DateTime value type represents dates and times with values ranging from 12:00:00 midnight, January 1, 0001 Anno Domini (Common Era) through 11:59:59 P.M., December 31, 9999 A.D. (C.E.) ‘

    This is the full range of ulong in ticks. A ‘0’ in the struct of DateTime is treated as 1/1/0001 when you convert to a string – the values aren’t 1 + 1 + …. – it’s a single 0.

    • 0