Re: Safety of unprotected concurrent operations on constant objects

["Randy Brukardt" <randy@rrsoftware.com>]

"Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote in message 
news:1jbqe7sd9d13f.1p91l6pdkv6fz.dlg@40tude.net...
> On Sun, 04 May 2014 14:25:26 -0600, Shark8 wrote:
>
>> On 04-May-14 01:46, Natasha Kerensikova wrote:
>>> I have some shared resources indexed by a string, described in a
>>> file, and considered as constant throughout the lifetime of the program
>>> (with the standard scheme "restart for changes to take effect").
>>
>> What's the problem, then, with making the entire map constant?
>
> That is not enough. Technically speaking all operations must be reentrant.
> All arguments being immutable does not make a subprogram reentrant.

Right. And "constant" doesn't have that effect (of immutability) for private 
types anyway. It just prevents assigning to the whole object *from the 
current view*. But it's easy to get a variable view for immutably limited 
types, and not hard for controlled types. Not to mention access-to-variable 
components. And you cannot assume that there aren't any controlled 
components or access-to-variable components for a private type.

                   Randy.