From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: ***
X-Spam-Status: No, score=3.8 required=5.0 tests=BAYES_00,INVALID_MSGID,
	RATWARE_MS_HASH,RATWARE_OUTLOOK_NONAME autolearn=no autolearn_force=no
	version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,3606a072272ee7d2,start
X-Google-Attributes: gid103376,public
From: "Pat Rogers" <progers@acm.org>
Subject: differentiating finalization causes
Date: 1997/10/29
Message-ID: <01bce4a2$b0ad17e0$020c6dce@my-pc.neosoft.com>#1/1
X-Deja-AN: 285823999
Organization: Software Arts and Sciences
Newsgroups: comp.lang.ada
Date: 1997-10-29T00:00:00+00:00
List-Id: <comp.lang.ada>


For a type descended from Controlled, I need to be able to differentiate
finalization due to object destruction from finalization due to assignment
("value destruction", if you will).  At present I use the subsequent call
to Adjust to set a flag, with other routines checking the flag to see if
the effect of Finalize was really due to assignment.  However, Finalize is
responsible for placing an item in a list, and Adjust must search the list,
looking for the corresponding node (in which to set the flag).  I don't
want to search this list each time Adjust is called, but of course I cannot
keep a pointer to the node in the controlled object itself, since
assignment will overwrite it with the value on the right-hand side.  Are
there any "tricks" or techniques for differentiating between the two causes
of finalization?