[NFN] Capomaestro writes:
How about the retrieval operation? 'Data Areas' maps the PXT and PPD with
VSM (IGV-prefixed) macros. But even the cell pool callable services
(CSRxxxxx) do not have a mechanism AFAICS.
As I'm sure you know,
o CSRP[4]BLD constructs a cell pool and sets a pointer to it
o CSRP[4]GET et al. obtrain a cell from a nominated pool, and
o CSRPFRE or CSRC4FRE et al. return a cell to its pool for notional reuse.
After that you're on your own. For my own purposes, I organize sets of
cells into (a) a set of hashed ordered sublists, (b) a binary-search trees,
or the like. I find it convenient to manage these lists in completely
decoupled fashion, using chaining blocks EXTERNAL to them of such forms as
declare 1 lcb based(cbp), /* list chaining block */
2 nxttp pointer, /* forward chaining */
2 cellp pointer, /* -> cell */
2 keybp pointer ; /* -> key block */
declare 1 bstcb based(bstcbp), /* BST chaining block */
2 leftp pointer, /* -> left subtree */
2 rghtp pointer, /* -> right dsubtree */
2 cellp[ pointer, /* -> -> cell */
2 keybp pointer ; /* -> key block */
and the like. But these are the schemes [and routines] I use to manage
other sorts of lists too, e.g., those that occupy STORAGE-allocated and
-freed storage. The point, to repeat myself now, is that you are on your
own and should be.
Worth mentioning---kudos to some group within IBM---is that the 64-bit
versions of these macros have been surprisingly trouble-free for me.
John Gilmore
Ashland, MA 01721
USA
_________________________________________________________________
Dont just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html