TMG-L Archives

Archiver > TMG > 2004-09 > 1094126517


From: "John Cardinal" <>
Subject: RE: [TMG] Entering / Selecting Places
Date: Thu, 2 Sep 2004 08:01:57 -0400
In-Reply-To: <6.0.2.0.0.20040902054606.02602640@pop.sprynet.com>


Darrell Martin wrote:
> However, I am quite certain that if F2 is pressed with the cursor in
> a place field "later" than the first, and then one of the "earlier"
> fields is edited, the behavior is not always what one would expect of
> a true Master Place List stored as a separate file (because TMG does
> not store its "MPL" that way). In this regard, I will have to review
> the specifics of what fields are, and are not, changed by Tag screen
> edits (L1 through L9 versus Short Place, Place Style, comments, and
> date range).

Darrell,

TMG does have a Master Place List stored as a separate file. The _p table in your project folder is precisely that. That table is the top of a structure that groups the place field data into distinct master place list entries. However, when you use the F2 function, you are not shown one-for-one entries from the MPL. I believe what you see is a list of the unique combinations of fields that occur in the MPL entries where the fields are limited to the remaining place fields on the data entry screen.

So, for example, if the cursor is in the L3 field (regardless of the label for it), the F2 list will show the unique combinations of the L3-L10 fields. If you select one of those entries, that sets the value of the L3-L10 fields on the tag entry screen. If you save the tag, TMG will analyze the place data to determine whether a new, unique master place entry is needed. If there is an exact match of the field data, then an existing MPL entry is linked to the tag. Otherwise, a new MPL entry is created and linked to the tag.

I know the MPL data structures quite well and I stand behind the statement that TMG has a true MPL. I am less certain about all the data entry behaviors and how the MPL is used during the F2 processing, but a couple tests this morning seem to support what I thought was the behavior and I described above.

John
That's my understanding


This thread: