[Date Prev][Date Next]
[Thread Prev][Thread Next]
[Date Index]
[Thread Index]
[New search]
Subject: Re: [austechwriter] Merging of Framemaker files on Clearcase
From: Chris Despopoulos <cud@xxxxxxxxxxxx>
Date: Mon, 08 Apr 2002 11:05:24 +0200
References: <3cb2e273.19956515@smtp.omsys.com>
Sender: owner-framers@xxxxxxxxx
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:0.9.4) Gecko/20011128 Netscape6/6.2.1
Just to clarify... I believe Hedley is talking about two different IDs - there's the unique ID and the table/aframe ID. The unique ID is used by Maker to track any document object persistently across sessions. I believe MIF is loose about that - if you have a duplicate UID in a MIF document then it should notice that and create a new UID for the conflicting objetc. If you have some data repository that makes use of these UIDs, then that will be hosed. But I'm not aware of this use for UIDs in actual practice... I don't believe it's very common. On the other hand, table and aframe IDs are used to identify the location in a flow for a table or aframe. These IDs are serial - tbl1, tbl2, tbl3, etc. I have received generated MIF files that get the sequence mixed up - the result is missing tables in the document. So, for example, if you have two tables in your MIF with the same ID, one of them is likely to never appear in your document. I personally view CVS (or other source control) on MIF with dread. I would rather think about SGML - it's safer. cud >Because the <Unique ...> tags occupy a single line, the change of the ID is >treated as a conflict. VSS recognizes the conflict, asks if conflicts have >been resolved (to which you lyingly reply 'yes' even though you haven't >resolved conflicts), and then applies the IDs from one of the files to the >merged result. > >Admittedly, I have tested this only on short documents of three or four >pages, to which I made substantial changes (paras, chars, tables, graphics) >before merging. In all cases FrameMaker opened the merged result without >complaint. > >What could be problematic is that different paragraphs or tables in the two >branches have the same IDs. Since the ID of a table is used in the anchor >marker to point to the 'real' table located at the beginning of the file, >there is a potential for the anchor marker to point to two different tables >with the same ID in the merged result. > >Hence my caveat to obtain an evaluation copy of FrameMaker to put your >repository manager (ClearCase, CVS, VSS, or whatever) before committing >(get it?) to purchase. > >If all else fails you can use FrameMaker's document comparison to merge >branched files. > >Regards, >Hedley > ** To unsubscribe, send a message to majordomo@omsys.com ** ** with "unsubscribe framers" (no quotes) in the body. **