[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference iosg::all-in-1_v30

Title:*OLD* ALL-IN-1 (tm) Support Conference
Notice:Closed - See Note 4331.l to move to IOSG::ALL-IN-1
Moderator:IOSG::PYE
Created:Thu Jan 30 1992
Last Modified:Tue Jan 23 1996
Last Successful Update:Fri Jun 06 1997
Number of topics:4343
Total number of notes:18308

1133.0. "CM ME Merge Element problems." by COMICS::TALBOT (Trevor Talbot) Tue Jul 28 1992 12:00

    Hi,
    
    	Can anyone confirm this as a typo in a script or is it a problem on
    my system:
    
    In the OA$LIB:CM_PRE_EDIT_FRM.SCP
    
    A line of: YESNO_PROMPT CM$_E_MERGE_CON , CM$_YESNO
    
    should this be: YESNO_PROMPT CM$_E_MRG_CON , CM$_YESNO
                                       ^^^
    				Change here?
    
    I found this whilst looking into a ME merge Element problem. This
    problem, I would also like help on, when merging elements from 3
    application areas and assuming I enter yes to edit on conflicts,
    the edit session contains the asterisks outling conflicts but also
    contains lots of reverse video, solid blocks and other garbage... how
    am I supposed to make sense of that and edit the element to remove the
    conflicts as the scanty documentation in this area advises? 
    
    -Trev
T.RTitleUserPersonal
Name
DateLines
1133.1CMS merge outputeFAILTE::LAAHSAn accumulation of CeltsTue Jul 28 1992 12:2317
    Trevor,
    
    Looking at OA$BUILD:CM.A1$MSG it would appear that the symbol should
    be _MER_ and NOT _MERGE_ OR _MRG_. Looks like a bugette.
    
    As to the actual contents of the element when it has a merge conflict.
    The contents are the direct output of the CMS MERGE operation so it
    must be putting in the reverse video stuff. I *assume* you are trying
    to merge a form? If so, it is the FLG that is actually merged.  Can you
    confirm that it is an FLG you end up with?
    
    You should have a file in the development area
    OA$SITE_DEV_<area>:<name>.FLG_MERGE  which
    contains the merged conflict. If you edit it directly does it still
    have the reverse video etc?
    
    Kevin
1133.2It's a bugMAULS::REDMONDThoughts of an Idle MindTue Jul 28 1992 12:249
Yes, it looks like a bug. CM$_E_MRG_CON seems to be the correct symbol to 
use. Easy to fix - insert the symbol into the SITE$CM literal symbol file and 
away you go.  Make sure that this bug is reported though, just to get it onto 
the bug list.

What editor are you using when attempting to resolve merge conflicts?  Where 
are the elements to be merged coming from?  

Tony
1133.3COMICS::TALBOTTrevor TalbotTue Jul 28 1992 18:1020
    Thanks for the quick responses,
    
    
    	Re: .1
    	Kevin,
    		Yes, you assumed correctly the elements being merged are
    type of FRM. Yes again, there was an output file breakdown.flg_merged
    created in the oa$site_dev_area directory. No, if edited directly using
    the EDT editor, no reverse video guff but still lots of unwanted chars.
    
    	Re: .2
    	Tony,
    		Initialy WPS-PLUS was my default editor, hence the reverse
    stuff etc, I switched to EDT and the same results as if edited
    directly. The elements are coming from some application areas that I
    set up for testing, the elements are both .FRM, both named the same the
    only diffs are some screen text added to force a conflict and again
    some named data designed also to conflict.
    
    	
1133.4Here's the breakdown.flg_merge fileCOMICS::TALBOTTrevor TalbotTue Jul 28 1992 18:151
    
1133.52nd try at getting file in!!COMICS::TALBOTTrevor TalbotTue Jul 28 1992 18:1991
    Here is the file:
    
**************** Conflict 1     ************************************************
 
 
!          FMS Form Description Application Aid 
!                     Version V2.4
 
FORM NAME='BREAKDOWN'
    AREA_TO_CLEAR=1:23
    WIDTH=80
    BACKGROUND=CURRENT
    ;


TEXT (1,1) 'breakdown'
    ;
TEXT (2,1) 'Created in ALL-IN-1 V3.0'
    ;
TEXT (3,1) 'TEST2'
    ;
TEXT (5,17) 'Invoice_no:'
    ;
TEXT (7,17) 'other:'
    ;

ATTRIBUTE_DEFAULTS FIELD
    CLEAR_CHARACTER=' '
    NOAUTOTAB BLANK_FILL NOBLINKING NOBOLD NOREVERSE
    NOUNDERLINE NODISPLAY_ONLY ECHO NOFIXED_DECIMAL
    LEFT_JUSTIFIED NOSUPERVISOR_ONLY NOSUPPRESS NOUPPERCASE
    ;

FIELD NAME='INVOICE_NO'  (5,28)
    PICTURE=6'X'
    ;
FIELD NAME='OTHER'  (7,23)
    PICTURE=30'X'
    ;

ORDER BEGIN_WITH = 1
    NAME='INVOICE_NO' 
    NAME='OTHER' 
    ;

NAMED_DATA INDEX=1 NAME='.TYPE'
    DATA='ENTRY /MODE=update/HARD=''GGGG''' ;
NAMED_DATA INDEX=2 NAME='.FILE'
    DATA='breakdown,TEST2.FDL' ;
NAMED_DATA INDEX=3 NAME='INVOICE_NO'
    DATA='/rse_valid = inv_ent.invoice_no with .cost_code = "YYYYYY"' ;
NAMED_DATA INDEX=4 NAME='.COMMENT'
    DATA='This was added to show up some diffs between CMS copy and CM copy.' ;
 
END_OF_FORM NAME='BREAKDOWN' ;
********************************************************************************
6
FMSFRMFED20028-JUL-1992 09:47:0328-JUL-1992 09:47:03�


66�
&
20$P4	BREAKDOWN

  
INVOICE_NO

  OTHER
*



,
.TYPEENTRY /MODE=update/HARD='GGGG'
"	
.FILEbreakdown,TEST2.FDL
	N


INVOICE_NO:/rse_valid = inv_ent.invoice_no with .cost_code = "YYYYYY"

T
.COMMENTBThis was added to show up some diffs between CMS copy and CM copy.

	breakdown

Created in ALL-IN-1 V3.0



TEST2

Invoice_no:

other:

************** End of Conflict 1     *******************************************
    
1133.6DIfference between .FRMs included in .FLG47215::EIJSAll in 1 PieceWed Jul 29 1992 13:2937
Hi Trevor,

The result you see it the merge of 2 FRMs into the FLG. Let me explain.

The original file is the .FLG. You then select the 2 elements of which you 
want the differences included into your .FLG. The difference should run on the 
2 .FLGs. However (yes, you found another one), when CM gets copies of the 2 
elements, instead of obtaining the .FLGs, it obtains the .FRMs. Differences 
are run between the 2 FRMs (which might be different), but obviously the result 
is always different from the contents of the .FLG. So the result is one big 
conflict, and the funny characters are binary (FRM) characters.

Fix:

CM_PRE_MERGE_FRM.SCP

.
.LABEL NOT_SDC
.
        GET #CM_GET_KEY = #CM_SEL_KEY\-
        GET #CM_SRC_LOC = "DEV"\-
-->     GET #CM_FORM_TYPE = "FLG"\-
        GET OA$FUNCTION = CM$ETYPES:NUMBER.GET_SOURCE[CM$SITELOG.NUMBER[#CM_SEL_KEY]]
.
        GET #CM_GET_KEY = #CM_SEL_KEY\-
         GET #CM_SRC_LOC = "DEV"\-
-->      GET #CM_FORM_TYPE = "FLG"\-
         GET OA$FUNCTION =CM$ETYPES:NUMBER.GET_SOURCE[CM$SITELOG.NUMBER[#CM_SEL_KEY]]

The lines starting with '-->' are missing. (CM_GET_FRM.SCP actually nullifies 
the symbol after procssing).

Sorry,

	Simon
                                            
1133.7PAULUS::BAUERRichard - ISE L10N Center FrankfurtWed Jul 29 1992 13:3335
Hi Trevor !

Merging an FRM will not give satisfactory results. This has primarily to do
with the structure of the FLG file.

It is more appropriate to handle forms as screen and named data seperately.

I'm convinced that handling the screen part manually will be quicker in 99% of
the cases. Hence, displaying the old form, editing the new form and applying
the changes.

However, the named data could be merged automaticaly. But again the FLG format
isn't suitable. It list each line of named data with an INDEX number. Any
additional line causes that the following lines can't be matched (and therefore
can't be merged) by CMS because the INDE number is incremented.

There is a solution to it. Using the FDNDEDT1.EXE and FDNDEDT2.EXE one would be
able to get the named data in the style GOLD N shows it (there are propably
other ways in V3 too). Now this file can be merged with good results !!!!

In order to implement it the following script need to be modified

OA$LIB:CM_MERGE_CMS_FRM.SCP

- splitting the form into screen and named data part
- calling the FMS editor for the screen part only
- calling CMS with the named data portion only
- resolving conflicts 
- combining the result again with the associated screen part

This is how we handle forms during the translation effort.

	hope this helps

		Richard