Names not being imported

[Guest Michael Dietz]

I am a complete loss. I am trying to import a relatively large GEDCOM file, 104,000 indiciduals. I have tried all sorts of combinations in the options for the import. I just completed another attempt with the following options checked: Assume marriage of parents, Convert widowed SOUR, Read NPFX/SURN/GIVN and accept Tags as given. Regardless of how I configure the options (I have tried about two dozen times) I get the same result. I even went through the .GED file and corrected as many of the warnings as I could. No avail. Always the same answer. I have been trying to import this file for over two months now.

 

The database is constructed and immediately has 103,984 individuals, all with the surname of UNKNOWN and no given name. The .ged file does have GIVN and SURN data fields. Parental links are built and marriage links are built. Witness links are built. Everything seems okay except there are no names.

 

What is very interesting is all the individuals have a Name-Var event in the Tag Entry list with a memo of the form: Maragret /Harcourt/ Unknown GEDCOM info: Maragret Unknown GEDCOM info: Harcourt. What is even more interesting is this tag will be marked with an asterisk signifying it is the primary name. My understanding is you could only have one primary name but it seems as if this file has two, the UNKNOWN name and the Name-Var.

 

Any suggestions are greatly appreciated.

 

Thank you

Mike

[Michael Hannah 0]

Hi Mike,

 

Since you indicate you are comfortable with editing the GEDCOM file, could you open the file in a text editor and copy/paste one of the people's lines of data in that GEDCOM? I would need to see everything for that individual from their "0 @num@ INDI" line, through and including the "1 NAME xxx" line plus all lines after the NAME line that begin with a number greater than 1, up to the next line that begins with either a 1 or a 0. Something about the combination of your NAME entries and the import options you are choosing does not seem to make sense.

[Guest Michael Dietz]

Hi Mike,

 

Since you indicate you are comfortable with editing the GEDCOM file, could you open the file in a text editor and copy/paste one of the people's lines of data in that GEDCOM? I would need to see everything for that individual from their "0 @num@ INDI" line, through and including the "1 NAME xxx" line plus all lines after the NAME line that begin with a number greater than 1, up to the next line that begins with either a 1 or a 0. Something about the combination of your NAME entries and the import options you are choosing does not seem to make sense.

 

Michael:

Thank you for the reply.

 

Here is the data from the ged file I used in the example above.

0 @I100763@ INDI

1 NAME Margaret /Harcourt/

2 GIVN Margaret

2 SURN Harcourt

1 SEX F

1 BIRT

2 DATE Abt 1581

2 SOUR @S1833@

1 DEAT Y

1 EVEN

2 TYPE Record Change

2 DATE 26 Jun 2003

1 _UID 745C4FA2B2AD463FAB8E83C2A880D56C8C36

1 CHAN

2 DATE 27 Oct 2008

3 TIME 14:14

1 FAMC @F40447@

 

 

One additional bit of information. Clicking on the Tag Entry for the Name-Var event does not give a Tag Entry Name display screen, i.e., with the presurname field, suffix field, etc. Rather it gives a normal tag entry event with the location fields, etc. What is doubly strange in clicking on the tag type button takes me to the Name-Var tag in the Tag TYpe List. If I replace the tag with the Name-Var tag (even though they are the same) it still show the tag entry event with the location fields. So it appears as if the name tags are treated as historic events, similar to birth and death. The UNKNOWN name tag also goes to the Name-Var tag in the Tag Type list but it returns to the Name tag data, i.e., presurname, suffix, etc.

 

Thank you again.

Mike

[Jim Byram 0]

You have tried the import using the default settings? You probably want to tick assume marriage of parents and create custom tag types for the non-standard and event misc GEDCOM events.

[Guest Michael Dietz]

You have tried the import using the default settings? You probably want to tick assume marriage of parents and create custom tag types for the non-standard and event misc GEDCOM events.

 

Thank you Jim:

I have vasically tried both suggestions. I have even done the Unassign All followed by Assign All. I have not assigned a custom tag to all the event misc GEDCOM events since there are about 200 of them Howeve I have unassigned them all. There is one event misc tag named Name which I have assigned to the Nave-Var tag. All to no avail.

 

Thank you again.

Mike

[Jim Byram 0]

How big is a ZIP of the GEDCOM?

[Guest Michael Dietz]

How big is a ZIP of the GEDCOM?

 

Jim:

 

11,775 kb

 

I am going to try another tactic today. There are 571 lines which have a code of _NAME. I think maybe because these are treated as event misc lines it triggers the GenBridge program to set them as event events and not name events and this triggering carries over into the normal NAME lines also. I am going to change the _NAME events to NAME events and reimport the file. However it takes over 10 hours to go through an import so I won't know until this evening if it worked. I will let you know as soon as it completes what happens.

 

Mike

[Jim Byram 0]

You haven't shown how the _NAME tags are used in the GEDCOM. I agree that assigning them to Name-Var could break things.

 

Another approach would be to import the GEDCOM into PAF to try to clean it up and then export a GEDCOM 5.5 ANSI file from PAF and import that to TMG.

[Guest Michael Dietz]

You haven't shown how the _NAME tags are used in the GEDCOM. I agree that assigning them to Name-Var could break things.

 

Another approach would be to import the GEDCOM into PAF to try to clean it up and then export a GEDCOM 5.5 ANSI file from PAF and import that to TMG.

 

Here is an example of the _NAME tag (or is this just To Be or Not To Be):

0 @I17391@ INDI

1 NAME Lulach "The Fool" King Of /Scotland/

2 GIVN Lulach "The Fool" King Of

2 SURN Scotland

1 NAME The Fool //

2 GIVN The Fool

2 SURN

2 NICK The Fool

1 SEX M

1 BIRT

2 DATE Abt 1032

2 PLAC Morayshire, Scotland

1 DEAT

2 DATE 17 Mar 1057-1058

2 PLAC Ambush, Burned Alive By Malcolm Iiiat Essie, Strathbogie

2 SOUR @S501@

3 PAGE Macbeth

3 QUAY 3

1 BURI

2 DATE Unknown

2 PLAC Isle Of Iona, Scotland

1 ADDR

2 _NAME Lulach "The Fool" King Of Scotland

1 EVEN

2 TYPE Info 2

2 PLAC Succeeded His Step-Father Macbeth, Reigned 15 Aug 1057-17 March 1058. Great Grandson Of Kenneth Iii. Succeded By Malcolm III

1 EVEN

2 TYPE Ruled

2 PLAC 1057-1058

1 _UID B271C874C9C74018AA36C80F22C035CFE4AA

1 CHAN

2 DATE 20 Aug 2008

3 TIME 20:25

1 FAMS @F3657@

1 FAMC @F9956@

1 NOTE @NI17391@

 

I did further searching and found EVEN NAME tags. My re-import of the file has been going for three hours now so I hate to stop it to fix that tag. Here is an example of it (quite long):

0 @I640679004@ INDI

1 NAME John /Sherman/

2 GIVN John

2 SURN Sherman

1 SEX M

1 BIRT

2 DATE 1396

2 PLAC Yaxley, Suffolk, England

2 SOUR @S992@

2 SOUR @S993@

2 SOUR @S995@

3 PAGE Compact Disk #4 ID#994008

3 QUAY 3

1 CHR

2 DATE 1401

2 PLAC Yaxley, Suffolk, England

2 SOUR @S995@

3 PAGE Compact Disk #4 ID#994008

3 QUAY 3

2 SOUR @S993@

1 DEAT

2 DATE 1487

2 PLAC Yaxley, Suffolk, England

2 SOUR @S992@

2 SOUR @S993@

2 SOUR @S995@

3 PAGE Compact Disk #4 ID#994008

3 QUAY 3

1 SOUR @S1037@

2 PAGE Pedigree Records - Chart: 301-5

2 QUAY 3

1 SOUR @S1058@

1 SOUR @S1066@

2 PAGE p. 19

2 QUAY 3

1 SOUR @S1065@

2 PAGE p. 680

2 QUAY 3

1 EVEN Name

2 TYPE Unknown

2 SOUR @S1037@

3 PAGE Pedigree Records - Chart: 301-4

3 QUAY 3

2 SOUR @S1066@

3 PAGE p. 19

3 QUAY 3

2 SOUR @S993@

2 SOUR @S995@

3 PAGE Compact Disk #4 ID#994008

3 QUAY 3

1 _UID 88EDCFEBBD67D911AF460010B59547512424

1 BAPL

2 DATE 12 Nov 1938

1 ENDL

2 DATE 1 Nov 1940

1 SLGC

2 DATE Done

1 CHAN

2 DATE 4 Sep 2008

3 TIME 08:38

1 FAMS @F554353395@

1 FAMC @F25003@

1 NOTE @HI640679004@

 

I will try the import through PAF and see how that works.

 

Thank you Jim for the help

Mike

[Jim Byram 0]

1 ADDR

2 _NAME Lulach "The Fool" King Of Scotland

 

This is a level 2 tag. I don't know how it would be handled but the data is subordinate to the ADDR tag. You don't want to change this to NAME. And the structure makes no sense. This reflects something about the data structure in the exporting program.

 

1 EVEN Name

2 TYPE Unknown

 

This makes no sense. It's an event tag but is improperly structured. And I have no idea how you'd 'fix' it.

[Guest Michael Dietz]

Jim:

Here is the header from the .ged file:

 

0 HEAD

1 SOUR Legacy

2 VERS 7.0

2 NAME Legacy ®

2 CORP Millennia Corp.

3 ADDR PO Box 9410

4 CONT Surprise, AZ 85374

1 DEST Legacy

1 DATE 9 Nov 2008

1 SUBM @S0@

1 FILE C:\Documents and Settings\Owner.YOUR-XHTR8HVC4P\My Documents\emsuggs.ged

1 GEDC

2 VERS 5.5

2 FORM LINEAGE-LINKED

1 CHAR ANSI

0 @S0@ SUBM

1 NAME Not Given

 

I imported the file into PAF v5 and received the following statistics

Created by: Legacy Legacy ® 7.0

Submitted by: Not Given

Individuals: 103983

Marriages: 52586

Sources: 1836

Repositories: 535

Errors: 127908

 

The most frequent error was trying to import two names for the same individual. I looked in the .ged file and what it seems to be are name variations, e.g., Smith and Smythe

 

In any case I am now importing the paf file into TMG. I will let you know what happens.

 

Thank you again.

 

Mike

[Jim Byram 0]

OK. Legacy creates many structures in the GEDCOM that are meaningful only to Legacy and are intended for reimporting the GEDCOM into Legacy.

 

1 ADDR

2 _NAME Lulach "The Fool" King Of Scotland

 

This is for the Name field on the Legacy address entry screen and is only of use to Legacy.

 

Checking the 'name' 1 EVEN 2 TYPE event. I have no idea how this was done. Legacy exports alternate names correctly as 1 NAME tags. Maybe the user created a custom event named 'Name'.

 

Legacy has various 'Produce file for:' GEDCOM export options. I'm guessing that the GEDCOM was created using the default 'Legacy' option which creates a file intended to be re-imported into Legacy. The GEDCOM should have been created using the 'GEDCOM 5.5 Only' option for sending to people not using Legacy.

 

If that's the case, you're fighting a losing battle.

 

btw...

1 EVEN Name

2 TYPE Unknown

 

This is trying to say that the name was unknown but doesn't make structural sense since 'Name' is not an event. It is meaningful only when imported into Legacy.

 

This whole discussion is a perfect example of the hard lesson that programs can create GEDCOM files containing structural elements that other programs can't deal with when importing. The structures can be perfectly legal in terms of the GEDCOM specs but are only meaningful when the GEDCOM is re-imported by the program that created it.

Edited June 27, 2009 by Jim Byram

[Guest Michael Dietz]

If that's the case, you're fighting a losing battle.

 

btw...

1 EVEN Name

2 TYPE Unknown

 

This is trying to say that the name was unknown but doesn't make structural sense since 'Name' is not an event. It is meaningful only when imported into Legacy.

 

This whole discussion is a perfect example of the hard lesson that programs can create GEDCOM files containing structural elements that other programs can't deal with when importing. The structures can be perfectly legal in terms of the GEDCOM specs but are only meaningful when the GEDCOM is re-imported by the program that created it.

 

Jim:

Thank you for all the time you have spent on this.

 

I think I am going to resolve myself to accept what gets imported from PAF and go with that. Right now it is 89% done with importing other information. I do have the error list from PAF and the one from trying to import directly from the ged into TMG. I can use those for other info which failed to import.

 

I can only second your last comment, don't expect wonders from GEDCOM, either to or from.

 

Again thank you and we will see you the end of August in Virginia.

 

Mike

[Michael Hannah 0]

Hi Mike,

 

The good news

I created a small GEDCOM file from your examples and imported it using the TMG Advanced import wizard. The structure for

1 NAME Margaret /Harcourt/

2 GIVN Margaret

2 SURN Harcourt

imports using the option you specified of "Read NPFX/SURN/GIVN" as a single primary Name tag with the appropriate Given and Surname. No problems that I can see.

 

I also tried

1 NAME Lulach "The Fool" King Of /Scotland/

2 GIVN Lulach "The Fool" King Of

2 SURN Scotland

1 NAME The Fool //

2 GIVN The Fool

2 SURN

2 NICK The Fool

and that gave the appropriate primary Name tag with the appropriate Given and Surname. In addition it produced an additional Name-Var tag with Given of "The Fool" and also created a Name-Nick tag with Given of "The Fool". This is just what I would have expected of these GEDCOM constructs, and see no problems. I get no surnames "Unknown" and no other problems.

 

The bad news

As far as the constructs of:

1 ADDR

2 _NAME Lulach "The Fool" King Of Scotland

 

and

 

1 EVEN Name

2 TYPE Unknown

 

The advanced import wizard defaulted to suggest the GEDCOM ADDR tag be imported into an Address tag, and the type "Unknown" GEDCOM tag be imported into an Event-Misc tag. For my first test I accepted these defaults.

 

Using default tag assignments

The resultant Address tag had no location or any other kind of data, but filled the memo with the text:

Unknown GEDCOM info: Lulach 'The Fool' King Of Scotland

This is what I would have expected, since the value of "_NAME" following the digit 2 is not a standard keyword for GEDCOM version 5.5. The only documented keywords subordinate to an ADDR tag other than continuation lines are:

ADR1, ADR2, CITY, STAE, POST, CTRY

Thus the keyword _NAME is not recognized. However, since this information is subordinate to this ADDR tag I think it is appropriate for TMG to put this into the TMG memo. However, whatever program is producing this construct is not producing recognizable GEDCOM.

 

The resultant Event-Misc tag loses the word "Unknown" since that was considered the name of the GEDCOM tag, and the text "Name" following the keyword EVEN is put in the memo. In my opinion this is also appropriate as the best TMG could do.

 

Forcing assignment to Name-Var tags

As a second test, in the advanced import Wizard I forced both of these TMG constructs to import not as the default but as Name-Var tags.

 

The first construct produced a Name-Var tag (in addition to the Name-Var up in the normal Primary spot) with its own Primary designation!! But as you noted when you open this tag it shows the Address fields, not the Name fields, yet the pull-down list of tag types to which it could be changed are only tags from the Name group!! In my opinion this is a bug in the TMG import. Since there are no subordinate "NPFX/SURN/GIVN" fields, which you claimed by the import option would exist, TMG import should not recognize this construct. It "sort of" created a Name-Var tag yet it "sort of" is not a Name-Var. TMG is really confused at this point with this GEDCOM construct that does not conform to the basic standard. I have reported this to WG as a bug, but since it is very rare and caused by a construct that does not comply with the basic GEDCOM standard I suspect its priority for a fix will be low.

 

TMG import is smarter with the second construct. It identifies the GEDCOM construct is not a recognized standard name structure and the import refused to create a Name-Var tag. Instead it created an Event-Misc tag. However, this time in the memo it puts the text "Name Type: Unknown" which accurately reflects that GEDCOM claimed this was some kind of Name tag with some kind of "type" that was the text "Unknown". I think this is the best TMG could do with this GEDCOM construct.

 

Conclusion

I believe the symptoms you are seeing are caused by two conditions. First, most importantly, is a GEDCOM file containing constructs that do not conform to the basic GEDCOM 5.5 standard and thus are not recognized by TMG. Second is the attempt to force TMG to create a Name-Var tag from a GEDCOM construct that is not a recognizable name structure. It tries, but gets very confused.

 

Hope this helps,

Edited June 27, 2009 by Michael Hannah

[Jim Byram 0]

Mike,

 

You could ask the creator of the GEDCOM to create a file using the appropriate GEDCOM export settings.

 

Alternatively, I'll import the GEDCOM into Legacy and recreate the GEDCOM in the correct format for you. Or send you the Legacy database and you can import that.

 

Jim

[Guest Michael Dietz]

Thank you both, Jim and Michael, for the time you spent on this. The problem stems from the original GEDCOM export and since I have imported it into TMG via PAF I at least have the primary names and dates I was looking for. A partial resolution is better than no resolution.

 

Again thank you very much.

Mike