Could you provide a test file that we could use to check what is going
wrong? I appreciate that you described the way your file was
constructed, but it did not have any of the linked data in it.
It would probably be best to send it direct to me (not to the list, as
I appreciate that you would not want the data too widely distributed),
because I was working on the GEDCOM import and correction of missing
bits of the GEDCOM data.
Tim (aka guy).
On 23 Aug 2012, at 16:43, <foberle@...> wrote:
> This is to follow up on my Gedcom import issue.
> I downloaded and installed Gramps 3.4.0-1 (didn't go well at first
> as I was unaware that the old version needed to be removed first -
> worked fine after I totally removed the original install).
> This time I had no repetition of the first person as both ID0 and
> ID1, so I guess that was solved. The import error log was very, very
> long and, among other things, contained many "tag recognized but not
> supported" messages - towards the end, each and every one of the
> family records was marked as ignored because they don't refer back
> to the family (I can't figure out what that means).
> Most importantly, each and every person continues to show up as a
> completely disconnected individual in the ancestors and descendants
> views (I can't recall the terms Gramps uses for those). Strangely
> enough, however, when I pick a person and choose Edit Person|
> Reference, the entire family is shown correctly, so Gramps obviously
> figured out the contents of the ged file. I'm lost.
> Although I greatly appreciate the feedback, I really don't have the
> patience to deal with this sort of thing any more, so I uninstalled
> the program and will keep looking for another Linux-based solution.
> So - no need to respond; I'll check back periodically as new
> versions appear, but for now I'm officially giving up.
> Thanks again,
> Frank Oberle
> ---- Original message ----
>> Date: Thu, 23 Aug 2012 09:59:22 +0200
>> From: Benny Malengier <benny.malengier@...>
>> Subject: Re: [Gramps-bugs] Reading GED(com) Files
>> To: foberle@...
>> Cc: guy.linton@...
>> Best keep posting to user list also, some gedcom
>> experts there that can perhaps comment on your
>> If you import a gedcom, and there are messages
>> during the import, you will see in the bottom left
>> corner of Gramps a small red icon appear. If you
>> click it, you see all the warning messages Gramps
>> generated during import.
>> 2012/8/22 <foberle@...>
>> Benny and Tim (Guy?)
>> Thanks for the responses. I'll try to download an
>> update and rebuild the database, after which I'll
>> report back (although this may take a few days as
>> life is interfering again). Here is the original
>> response I sent to Peter which should answer your
>> questions about where the ged file originated:
>> Hi Peter:
>> I guess you've caught me red-handed... but thanks
>> for responding.
>> I keep all of my raw data externally, since I like
>> to edit it in what is, at least to me, a much more
>> straightforward fashion than using a GUI (I'm an
>> old, but efficient old fogey).
>> I then use a rather large set of what are
>> essentially GREP macros to generate new
>> replacement sections for the GEDCOM file. There is
>> a header, of course, followed by the INDI section.
>> After this comes a section containing notes,
>> followed by the family definition section.
>> Individual records look like this, although
>> obviously the level of detail varies widely:
>> 0 @I2667@ INDI
>> 1 IDNO 2667
>> 1 NAME Kiziah /Gonce/
>> 2 SURN Gonce
>> 2 GIVN Kiziah
>> 2 _MARNM Rogers
>> 2 SOUR @S64@
>> 3 PAGE 34
>> 3 QUAY 2
>> 1 SEX F
>> 1 BIRT
>> 2 DATE 1822
>> 2 PLAC Hawkins County, TN
>> 1 DEAT
>> 2 DATE 1848
>> 1 REFN 2667
>> Family records look like the following:
>> 0 @F5@ FAM
>> 1 HUSB @I3@
>> 1 WIFE @I129@
>> 1 MARR
>> 2 DATE 28 JUN 1943
>> 2 PLAC Blessed Sacrament Church; Baltimore MD
>> 1 CHIL @I133@
>> 2 _FREL Natural
>> 2 _MREL Natural
>> 1 CHIL @I136@
>> 2 _FREL Natural
>> 2 _MREL Natural
>> 1 CHIL @I138@
>> 2 _FREL Natural
>> 2 _MREL Natural
>> In this example, for instance, I suspect that the
>> fact that there are no family lines within the
>> INDI segments for IDs 3, 129, 133, 136 and 138
>> must indicate that Gramps wanted those entries to
>> be present. What confuses me is that the error
>> output clearly showed that Gramps knew that 133,
>> 136 and 138 were children of 3 and 129, because it
>> said (as an example) that's what it was removing
>> and even gave the names of the individuals
>> Lastly, there are the two sections for sources and
>> repositories that look like the following:
>> 0 @S172@ SOUR
>> 1 TITL Pennsylvania Marriages Prior to 1790
>> 1 AUTH John B. Linn and William H. Egle, Editors
>> 1 PUBL 1984; Genealogical Publishing Co., Inc.
>> Baltimore, MD.
>> 1 REPO @REPO4@
>> 2 CALN ISBN: 0-8063-0709-9; LoC 75-37471
>> 3 MEDI Book
>> 0 @REPO12@ REPO
>> 1 NAME Harford Memorial Gardens
>> 1 ADDR 300 Moore's Mill Rd
>> 2 CONT BEL AIR MD 21014
>> 1 PHON 410-838-6667
>> I believe these are all generated to conform to
>> the specs (as I understand them; those specs
>> aren't always as clear as I would like). The
>> automated indentation also helps me read the
>> gedcom file more easily (again, being a Luddite
>> and command line user, I tend to do lookups on the
>> ged file as it is just faster and more convenient
>> - my editor lets me use GREP and highlights the
>> appropriate lines and so forth). I have a certain
>> level of confidence in the assertion that the
>> formatting is at least acceptable, if not correct,
>> since the generated files load in the various
>> programs I use (Why so many? Simply because
>> different ones have different strengths and levels
>> of flexibility in presentation and
>> error/discrepancy checking.)
>> I know you were looking for something like
>> "Program X," but hope this helps answer your
>> Any comments would certainly be appreciated.
>> Frank Oberle
>> ---- Original message ----
>>> Date: Wed, 22 Aug 2012 14:55:50 +0200
>>> From: Benny Malengier <benny.malengier@...>
>>> Subject: Re: [Gramps-bugs] Reading GED(com) Files
>>> To: foberle@...
>>> Cc: gramps-bugs@...
>>> Is it not possible you imported the gedcom
>>> second time in the family tree you did the
>>> attempt for?
>>> As the first import crashed on the empty
>> lines, the
>>> families of the people would never have been
>>> and everybody would have incomplete families.
>>> Your second import would duplicate people.
>>> So, if this was the case, delete the family
>>> make a new one, and do the import in this
>>> 2012/8/22 Tim Lyons <guy.linton@...>
>>> We discussed what is probably the same
>>> The GEDCOM import has been improved in
>> 3.4.0, and
>>> it is possible that the
>>> problem you encountered has been fixed.
>>> I know it is annoying to be told to try
>> the newer
>>> version, because the
>>> problem _may_ be fixed there, but in this
>> case, I
>>> think it would be
>>> worthwhile. Anyway, if the problem is ever
>>> to be fixed, it will not
>>> now be fixed in 3.3.x.
>>> I am not sure why you found a duplication
>> of the
>>> first individual, but the
>>> allocation of ID Numbers has also been
>>> so this problem may also
>>> have gone away.
>>> I too would be interested to know which
>> tool was
>>> used to export the GEDCOM.
>>> View this message in context:
>>> Sent from the gramps-bugs mailing list
>> archive at
>>> Live Security Virtual Conference
>>> Exclusive live event will cover all the
>>> today's security and
>>> threat landscape has changed and how IT
>>> can respond. Discussions
>>> will include endpoint security, mobile
>>> and the latest in malware
>>> Gramps-bugs mailing list