* "Cannot load file" error

[jsnyder]

I have found a .ged file through GEDCOMindex.com. It is a very large file (319 MB). When trying to open or import the file into FH 6.2.7, it begins to load. After a minute or so of "loading file ... " the screen reads "Initialising data ... ". But after a few seconds the messages "An error has occurred. Cannot load file" and "Unable to import file" pop up. I don't know the age of the file and don't have any way of contacting the person who created it. It has passed all malware tests, so I'm comfortable with its safety. My Windows 7 PC has plenty of RAM (8 of 12 GB available). The file's properties show all appropriate permissions. Other .ged files (not necessarily from the same person) from the web site can be imported without issue.

The file is too large to open in any text editor that I'm familiar with. There are file splitter programs that could break the file into more manageable parts, but trying to add headers and end-of-file statements to each part to make them work is a little beyond my FH experience.

This may be an issue that can't be resolved without knowing more about the file's contents or how it was created, but I thought I'd try. I'm certain the file could be very useful if it can be opened. Are there any suggestions about how to tackle the problem?

Thank you,
Joel

[Valkrider]

Joel

Can you post a link to which file you actually are talking about as there are many hundreds on that site.

Programmes such as Notepad++ can handle files with a large number of rows. Other text editors such as UltraEdit, SublimeEdit say they can handle files of up to 2GB so filesize may not be the issue you may need some more memory.

[jsnyder]

Colin,

Thanks for responding so quickly and thanks for the text editor recommendations. I'm not sure I'll know what to look for if I am able to open it in an editor but I'll certainly give it a try.

In the meantime, the file I am trying to open in FH is for Nathaniel Felch on the page http://gedcomindex.com/Pages/US/USME0389.html.

The file link itself is: http://gblcreations.com/Resources/Genea ... 524482.ged

I would be surprised if memory is an issue as 8-10 GB of free RAM seems like it should be enough and I am able to work with other very large multi-GB files in memory-intensive image editing programs. Although if all else fails I'll look at the option of more memory. Thanks.

[PeterR]

I was able to load the file OK; it took several minutes and generated a massive exception report, but no error messages. FH Focus Window at bottom right has: Individuals: 879234 Families: 305277. My 9-year-old laptop has 6 GB RAM and I recently upgraded (free) from Windows 7 to Windows 10.

[tatewise]

My even older desktop ex-Win 7 now Win 10 with only 3GB RAM took even longer but with similar results.
It is then VERY SLOW to respond.
Although there are a lot of Exception Reports, most are relatively benign.

[jsnyder]

Thanks Peter. At least it's good to know that someone was able to open the file in FH, suggesting that it's not the file itself. I figured it contained hundreds of thousands of individuals and the size of the exception report isn't surprising. I was able to easily open the file in Notepad++, so at least I can see the contents. Now I just need to know what to do with it.

I'm still puzzled why I get the error messages when I just opened a relatively large .ged file (170 MB with 270,000 individuals and 105,000 families). I guess I'll try breaking the text file up into a few smaller files and see if that works. Problem may be in losing individual and family relationships, not to mention trying to figure out what to add as headers and end-of-file statements.

Unfortunately, upgrading to Windows 10 isn't an option since I run a couple other special programs that run only with Windows 7 (very expensive to purchase Win 10 version).

I appreciate your feedback.

[jsnyder]

Mike, I just saw your post. It seems clear that the problem is on my end. With a file size this large, I imagined it would be cumbersome to work with, but at least you were able to open it. I'll keep working on it. Thanks.

[Jane]

I think just 800,000 people I suspect is at the top of what is usable in a single tree on FH, although it loaded fine and works ok on my desktop PC.

I have saved the file off in V6 format in case that helps in making it load on your PC as it won't have to go through the conversion routine.

https://www.dropbox.com/s/w8g4ci9myzufc ... 2.zip?dl=0

From what I could see there are no sources or other supporting evidence in the file.

[tatewise]

This is the largest GEDCOM ever seen by me, and FH is VERY slow ~ taking minutes to perform many functions.
It is not easy to edit a GEDCOM file into a logical subset, because usually there are too many interlinked records.

The File > File Statistics say:
Record Counts...
Individual Records:879234
Family Records: 305277
Note Records: 204500
Source Records: 15302
Repository Records: 223
Submitter Records: 1
Submission Records: 0
Media Records: 752
Place Records: 132111

So contrary to Jane, there seems to be plenty of Note, Source, Repository, and Media records.
The Source records have many Citations and many have a Repository.

[Jane]

Ok, I only checked the first few records in the file.

If you know which branches you are interested you should be able to use the Export Gedcom to split out the information you need. Just be very patient with FH.

[johnmorrisoniom]

There are 1641 pools (Several are only one person) in the file, so splitting by pool could also be done

[tatewise]

Yes, exporting by Pools could work, but jsnyder's problem is he cannot load the GEDCOM.
Editing the GEDCOM file into Pools is not feasible.
So it needs some kind soul whose PC is not too slow to run the FH exports for him (not me as my PC is far too slow).

[jsnyder]

Thanks to everyone. You all have been very helpful. Through Notepad++ (thanks, Colin, for that suggestion) I've been exploring some alternatives by eliminating all non-ASCII characters, encoding various characters sets, encoding the file as ANSI, UTF-8, and others. Nothing seems to work. Jane, I downloaded your Dropbox file but it doesn't work either. However, from looking at my file through Notepad++ I do see sources and other data as Mike mentioned.

Since several of you have opened the file successfully, it seems that the problem remains with my system or FH program. I'm certain that I have ample RAM, so I'll continue to investigate other alternatives. I may consider reinstalling FH but I want to explore what that entails before I go that route, making sure I don't lose settings or anything else I've created. I have to be away and won't have time to pursue this for a day or so. I'll post the results of reinstalling FH when I've completed that task. Again, thanks to all for all your suggestions.

[Ron Melby]

https://www.editpadlite.com/

this will translate into utf-8 for you.

given the hog slaughterer file you have, it might take a loooooooooooooong time. but thats one thing done.

[Valkrider]

How do you want the file @Joel?

It only took about 1 minute to load on my iMac into Family Historian.

[tatewise]

Colin, as was said earlier, the Individuals appear to be in many unrelated Pools.
So by using File > Export > GEDCOM File and selecting say the largest Pool 1 that will create a smaller GEDCOM file.
Repeat that process with each single or cluster of Pool numbers to yield a number of more manageable GEDCOM files.
Then it is a matter of transferring those files to Joel, perhaps via Dropbox.

That process will retain all Family relationships, and links to Notes, Sources, Media, etc.

[jsnyder]

Thanks, Ron. With Notepad++ I was able to encode the file in various ways, including UTF-8. It still caused problems so my guess is that it's something else that's the issue.

How do you want the file @Joel?

It only took about 1 minute to load on my iMac into Family Historian.

Colin, I'm not quite sure what you mean, but I'm running Windows 7 and if you're able to use Dropbox as Jane did in her post, then that would work perfectly. If you are talking about something else, I apologize.

....
I have saved the file off in V6 format in case that helps in making it load on your PC as it won't have to go through the conversion routine.

https://www.dropbox.com/s/w8g4ci9myzufc ... 2.zip?dl=0
....

[johnmorrisoniom]

I got an error when I tried to start a new project using the file, but I could open it through gedcom tasks as a stand-alone gedcom.

[wianb]

You could try GENViewer 2 from http://www.mudcreeksoftware.com/ It has a free trial period.

[tatewise]

I have had a 'trial' version of GENViewer 2 for ages and it has never expired.
It reads that enormous GEDCOM file in 12 seconds and allows records to be searched & viewed quickly.

It has an Export Islands option (like FH Pools) but the exported GEDCOM file format is extremely corrupt & unusable.

[jsnyder]

Thanks, everyone, for your responses. I was unexpectedly called away for several days but I am back at this issue. Regarding looking at the GEDCOM file as a text file, I was easily able to open and edit it in Notepad++

I tried opening the original large file in my Surface Pro tablet running Windows 10, which has about 5 GB of free RAM out of 8 GB total. It took less than a minute to load and I could see all 879k individuals. I selected Nathaniel Felch (b. 1699) and easily created an Ancestors & Descendants tree. I then tried to create an All Relatives tree for him but FH locked up after a few minutes. Next, I ran the All Relatives query, but FH also locked up. Lastly, I tried running the All Pools Relatives plugin and got the same FH lock-up.

So, the bottom line appears to be that the file is too large to use efficiently and that there is probably a practical limit on file size/number of individuals that FH operating in Windows 7 can handle. Consequently, I'll use my Surface Pro running Windows 10 to select an individual or family to work with along with a limited number of their relatives, then try to export them to a file that can ultimately be brought into my tree that's loaded in FH on my Windows 7.

Again, thank you all for your contributions.

[tatewise]

It may be worth reporting your problems and solutions for Calico Pie to investigate.
See how_to:about#problem_reporting|> Problem Reporting for how to submit report.
Just point them at the GEDCOM download page so they can get a copy.

[jsnyder]

I'll do that and report back if I learn anything.
Thanks

[jsnyder]

Calico Pie's response was as expected:

"The file is exceptionally large, but as you say does load in Windows 10, I have not yet been able to get Family Historian to lock up, but it does go non responsive when trying to compute relationship, however my tests suggest it's still working on the results and if left long enough might well come back.

The diagram may be a step too far though. We have not tested with files much larger than 250,000 as a working tree."

I think I'm relegated to working with the file on my Surface Pro in Windows 10 on a limited basis, selecting a few individuals at a time, then exporting reasonably sized families to FH on my PC running Windows 7. At least I should be able to cobble together a few pieces of this very large pie.

Thank you, again, for your feedback.