* Seaching Multiple Projects

[koornalla]

Hi all,

I am currently assessing the suitability of FH with data exported from TMG. A really useful feature in TMG is "Data Sets. I have allocated a Data Set to each of my unconnected "THURLEY" trees (twigs). All THURLEY information is in the one TMG THURLEY Project but each separate twig is stored in it's own Data Set. This allows me the separation to easily work on an individual THURLEY trees or to search all THURLEY trees when new THURLEY information comes in.

I am happy to have all my unconnected THURLEY twigs in separate FH projects but, if that is my only option, then I need the ability to search for a THURLEY related individual across multiple FH THURLEY projects.

I am sure I am not the only person with this particular requirement.

How would I best achieve this in FH?


Wayne Thurley

[Jane]

Unfortunately can't search across multiple projects currently, but you could simply merge all your datasets into one Project, but add a "branch" attribute or flag to each twig so you can easily filter or split off the twigs when sharing them with other people.

It's worth remembering as well that FH can have multiple instances open so you could do the same search on each twig (although this is obviously not as efficient.

[jimlad68]

I remember the debate in TMG, separate projects or separate datasets within 1 project, the latter, as you are using, being the preferred method. However I found that it required a lot of maintenance for those individuals that appeared in multiple datasets (as I suspect you will if you use separate FH projects), so I kept them altogether in one dataset in one project and just used alternate datasets for splitting, merging etc. I don't know how big your DB is, but you will find FH much faster than TMG, although a large DB does not always equate to a much slower speed.

Otherwise Jane's comments seem the best option with the hope that a future release may include it. Another one for the wish list as it would be a nice feature, not sure how much effort would be required from the developers; or possibly a Plugin.

[mjashby]

Wayne,

I believe that you have hit upon one of the truly unique facilities of TMG that no other family history software has yet been able to replicate (unless anyone knows different), i.e. the apparent ability to wrap individual/distinct family data files within a single database/project structure. Presumably one of the reasons TMG was/is extremely popular within the One Name Study population.

The fact that FH Project data is contained within entirely separate/discrete gedcom text/datafiles means that such a facility could not be directly implemented unless FH could be programmed to load and switch between multiple live Projects. Don't know enough about Lua Plugins, but I don't think a search facility could be written that could be made to include searches within unconnected gedcom files, i.e. it doesn't seem logical.

The nearest you could get is probably some form of third-party text searching software that could be directed towards examining the contents of multiple gedcom files for matching information - There is a plethora of free utilities and one of them may be able to work in this way. In theory you should also be able to use Windows 'Advanced Query Syntax' but.....

Mervyn

[tatewise]

If I understand correctly, then each Data Set twig is a sub-tree with no relationship link to any other sub-tree.
In other words each twig is an isolated family tree disconnected from the rest of the database.

In FH this is called a Relation Pool and each Pool is automatically assigned a different number.
This can be demonstrated using the oddly named View > Standard Queries > Search for Orphans that lists Individuals according to their Pool number.
This uses the =RelationPool(%INDI%) function that returns the Pool number of any Individual.
So it is perfectly feasible to identify one Pool/Data Set twig of Individuals and create queries/searches on any one or combination of Pools.

In addition to all that, it is possible to put all Individuals of any one Data Set into an FH Named List, which effectively forms a Data Set on which queries/searches/diagrams/reports/exports can be performed.

It sounds like you have a One Name Study and other FH users will no doubt give you advice about how they manage a single FH Project ONS database for all their sub-trees.

[AnneEast]

I don't pretend to any advanced knowledge of the deeper workings of FH. However I do run two separate one name studies and am very happy with them. They are completely different names so I have one Project for each. Within the 'Name' project I don't have any trouble selecting a family group. I just pick a person and generate a tree of whichever sort is appropriate for the job.

This is one of the beauties of FH ....... the ability to work from a tree diagram!

Anne

[tatewise]

Having looked at the TMG Data Set documentation, one of its features is the ability to compare Records in different Data Sets.
So any database/Gedcom can be imported into a new Data Set and compared/merged with another Data Set.

This is somewhat similar to the FH Merge/Compare File feature, which allows one Project to be merged/compared with another Project and interactively match compatible Records.
So any database/Gedcom can be imported into a new Project and compared/merged usually with a copy of your master Project.
Once complete, the merged Project can become the new master composite Project.

[koornalla]

Hi all,

Thanks for the prompt and very informative response to my question. I guess I am using TMG Data Sets in a one name study context. I have had in excess of 30 THURLEY Data Sets at points in time but are now down to about 25. There is never common people across Data Sets because, as soon as I find a common individual, I merge the two Data Sets. In a perfect world I would like to get down to one big Data Set.

With the current TMG setup, other than being able search across multiple Data Sets, I am able to produce reports/web pages on individual twigs as well as export a Gedcom file for an individual twig and send it to an interesting party (or upload to Ancestry.com). I use Second Site to produce a web site and this works well with TMG Data Sets. Unfortunately, with the demise of TMG, the future of Second Site is uncertain. I am looking for a web solution and that will be another query I have to this Forum.

So, because I what I have indicated above, I am not in the situation of the person who raised the complexity issue with multiple TMG Data Sets.

The Attributes/Flags suggestion has merit but raises the issue of always having to remember set a flag/attribute when adding a new individual. One lapse of memory and the new person added is (under certain circumstances) lost. I just think this idea raises a risk issue that I am not willing to take. If a clever script can be employed that reminds you to add a flag when adding a new individual then I might reconsider this option.

Mike Tate's suggestions of Relation Pools seems to have merit as long as I can easily report on and export individual Pools.

If combining all my twigs in the one Project becomes too problematic then I will have a separate project for each and then work out a clever way to create a Master List. An Excel spreadsheet regularly updated with data from all related projects might be as good enough (if exporting to csv is possible?).

I do hope others with similar requirements have solutions they can offer me.




Wayne Thurley

[Valkrider]

Given your last update I would merge all your projects now into one FH database.

FH allows you to have unlinked individuals and later link them when more data becomes available.

I use FH for my one-name study in exactly this way.

[tatewise]

Wayne asks:

Relation Pools seem to have merit as long as I can easily report on and export individual Pools.

There is a custom query in the fhugdownloads:contents:query_list_single_relation_pool|> FHUG Downloads > Query:Individual ~ List Single Relation Pool that lists all Individuals of a chosen Pool number.

There is a standard feature that allows Reports and Exports to be confined to the Individuals listed by a Query.
So, yes, it is easy to Report or Export individual Pools or any combination of Pools.

[koornalla]

Hi Colin,

Do you use Pools as a tool as Mike suggests in this topic to do practical day to day things such as exporting twig data (as a GEDCOM) and reporting on individual twigs or to you use some other method/identifier to to distinguish between unconnected groups of people?

I am just trying to ascertain the best way to get value out of one combined project.

[Valkrider]

I don't use the Pools. I do use the Gedcom export for people enquiring about their branch. I make a lot of use of the Find Duplicates plugin to identify and match individuals.

At the moment I have one project per country and export the twigs if they emigrate. I am going to look at pulling all these projects into one and using Pools when I have some free time.