Treewide searches?

Is there a way to search the entire Tree for a string?

My Tree is starting to be complex enough that I need to re-assess & harmonize its structure. There are enough different places to have tucked tidbits of data that they can become lost.

Let’s take a concrete example.

I’ve changed my mind several times about how to use FindAGrave memorials. Initially, because there is little visual indication of whether a particular source had been located for an Event, I put a FindAGrave notation in Burial Event descriptions.

But I’ve also tried FindAGrave as a source with the memorial ID as the citation page number. Then added a FindAGrave custom attribute because their service doesn’t meet a lot of criteria for a ‘quality’ source. And as an Internet object to see if that would make them easier to re-visit. (It does not save on click-throughs.)

But I’ve also set up a People source for FindAGrave Graver IDs… and a Place source for FindAGrave Cemetery IDs.

My notes also include some cousin eMail conversation transcripts which mention memorials.

Finally, for collateral lines (or distant cousin spouses) where more research is needed, instead of relying on Gramps’ estimated dates, I’ve scraped death year from FindAGrave as a Burial Event with the URL as the Description.

So I want to search the entire database for incidences of “FindAGrave” or “Find A Grave”.

In can see other places where such a search would be useful. Such as searching for an ID (or its internal handle equivalent) that might find the Object, references in Associations, Notes, Custom Filters, or report Books.

As an aside, I’m looking forward to exploring whatever the Developers conjure up for custom citations. FindAGrave has a form they prefer and I’m hoping that we might be able to scrape & store it as an override string. It includes hyperlinked markup. An example:

Find a Grave , database and images (George Thompson (1752-1819) - Find a Grave Memorial : accessed 24 August 2021), memorial page for George Thompson (1752–6 Feb 1819), Find a Grave Memorial ID 42861655, citing Hopewell Cemetery, New Bedford, Lawrence County, Pennsylvania, USA ; Maintained by ctwardo (contributor 47093149).

When I have to make big searches like this, I save my database and open the gramps xml file in notepad++ to make it search for what i want. I’ve gramps opened near to it so i retrieve easily what I’m looking for with the gramps id of the object found in N++.

1 Like

The simplest (not necessarily most pleasant) way is to view the XML or GEDCOM in a text editor.

Or you could create custom filters for each of the variations you’ve used.

As you know, custom filters are stored separately. I imagine report books are as well, but I haven’t used them.

1 Like

That is how I currently do it. I use the citation for dates (if present) of birth and death events, and place of burial event. For place of death event and date of burial event, I try to rely on other sources (death certificate, cemetery record, newspaper article, etc.), since the remains may have been transferred from a distant place of death or a previous burial location.

Not sure what you mean about the attribute but I agree about the questionable quality of the source. It varies a lot according to the contributor. The confidence level in the citation might help.

I attach a “Link” type Note to the citation, containing the URL. (I do that for all sources where possible.)

Not sure what you mean by a Place source. I have cemeteries as Places, and I guess I could add a Note or Internet link to the cemetery’s FindAGrave page, but I haven’t done that yet.

1 Like

It is just a Source applied to a Place. I added Cemetery as a custom place type. (Which seems like it should be a standard type for genealogy software.) And the cemetery listings on FindAGrave often have extra info beyond the absolutely needed Lat/Long coordinates.

Some have alternate names, history of the cemetery, contact info, etc. All of those extras go into a private Source note. (Since I don’t want such trivia to print for every interment there.) And the FindAGrave cemetery ID goes in the citation vol/page.

Example #49269: Arlington National Cemetery listing

I attach an Internet link to the place record for both FindAGrave and BillionGraves. If becomes a quick way to get back to the websites. If the cemetery has their own gravesite search, I will add this as well. I find this sufficient to document the place.

1 Like

Here’s what I have for custom filters:

  1. A source filter for my FindAGrave memorials source ("Source with ").
  2. An event filter for events having citations with the source from #1.
  3. A person filter for people having the source from #1 or events from #2.
  4. A place filter for places having the source from #1 or events from #2.

The approach could be expanded for your situation (multiple Sources).

Let’s not get mired down in the example of FindAGrave. That case for my tree is well-defined although somewhat diversely implemented. I just gave it as an example of how the same data can sprawl into a variety of structured data buckets.

So far, the best option for moderately technical users is: backup to uncompressed XML without Media and searching with Notepad++. (And expanding to do parallel searches in filters & books.)

That works for me as I’m very comfortable reading XML. I’ve been reading chunky formats since the days of Interchange File Format (IFF) in the mid-80s on the Amiga.

Is it going to be workable for our allegorical Aunt Martha? Probably not. But sometimes, I suppose Aunt Martha is just going to need help from a techie.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.