Correcting the spelling of Names

The spelling in my case is that all the same family names are spelt the same way but incorrectly. They are all listed under the same title so I thought that if I could find a way to change the title surname all the surnames under that title would be changed.

Unfortunately you can’t. A name and all its components are “private” data in a Person record. A “name”, if considered as a record by itself, cannot be shared between persons (design choice). Consequently, you’re doomed to edit every person.

In addition, in old records, you can never be sure about a name spelling because they were written in an essentially oral society. Spelling would highly depend on the culture and literacy of the clerk. The same person may be designated with a spelling at birth, another one at marriage and even another one at death, not speaking of the records for children birth. My practice is to record all these variants as alternate names, birth name being the main one.

Thank you for the explanation. I hoped there was a way but understand why there can’t be. I’ll deal with it.

The Name Editor Tool and Name Merge Tool addons are specifically for harmonizing names.



They are available through the Isotammi project’s list of Addons for Gramps.

As with all tools that do mass edits, MAKE A BACKUP before using this tool.

They appear quite similar. The Name Editor has powerful search & replace and could be complex to learn. The Name Merge ensure that the names will match an existing one.

(Looks like there are a couple untranslated Finnish labels in the Name Merge. Here’s Google Translate’s interpretation: Nimiä = Names ; Henkilöitä = Personal age. I have NO idea what “personal age” could mean. )

But @kku , @PeterPower , @jpek-m , and @TimNal could tell us if we’re missing an alternative (besides SuperTool… which can do almost anything!) The Isotammi project has been coordinating efforts of multiple users towards a collective genealogical project. So they have a lot of tools for normalizing and merging data.

Export to a text format (e.g. a ‘gramps’ file), do a search-and-replace (uncompressing and then recompressing if necessary), and then reimport.

Brian, it is “Henkilöitä”, not “Henkilöikä” and it means simply “Persons” (or People), ie. “The number of people”. Will be fixed at some point.

ti 14. maalisk. 2023 klo 13.56 Brian McCullough via The Gramps Project (Discourse Forum & Mailing List) (notifications@gramps.discoursemail.com) kirjoitti:

1 Like

We have a case worse, a family name that is spelt differently, with a C in “proper French” and with a Q in the regional language “Auvergnat”. Before 1850, in most records which contain the name more than once, there are both spellings in the same single document.
When editing, it is a royal pain to have the individuals in two hierarchies. I’m not sure how to fix this.

In the Name Editor, there is the Group As option. It allows bringing various spellings of a surname into the grouping. In the example I used was some branches of a family using “Clarke” while most stayed with “Clark”

1 Like

The Grouping override is very useful for collating Surname variants. But unfortunately, there isn’t a corresponding “ignore surname on Group subSort” functionality.

So your Grouped ‘Smith/Smythe/Smyth’ will have separate subgroupings for each Surname Variant (and each multiple Surname combo) subsorted on each Given name.

This why I set my default display name to Given Surname. This brings John Clark together with John Clarke. This has the additional benefit of having the name displayed in reports without the comma.

See the fix in feature request 0011788: Sorting of people with grouped surnames,

1 Like

I had forgotten about this fix. It should be merged into the code.

1 Like

I had been thinking about sorting of names where I had set various surnames together using Group As. I just made a major change.

I just set ALL name records to use the Sort override in the name editor using the name display option “Given”. (Lots of search and replace of the XML backup.)

Using this setting and removing the FR 0011788 has all the grouped surnames mixed together with a proper sort by the Given name. My default display name remains “Title Given "Nickname" Surname, Suffix”.

With all records set, adding a child to a family, or a father to a child the alternate sort option carries over with the father surname guessing set. The only new records that will need to be manually set will be new mothers of the family or a new spouse.

Thank you for this hint, but since it doesn’t sort the result right, and since I am not comfortable with implementing the “fix in feature request 0011788”, I’ll wait until the feature is implemented.

That sounds like it will work well for you in this case. But are there other places in Gramps (besides the People category view) where names are sorted, but not grouped?

In the People view, if you sort on Name it will sort the list based how the name as displayed. Lists People will show in various References tabs but again the default would be something other than alphabetical by name.

Maybe the NarrWeb when a list of other people share a Place record. I would have to see how this type of list is currently ‘sorted’.

I haven’t used the Book feature, but I understand it can include an alphabetical index.

In many genealogies, the are a LOT of namesakes. So they qualify names thoughout with Birthyear-deathyear or with pedigree breadcrumbs.

Perhaps Gramps could have something for People that is similar to fully qualified Place titles which uses those formats, then we could make that the 1st (sorting) column?

Sometimes it’s good to have firstnames unsorted: in not grouped people view and in some other cases.

To avoid these issues, I’ve created this SuperTool script to set or unset this sorting mechanism:

[Gramps SuperTool script file]

version=1

[title]
Individus - Sort_as des individus groupés

[category]
People

[initial_statements]
# The last Format entry in name preferences must be / La dernière entrée du Format des noms dans les Préférences doit être: Prénom, Patronyme[nom]
# /!\ Do not checked Commit Changes, it's implicit in the script/ Ne pas cocher Commit Changes, le commit est explicite dans le script

sort_by_fn = True # By firtsname / Par prénom -> True; By lastname / Par nom -> False 

if sort_by_fn:
    sort_order = -1
else:
    sort_order = 0

[statements]
nobjs = nameobjs
if "<<" in db.get_name_group_mapping(nobjs[0].get_group_name()):
    modifie = False
    for n in nobjs:
        if n.get_sort_as() != sort_order:
            n.set_sort_as(sort_order)
            modifie = True
    if modifie:
        db.commit_person(person, trans)

[filter]

[expressions]

[scope]
all

[unwind_lists]
False

[commit_changes]
False

[summary_only]
False

Note that I use if “<<” in … as test to find grouped people because I name them all like that. You should probably need to change this.

1 Like