Gramps ID re-ordering

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Gramps ID re-ordering

brian fitzgerald
Hi everyone,
 I have just performed a re-ordering of the Various Gramps IDs. It became desireable after a series of gedcom imports which created a number of unusual, out of sequence IDs. It went well but there was one drawback

Afterwards my custom filters were all confused since the renumbering utility does not take into account the various filters which depend upon a rule combination referencing one or more IDs, such as "descendant filter of <person ID>". There are quite a lot of them which can reference an ID and since these rules are not processed by the utility they break everything on the next run.

It is probably not an easy thing to fix but one thing to help someone recover from this scenario would be to have an additional text field in such rules where the user can free-form annotate the entity which they intend to be represented by the (now wrong!) ID. It would be an optional reminder. That way the new ID (post reordering) can be found after reordering by manually looking up the labeled entity . Clunky, but better than having to search through archive reports for 'who really was I12345 after all, because it now points to Ghengis Khan instead of uncle Ned, or was it uncle George?'.

Another thing would be to fully process these rules from the reordering utilty itself. The utility would need to iterate through all custom rules with an ID which was being changed to update them to the new ID. 

Lastly a simple reporting utility might capture the existing custom filters' architecture, their interrelationships and their IDs and output them in a narrative format to a pdf for printing. Something which could be done prior to using the renumbering utility, or part of it.

It might output something like;
(a) rule X uses rule Y and rule Z , both must apply. 
(b) Rule Z is "persons with an ID of <ID>" where ID is I12345, pointing to "uncle Ned". 
etc, etc

Thinking about it now it seem like all of these these might be useful to help plan, manage, administer and back-up the custom filters.

I recovered I think after a few hours, a paper cut and some frayed nerves.

Any thoughts? Is any of this worthy of a feature request?

Brian



_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Gramps ID re-ordering

GRAMPS - User mailing list
Brian Fitzg,

I think I'd be more worried about Markup links in my Notes text.  Have you've used the Link Editor to add a Person link type? Does the Link point to the right person after a re-order? (I suspect those links will be unaffected because the Link Editor creates "gramps://Person/handle/” links that use Handles rather than IDs. Handles to change with the IDs.
I hope Links survive a re-order because you can't search or filter Note content for embedded links.)


Can't tell if Bookmarks are ID or Handle based.

Maybe you could put in a request to convert Rule to Handles? (but hopefully continue to display the more user-friendly IDs.)

Personally, I dislike creating Rules based on a Person ID. I'm always having drill down so far to run a Filter on a different focus Person.  8t would be nice if I could set the rule for the current "Home Person" or "Active Person". You can change those quickly. ("Active Person" might be unstable because of the the danger of recursive re-filter loops. The Filters re-apply when Windows or Sort columns are changed.)

Brian McC


On Mon, Mar 11, 2019 at 14:59, brian fitzgerald
Hi everyone,
 I have just performed a re-ordering of the Various Gramps IDs. It became desireable after a series of gedcom imports which created a number of unusual, out of sequence IDs. It went well but there was one drawback

Afterwards my custom filters were all confused since the renumbering utility does not take into account the various filters which depend upon a rule combination referencing one or more IDs, such as "descendant filter of <person ID>". There are quite a lot of them which can reference an ID and since these rules are not processed by the utility they break everything on the next run.

It is probably not an easy thing to fix but one thing to help someone recover from this scenario would be to have an additional text field in such rules where the user can free-form annotate the entity which they intend to be represented by the (now wrong!) ID. It would be an optional reminder. That way the new ID (post reordering) can be found after reordering by manually looking up the labeled entity . Clunky, but better than having to search through archive reports for 'who really was I12345 after all, because it now points to Ghengis Khan instead of uncle Ned, or was it uncle George?'.

Another thing would be to fully process these rules from the reordering utilty itself. The utility would need to iterate through all custom rules with an ID which was being changed to update them to the new ID. 

Lastly a simple reporting utility might capture the existing custom filters' architecture, their interrelationships and their IDs and output them in a narrative format to a pdf for printing. Something which could be done prior to using the renumbering utility, or part of it.

It might output something like;
(a) rule X uses rule Y and rule Z , both must apply. 
(b) Rule Z is "persons with an ID of <ID>" where ID is I12345, pointing to "uncle Ned". 
etc, etc

Thinking about it now it seem like all of these these might be useful to help plan, manage, administer and back-up the custom filters.

I recovered I think after a few hours, a paper cut and some frayed nerves.

Any thoughts? Is any of this worthy of a feature request?

Brian


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Gramps ID re-ordering

Dave Scheipers
The Link editor in notes works off of the internal database id string.
So altering the User's ID numbers should have no affect.

The first time I ran the new (5.0) Reorder ID tool I noticed that the
new ID order bore no resemblance of the old order. The new numbers
were scrambled. My own record went from I000025 to I001497. I quickly
restored from the backup and now when I run the tool I make sure that
I exclude People.

I filed a bug https://gramps-project.org/bugs/view.php?id=10641

I suggested a stronger warning be issued. Now that I have explored the
underlying code files, I can come up with the suggestion and which
lines would need to be changed.

Dave

On Mon, Mar 11, 2019 at 4:49 PM Emyoulation--- via Gramps-users
<[hidden email]> wrote:

>
> Brian Fitzg,
>
> I think I'd be more worried about Markup links in my Notes text.  Have you've used the Link Editor to add a Person link type? Does the Link point to the right person after a re-order? (I suspect those links will be unaffected because the Link Editor creates "gramps://Person/handle/” links that use Handles rather than IDs. Handles to change with the IDs.
> I hope Links survive a re-order because you can't search or filter Note content for embedded links.)
>
> https://www.gramps-project.org/wiki/index.php/Gramps_5.0_Wiki_Manual_-_Entering_and_editing_data:_detailed_-_part_2#Link_Editor
>
> Can't tell if Bookmarks are ID or Handle based.
>
> Maybe you could put in a request to convert Rule to Handles? (but hopefully continue to display the more user-friendly IDs.)
>
> Personally, I dislike creating Rules based on a Person ID. I'm always having drill down so far to run a Filter on a different focus Person.  8t would be nice if I could set the rule for the current "Home Person" or "Active Person". You can change those quickly. ("Active Person" might be unstable because of the the danger of recursive re-filter loops. The Filters re-apply when Windows or Sort columns are changed.)
>
> Brian McC
>
>
> On Mon, Mar 11, 2019 at 14:59, brian fitzgerald
> <[hidden email]> wrote:
> Hi everyone,
>  I have just performed a re-ordering of the Various Gramps IDs. It became desireable after a series of gedcom imports which created a number of unusual, out of sequence IDs. It went well but there was one drawback
>
> Afterwards my custom filters were all confused since the renumbering utility does not take into account the various filters which depend upon a rule combination referencing one or more IDs, such as "descendant filter of <person ID>". There are quite a lot of them which can reference an ID and since these rules are not processed by the utility they break everything on the next run.
>
> It is probably not an easy thing to fix but one thing to help someone recover from this scenario would be to have an additional text field in such rules where the user can free-form annotate the entity which they intend to be represented by the (now wrong!) ID. It would be an optional reminder. That way the new ID (post reordering) can be found after reordering by manually looking up the labeled entity . Clunky, but better than having to search through archive reports for 'who really was I12345 after all, because it now points to Ghengis Khan instead of uncle Ned, or was it uncle George?'.
>
> Another thing would be to fully process these rules from the reordering utilty itself. The utility would need to iterate through all custom rules with an ID which was being changed to update them to the new ID.
>
> Lastly a simple reporting utility might capture the existing custom filters' architecture, their interrelationships and their IDs and output them in a narrative format to a pdf for printing. Something which could be done prior to using the renumbering utility, or part of it.
>
> It might output something like;
> (a) rule X uses rule Y and rule Z , both must apply.
> (b) Rule Z is "persons with an ID of <ID>" where ID is I12345, pointing to "uncle Ned".
> etc, etc
>
> Thinking about it now it seem like all of these these might be useful to help plan, manage, administer and back-up the custom filters.
>
> I recovered I think after a few hours, a paper cut and some frayed nerves.
>
> Any thoughts? Is any of this worthy of a feature request?
>
> Brian
>
> _______________________________________________
> Gramps-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gramps-users
> https://gramps-project.org


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|

Re: Gramps ID re-ordering

Doug-11
In reply to this post by GRAMPS - User mailing list
Brian McC

I extract a comment from your post:

<quote>
Personally, I dislike creating Rules based on a Person ID. I'm always having drill down so far to run a Filter on a different focus Person.  8t would be nice if I could set the rule for the current "Home Person" or "Active Person". You can change those quickly.
<endquote>

This has cropped up repeatedly over the years in feature requests (e.g. bugtracker #7881) and it would greatly improve the usefulness of custom filters,

so +1

Doug

On 11/03/2019 20:48, Emyoulation--- via Gramps-users wrote:
Brian Fitzg,

I think I'd be more worried about Markup links in my Notes text.  Have you've used the Link Editor to add a Person link type? Does the Link point to the right person after a re-order? (I suspect those links will be unaffected because the Link Editor creates "gramps://Person/handle/” links that use Handles rather than IDs. Handles to change with the IDs.
I hope Links survive a re-order because you can't search or filter Note content for embedded links.)


Can't tell if Bookmarks are ID or Handle based.

Maybe you could put in a request to convert Rule to Handles? (but hopefully continue to display the more user-friendly IDs.)

Personally, I dislike creating Rules based on a Person ID. I'm always having drill down so far to run a Filter on a different focus Person.  8t would be nice if I could set the rule for the current "Home Person" or "Active Person". You can change those quickly. ("Active Person" might be unstable because of the the danger of recursive re-filter loops. The Filters re-apply when Windows or Sort columns are changed.)

Brian McC


On Mon, Mar 11, 2019 at 14:59, brian fitzgerald
Hi everyone,
 I have just performed a re-ordering of the Various Gramps IDs. It became desireable after a series of gedcom imports which created a number of unusual, out of sequence IDs. It went well but there was one drawback

Afterwards my custom filters were all confused since the renumbering utility does not take into account the various filters which depend upon a rule combination referencing one or more IDs, such as "descendant filter of <person ID>". There are quite a lot of them which can reference an ID and since these rules are not processed by the utility they break everything on the next run.

It is probably not an easy thing to fix but one thing to help someone recover from this scenario would be to have an additional text field in such rules where the user can free-form annotate the entity which they intend to be represented by the (now wrong!) ID. It would be an optional reminder. That way the new ID (post reordering) can be found after reordering by manually looking up the labeled entity . Clunky, but better than having to search through archive reports for 'who really was I12345 after all, because it now points to Ghengis Khan instead of uncle Ned, or was it uncle George?'.

Another thing would be to fully process these rules from the reordering utilty itself. The utility would need to iterate through all custom rules with an ID which was being changed to update them to the new ID. 

Lastly a simple reporting utility might capture the existing custom filters' architecture, their interrelationships and their IDs and output them in a narrative format to a pdf for printing. Something which could be done prior to using the renumbering utility, or part of it.

It might output something like;
(a) rule X uses rule Y and rule Z , both must apply. 
(b) Rule Z is "persons with an ID of <ID>" where ID is I12345, pointing to "uncle Ned". 
etc, etc

Thinking about it now it seem like all of these these might be useful to help plan, manage, administer and back-up the custom filters.

I recovered I think after a few hours, a paper cut and some frayed nerves.

Any thoughts? Is any of this worthy of a feature request?

Brian


_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org




_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org