Please do NOT expect all Feature Requests to be actioned automatically. Describing your proposal here will ensure the development team are aware of it, and they will give it careful consideration.
  • Page:
  • 1
  • 2

TOPIC:

Handling of non-standard tag _PRIM 6 days 10 hours ago #1

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752
1 BIRT
2 DATE 1805
1 BIRT
2 DATE 1804
2 _PRIM Y
results in the display attached.

However, without the 2 _PRIM, it still puts the 1804 first, giving the impression that it is preferred. In both cases, in search results, the 1805 is on top.
--
Wes Groleau
UniGen.us/
Attachments:

Please Log in or Create an account to join the conversation.

Last edit: by WGroleau.

Handling of non-standard tag _PRIM 6 days 7 hours ago #2

  • bertkoor
  • bertkoor's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Greetings from Utrecht, Holland
  • Posts: 2444

2 _PRIM Y


I'd say this is illegal GEDCOM. As the translation shows, this tag was intended for images. You used it outside of that context, so ofcourse it won't work.
The preferred one should be the first listed, so you need to change the order.
stamboom.BertKoor.nl runs on webtrees v1.7.13

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 6 days 6 hours ago #3

  • fisharebest
  • fisharebest's Avatar
  • Offline
  • Administrator
  • Administrator
  • Posts: 14947
PhpGedView and webtrees 1 used this tag to control the order of media objects.

Since webtrees 2, we use the order of the media within the individual record.

There is a datafix in the control panel which will re-order your media according to these tags, and then remove them.
Greg Roach - This email address is being protected from spambots. You need JavaScript enabled to view it. - fisharebest.webtrees.net

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 6 days 28 minutes ago #4

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752

2 _PRIM Y


I'd say this is illegal GEDCOM. As the translation shows, this tag was intended for images. You used it outside of that context, so ofcourse it won't work.
The preferred one should be the first listed, so you need to change the order.


This came from a test suite. It's obviously not in the spec (custom tag), but the creator of the test had some explanation which I'll have to read again. However, in the following test, the 1805 was first in the GEDCOM and webtrees put the 1804 first without the _PRIM.
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 23 hours ago #5

  • norwegian_sardines
  • norwegian_sardines's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 2164
Wes,

My assuption is that the _PRIM Y as it relates to a FACT (rather than an image) is that the software that produced it uses it to prioritize duplicate "Fact Type", for example if you enter two BIRT fact for a person the "Primary" or most likely fact is the one marked _PRIM Y.

This is because they are not working with GEDCOM rules (first fact of two fact types) rather they put an attribute on the fact saying "this is the one I want to use in reports and other places that expect or write only one fact of each type".

GEDCOM just says "Use the first one in the list"!


The real question is: "When they generate a GEDCOM does the _PRIM Y fact always get listed first or are they mixed up and they expect the receiving program to honor the tag rather than the order?
Ken

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 23 hours ago #6

  • bertkoor
  • bertkoor's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Greetings from Utrecht, Holland
  • Posts: 2444

This came from a test suite.


Aha. The purpose then was to have lots of dubious edge cases and look how the software behaves. That's what test suites are for.

When outside of the specs, it's undefined. Or rather in this case: the first tag gets shown. Like Ken wrote, that's the official expected behaviour.
Like Greg wrote: there is a datafix for images with _PRIM Y.

I'd say webtrees passed this particular test. You may ofcourse disagree.
stamboom.BertKoor.nl runs on webtrees v1.7.13

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 23 hours ago #7

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752
The test specifically says

This test is intended to determine if the target application supports the custom _PRIM GEDCOM tag.

So, I would say that webtrees does support it, but that if we choose to support it for items other than media, then the translation for it should be different.

And supporting it instead of using the order makes us technically noncompliant with the spec (which is sometimes a good thing).

But putting the second BIRT first when the _PRIM is missing violates the spec without good reason. In search results, the dates are both displayed in the GEDCOM order, but in the INDI page, they were out of order in both tests.
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Last edit: by WGroleau.

Handling of non-standard tag _PRIM 5 days 22 hours ago #8

  • bertkoor
  • bertkoor's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Greetings from Utrecht, Holland
  • Posts: 2444
webtrees did support it in v1.7.x, only for images.
Support was dropped since v2.0
stamboom.BertKoor.nl runs on webtrees v1.7.13

Please Log in or Create an account to join the conversation.

Do you need a web hosting solution for your webtrees site?
If you prefer a host that specialises in webtrees, the following page lists some suppliers able to provide one for you: 

Handling of non-standard tag _PRIM 5 days 22 hours ago #9

  • fisharebest
  • fisharebest's Avatar
  • Offline
  • Administrator
  • Administrator
  • Posts: 14947
We have a datafix that applies _PRIM to media links.

I wasn't aware that other applications apply it to other record types.

It would be trivial to add this - github.com/fisharebest/webtrees/issues/4068

A question for the community:

Some data fixes could be applied automatically/silently during import. This is one of them.

Should webtrees

* automatically "fix" unambiguous cases like this
* leave data exactly as it is and let the user be responsible for applying fixes.

It depends on whether you consider webtrees to be a genealogy application or a GEDCOM editor/viewer.

My preference is for the first case. We already have a small number of silent auto-fixes. For example, converting tags such as URL to WWW, etc.
Greg Roach - This email address is being protected from spambots. You need JavaScript enabled to view it. - fisharebest.webtrees.net

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 22 hours ago #10

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752
I'd vote to delete _PRIM on import or ignore it and omit it from export.

But also, when _PRIM is not there, why do I get the second BIRT first on the INDI page and second in the search results?
unigen.us/tree/Assessment/individual/I16...5-Event-Primary-I164
unigen.us/tree/Assessment/search-general?query=PRIM
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 22 hours ago #11

  • norwegian_sardines
  • norwegian_sardines's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 2164
Greg asked:

Should webtrees

* automatically "fix" unambiguous cases like this
* leave data exactly as it is and let the user be responsible for applying fixes.

It depends on whether you consider webtrees to be a genealogy application or a GEDCOM editor/viewer.

1) I consider webtrees a genealogy program, this is not PGV!

2) Leave data exactly as it is, don't fix anything Automatically without a positive 'do this for me'! If (as I think we are) implementing dialects of GEDCOM, webtrees needs to support what the users send us rather than trying to think for them what it should be. If they move to webtrees as their new genealogy program (because it does genealogy great and has great reports) then they will adjust their data over time to support the way webtrees does things just like they did when they first picked up their current program. If they use webtrees as a viewer then maybe the program they use for data entry does things that they don't want webtrees messing with (I don't know what that might be but I'm not buying other programs to find out!).

Again just my thoughts, toss them out if you disagree!!

EDITED in red
Ken

Please Log in or Create an account to join the conversation.

Last edit: by norwegian_sardines.

Handling of non-standard tag _PRIM 5 days 21 hours ago #12

  • hermann
  • hermann's Avatar
  • Offline
  • Senior Member
  • Senior Member
  • Posts: 357
I vote to leave data exactly as it is (keep the _PRIM) but do not interpret this custom tag automatically. The standard is clear on this point and there is no need for a custom tag _PRIM, but if it is there: ok keep it.

In my opinion, webtrees has a unique value, that it keeps the imported GEDCOM records nearly untouched even if they could not be interpreted. So you can export in webtrees an imported GEDCOM file and nearly everything is as it was before. I don't know any other genealogical program with this behavior.

Maybe it makes sense to offer an additional option in the data fix to not only correct the sequence of images but also to correct the sequence of other record types. Then it is the user who decides how to use (or better not use) the _PRIM tag.
Hermann
Designer of the custom module "Extended Family"

webtrees 2.0.17 (all available custom modules installed, php 7.4.15, MySQL 5.6) @ ahnen.hartenthaler.eu

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 20 hours ago #13

  • JustCarmen
  • JustCarmen's Avatar
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1018

I'd vote to delete _PRIM on import or ignore it and omit it from export.


No, no please don't omit it from export! I know webtrees doesn't support the _PRIM tag but other programs do. From time to time I make an export from my webtrees site in order to import it into a Dutch genealogy site. This site uses the _PRIM tag in order to determine which image to use as the main image for an individual. So I need this tag on export.

Carmen
Designer of the JustLight theme (comes with a light and dark color palette), Fancy Imagebar and Fancy Research Links for webtrees 2


Check my website at www.justcarmen.nl

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 18 hours ago #14

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752

I'd vote to delete _PRIM on import or ignore it and omit it from export.


No, no please don't omit it from export! I know webtrees doesn't support the _PRIM tag but other programs do. From time to time I make an export from my webtrees site in order to import it into a Dutch genealogy site. This site uses the _PRIM tag in order to determine which image to use as the main image for an individual. So I need this tag on export.


How about: keep the tag, but DON'T let it change behavior. BUT, during import, issue a message stating that it is in opposition to the
GEDCOM spec and therefore webtrees will not honor it. I can't tell whether webtrees supports it, because it puts things in the same order whether or not _PRIM is there. And the order in search is different from the order in details.
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Last edit: by WGroleau.

Handling of non-standard tag _PRIM 5 days 17 hours ago #15

  • Peter_S
  • Peter_S's Avatar
  • Offline
  • Junior Member
  • Junior Member
  • Posts: 212

Should webtrees

* automatically "fix" unambiguous cases like this
* leave data exactly as it is and let the user be responsible for applying fixes.

It depends on whether you consider webtrees to be a genealogy application or a GEDCOM editor/viewer.


I am undecided about the correct behavior of webtrees when importing a GEDCOM file.

On the one hand, we should follow the GEDCOM standard and adjust tags and structures if they do not comply with the standard. This is because otherwise, when exporting, a bad GEDCOM file will be generated. This "bad data" will then be blamed on webtrees, because the recipient of this exported GEDCOM file will only see the name of the creating program in the header.

On the other hand, we want to ensure that the user does not suffer any loss of data when importing a foreign GEDCOM file.

Even if webtrees with version 2.1 is able to map custom tags and structures of other programs, this should still be in accordance with the standard. We should not still support bad programs and ennoble their faulty code by representing wrong structures in webtrees.

However, the modification of the data should be documented. An import log would be useful.

Alternatively, we can consider a checkbox to activate data cleansing during import. However, this does not solve the problem that webtrees exports incorrect data under its own name.

webtrees is an independent program with a high quality reputation. We should not carelessly put this reputation at risk.

In the case of _PRIM, it is a standard-compliant custom tag. It is possible to issue a warning during import (import log) that a different solution is provided for this tag in the standard and that there is a corresponding cleanup program for a conversion.
Peter

webtrees 1.7.18 and 2.0.17, vesta modules
PHP 7.4.3, MySQL 5.7.25
Webhosting: genonline.de

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 16 hours ago #16

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752
GEDCOM 5.5.1, page 20:

The occurrence of equal level numbers and equal tags within the same context imply that multiple opinions or multiple values of the data exist. The significance of the order in these cases is interpreted as the submitter's preference. The most preferred value being the first with the least preferred data listed in subsequent lines by order of decreasing preference. For example, a researcher who discovers conflicting evidence about a person's birth event would list the most credible information first and the least credible or least preferred items last.

So, for BIRT or OBJE or any other page with the same level number (under the same higher numbered tag), if we honor the _PRIM tag, we are non-compliant. If we export in the wrong order, we are non-compliant, whether or not we include a _PRIM tag. If we choose to provide a _PRIM tag, we should still use the correct order.

And, if an INDI has more than one BIRT or other level one item, those of the same type should be in the order imported, not in chronological order. But then, how complex would it be to sort BIRT…DEAT…BIRT…DEAT to comply with the above paragraph AND to not confuse a viewer who never even heard of GEDCOM?

And for the same person, should there be some sort of pop-up or footnote to explain why there are two births?
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 15 hours ago #17

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752

Even if webtrees with version 2.1 is able to map custom tags and structures of other programs, this should still be in accordance with the standard. We should not still support bad programs and ennoble their faulty code by representing wrong structures in webtrees.


Agree with not enabling non-standard behavior, to avoid the word "bad."  And yet most of us agree that in some places, obeying the published spec is bad behavior because the spec is so sloppy. But deviations, however well-intentioned, just put us on the same level as all the other incompatibilities out there.

However, the modification of the data should be documented. An import log would be useful.

In the case of _PRIM, it is a standard-compliant custom tag. It is possible to issue a warning during import (import log) that a different solution is provided for this tag in the standard and that there is a corresponding cleanup program for a conversion.


And such a message should be provided. For export, can the header include a NOTE stating what is non-compliant (if anything)? And if the DEST specifies a particular program, is it "legal"
to include a NOTE saying why certain things might be imported incorrectly by the DEST?

I think most recipients will not even know there's anything wrong, and only those that can actually read the HEAD will blame webtrees.
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 14 hours ago #18

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752
To further muddy the waters:

(1) I verified (by moving the _PRIM from one record to another) that webtrees 2.0.17 uses the order the spec requires, not what the non-standard _PRIM suggests. However, it does add a separate row saying "Highlighted image: Yes" on the item with the _PRIM tag. Bad? Good?

(2) in many cases, "edit raw GEDCOM" showed that the input was preserved. However, in one test, the input was
0 @I183@ INDI
1 NAME 04-5.5.1 Direct (v2) /Exhibit Person/
1 SEX M
1 OBJE
2 FILE C:\GedcomAssessment\exhibit-person-04P.jpg
3 FORM JPG
3 TITL 04-5.5.1 Direct (v2) Primary Title
2 NOTE 04-5.5.1 Direct (v2) Primary Note
2 _PRIM Y
1 OBJE
2 FILE C:\GedcomAssessment\exhibit-person-04N.jpg
3 FORM JPG
3 TITL 04-5.5.1 Direct (v2) Title
2 NOTE 04-5.5.1 Direct (v2) Note
but "edit raw GEDCOM" showed
0 @I183@ INDI
1 NAME 04-5.5.1 Direct (v2) /Exhibit Person/
1 SEX M
1 OBJE @X2656@
1 OBJE @X2657@
So, we are converting the legal "direct" items to the preferred cross-referenced version. Good? Bad?
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

Handling of non-standard tag _PRIM 5 days 10 hours ago #19

  • norwegian_sardines
  • norwegian_sardines's Avatar
  • Offline
  • Platinum Member
  • Platinum Member
  • Posts: 2164
Gentlemen!

Greg made this statement earlier and I wanted to point it out again:

It depends on whether you consider webtrees to be a genealogy application or a GEDCOM editor/viewer.

I answered this statement as a genealogist that thinks GEDCOM has almost everything I need to record my research for my family as well as the people that ask me to help them with their own family research. It is a genealogy program.

However, and this is important, the trend as I see it is that most of the new users of webtrees appear to want to use the software as a way to expose their research (done with other software) to their family on the web. They may be hoping that some family members will use webtrees to reply with additional data which may be a wish that rarely is fulfilled! These people are either "Up Loaders" or "Round Trippers".

This group of webtrees users either want webtrees to not touch their GEDCOM at all so it stays compatible with their primary genealogy program (Round Trippers) or display their GEDCOM with a hands off upload (the Up Loaders) that will be repeated on a regular basis. Neither of these two type want or care what the GEDCOM Standard is, they just want webtrees to display online using the data they generated on their primary system.

I will most likely never load a new GEDCOM into webtrees, or if I do it will be the rare occasion when a client gives me data they have about their family and my work will be to add any additional information I find and return the GEDCOM back to them. These GEDCOMs will be loaded on my server, but will never be exposed to the outside world, so I don't care if they look pretty on the web! I will also probably not be looking to download a GEDCOM, unless I leave webtrees or I offload that rare GEDCOM that a client has given me. If I need information out of webtrees I can run code that accesses the database directly!

My position has always been to use GEDCOM as it is, for the most part it works real well.

So.... If webtrees is a GEDCOM viewer (the trend is clear) then it needs to support every dialect of GEDCOM that is out there and do it well without data loss. Users of webtrees will continue to use their primary program for generating reports, creating books automatically, and doing research by reading hints and shaky leaves from their primary research tools. They will want to have facts that are less about genealogy and more about Family History, Address books, Social Media, Social Relationships, Photo Galleries and Manipulation, and probably more. Many of these things people have asked for in the forum! I see no problem in this much needed use!

Again, these are my thoughts and you can do what you want with them.
Ken

Please Log in or Create an account to join the conversation.

Last edit: by norwegian_sardines.

Handling of non-standard tag _PRIM 5 days 9 hours ago #20

  • WGroleau
  • WGroleau's Avatar Topic Author
  • Offline
  • Elite Member
  • Elite Member
  • Posts: 1752

I answered [Greg's question] as a genealogist that thinks GEDCOM has almost everything I need to record my research for my family as well as the people that ask me to help them with their own family research. It is a genealogy program.

…[snip]…

My position has always been to use GEDCOM as it is, for the most part it works real well.


That has always been my approach as well. Until I found webtrees, I edited my GEDCOM file directly and only used other programs as viewers (because raw GEDCOM put fewer limitations on me than any program I have ever seen (other than webtrees).

So.... If webtrees is a GEDCOM viewer (the trend is clear) then it needs to support every dialect of GEDCOM that is out there and do it well without data loss.


On the other hand, if the few people (anyone bedsides Greg?) actually maintaining webtrees agree with you and me, they are under no obligation to support the "trend." But if they choose to do so, those that agree with us have the option under open source to not upgrade or even to maintain a diverging version (but hopefully give it a new name).
--
Wes Groleau
UniGen.us/

Please Log in or Create an account to join the conversation.

  • Page:
  • 1
  • 2
Powered by Kunena Forum