#744 GedCheck Option not to show certain standard deviations

[Meliza Amity]

I would like optionally to disregard in the GedCheck Gedcom problems that applications, especially PhpGedView, create.

The following is a list of comments that I would like not to see in the GedCheck Error list.
I need a maneagable list of real problems.
The problems are now lost among the many standard deviations.

I have no objection if PGV would create cemetery records with for instance _CEM instead of CEME.
(_CEM should be added to the facts.en.php file).
The problem is that desktop applications do not follow the rule that any non standard facts should be of format _XXX.

1. Header
0000001[[0 HEAD]] missing HEAD:SUBM; see HEAD

0000005[[1 GEDC]] missing HEAD:GEDC:FORM; see HEAD

2. 0002407[[2 CEME epitafium i Romfartum kyrka]] invalid tag; see INDI @I181@

3. Addresses
0000112[[1 ADDR Stematsky 8 apartment 78]] invalid tag; see INDI @I1@
0000113[[2 CONT 69362 Tel Aviv]] invalid tag; see INDI @I1@
0000114[[2 CONT Israel]] invalid tag; see INDI @I1@

4. Addresses
0000138[[1 ADDR]] invalid tag; see INDI @I2@
0000139[[2 ADR1 Levitan 6]] invalid tag; see INDI @I2@
0000140[[2 CITY Tel Aviv]] invalid tag; see INDI @I2@
0000141[[2 POST 69204]] invalid tag; see INDI @I2@
0000142[[2 CTRY Israel]] invalid tag; see INDI @I2@
0000143[[1 PHON 03-6424830]] invalid tag; see INDI @I2@
0001410[[1 FAX 07-6517 966]] invalid tag; see INDI @I32@
0001366[[2 RESN confidential]] invalid tag; see INDI @I30@
0001810[[1 ADDR Latskova]] invalid tag; see INDI @I63@
0011058[[1 ADDR Person Name]] invalid tag; see FAM @F392@
0011059[[1 EMAL rklugman@xxx.gov]] invalid tag; see FAM @F392@
0094586[[1 ADDR]] invalid tag; see INDI @I2488@
0094587[[1 EMAIL xxxx@aol.com]] invalid tag; see INDI @I2488@
0139115[[1 WWW http://xxxx.yyyy.fidisk.fi/\]] invalid tag; see INDI @I4659@
0223021[[2 URL http://xxx.se/denise_grunstein/book1/\]] invalid tag; see INDI @I8371@

5. Media
0002495 0 @M9@ OBJE
0002496[[1 FORM jpeg]] invalid tag; see OBJE @M9@
0002497[[1 FILE media/anschell_jonathan.jpg]] missing OBJE:FILE:FORM; see OBJE @M9@
0002498[[1 TITL Jonathan Aaaa]] invalid tag; see OBJE @M9@

6. 0000036[[3 TIME 4:05]] invalid tag; see INDI @I1@

7. FACTs from descktop programs defined in the PGV facts.php
0000226[[1 COMM from Suomi Seura]] invalid tag; see INDI @I10000@
0002725[[1 DATA Jan told to show as anonymous]] invalid tag; see INDI @I10148@

8. Note
0000701[[1 EMAIL ooooo@teleline.es]] invalid tag; see INDI @I10019@
0000702[[2 NOTE old?]] invalid tag; see INDI @I10019@

9. ASSO
0001432 2 DATE 1880
0001433 2 PLAC Helsinki, Finland
0001434[[2 ASSO @I1264@]] invalid tag; see INDI @I10065@

10. More than one NICKname
0003606[[1 NAME Hirsch /Www/]] too many INDI:NAME:NICK; see INDI @I102@

11. Media
0169393 1 OBJE @M184@
0169394[[2 NOTE Icke känd som: Josef Kkkkk]] invalid tag; see INDI @I6088@
0274348 1 OBJE @M1854@
0274349[[2 SOUR @S104@]] invalid tag; see FAM @F4081@

0299759[[1 FILE media/pictures/Skurnik/xxx1.jpg]] missing OBJE:FILE:FORM; see OBJE @M1290@

[Greg Roach]

Logged In: YES
user_id=1466942
Originator: NO

Hi Meliza,

A list of which deviations from the specification are "acceptable" is very subjective; it will be impossible to satisfy everyone with this sort of request. I've coded a "Meliza apecial" version of gedcheck.php. It should address all the points you raise, except for:

1) I've got an RFE open to fix the header format. Plus you'll only get one occurrence of this error.

7) factarray[] doesn't tell me anything about the context of the tag, so I can't use it directly. Instead I've added a hard-coded list of additions. See line 732 for INDI facts and line 736 for FAM facts. Add to the list as you see fit.

I don't have permission to attach anything to this tracker item. Can you assign it to me? I think that will let me do it. Otherwise I'll have to email it to you....

I've grouped all the changes together (seach the code for "Meliza"). Even if you're not a programmer, you should be able to figure out how things work and copy/paste to tweak it yourself. Alternatively, just ask.

Hope this gives you what you want. If not, I'm sure you'll let me know! :-)

[Meliza Amity]

Logged In: YES
user_id=959928
Originator: YES

I have assigned this RFE.
Sounds great.

Meliza

[Greg Roach]

"Meliza Special" version of gedcheck.php

[Greg Roach]

Logged In: YES
user_id=1466942
Originator: NO

File Added: gedcheck.php

[Greg Roach]

Logged In: YES
user_id=1466942
Originator: NO

Meliza,

Can I close this one? I'm not going to submit this change to SVN, as having exceptions would defeat the purpose of a syntax checker.

The attached file should be enough of a hint to show people how to make these sorts of customisations.

Greg

[ColoredPixels]

Logged In: YES
user_id=1337066
Originator: NO

Since I also want what Meliza asked for, I would beg to not close out this ticket. Looking over the coding changes you did for her, it looks like you have 90% of it done already. So, to "formalize" the request and get away from what you would see as subjective, maybe this would help.

I would suggest adding another selection box for:
1) Absolutely strict Gedcom 5.5
2) Absolutely strict Gedcom 5.5.1 (it's still just a draft, right?)
3) Acceptable for PhpGedView (very loose, what we display without errors)

and then if/when you get enough input as to the deviations in the other programs out there,
4) clean enough to export to PAF
5) clean enough to export to Legacy
6) clean enough to export to Family Tree Maker

The languages/facts.en.php at least gives you a start as to what is wanted/accepted by the other programs.

So, where you had the extra code for Meliza, you would simply have
Standard default as strict55.
if(strict551) {add MAP/FAX/WWW }
if(phpGedViewLoose) { add a whole lot. See languages/facts.en.php }
if(PAF) {add a smaller subset from facts.en.php, augmented by user submissions }

Since I only use phpGedView for all of my work, #3 is the checker I would run. Others who export/import their work would like the more strict versions.

And now here is an off topic oddity. If I choose Critical/New/2 lines/errors, I get a blank page after the headers. If I chose 4 lines instead of 2, I get the results I expect. Any other level works with 2. It may just be my system though.

[Greg Roach]

Logged In: YES
user_id=1466942
Originator: NO

After much thought, I'm not going to make these changes. IMHO, the whole point of checking against a spec is to be as strict as possible.

I won't close it, but I'm going to unassign myself from it.

[Meliza Amity]

Logged In: YES
user_id=959928
Originator: YES

For me the gedcheck tool is part of PhpGedView.

It does not make sense that one part of a package reports as an error something that another part creates.

In order to take advantage of this tool, one has to be able not to show variations to the GEDCOM standards that are caused by desktop applications that created the input GEDCOM file.
PGV is supposed to present on the web data that contains these variations.

We want a possibility to highlight actual unknown problems.

After they are solved we can gradually show other problems, so that they too may be solved.

Too many comments at once makes users not want to use the tool.