Error when writing metadata in JPG files

[jarraun]

Hi Mario

IMatch 2021.102
NEF files from Nikon Z6 edited with NX Studio

After introducing metadata in NEF master files and then writing and propagate to JPG versions, some of the JPG files get tagged with error with this Exiftool output:

-execute9999
    1 image files updated
    1 image files updated
    0 image files updated
    1 files weren't updated due to errors
Warning: Truncated NikonScanIFD directory. IFD dropped. - D:/JAVIER/Mis imágenes/Flujo_Imagenes/Descarga_Tarjeta/img2021/img20211013_pasaia_hondarribia/japirivis20211013_102523_web.jpg
Error: Error reading StripOffsets data in IFD1 - D:/JAVIER/Mis imágenes/Flujo_Imagenes/Descarga_Tarjeta/img2021/img20211013_pasaia_hondarribia/japirivis20211013_102523_web.jpg

If I force update of one of the JPG files the error is gone and then the NEF master shows unwriten metadata:

"xmp::xmpmm\instanceid"

The errors also disappear when I propagate data to versions.

Find attached files whith logs and error fron Metadata analyst. Please dig in the issue because its very annoying. Thank you.

Best regards
Javier

[Mario]

Not sure what I can do here.

The warning error is issued from ExifTool and indicates that there is a problem with the existing metadata in your JPG.
How did you create it? With which software? What does the JPEG contain before ExifTool updates it.

Problems with writing to files are rare but perfectly normal.
Often it helps to re-save the affected file in the originating application.
Since maker notes are affected, maybe the software (?) that has produced the JPG has damaged the maker notes?
Or your propagation rule copies data other than XMP, which can cause problems (especially if you copy willy-nilly EXIF metadata between files).

[jarraun]

Thnks Mario.

I explain my workflow:



• I process my NEF files with NIKON NX Studio (good coors) and save them as TIF 16 bit
• Then I make final touches in these TIF´s with DXO Photolab 5 and exprt them as JPG
• Finally I introduce metadata and key words in IMatch

Supposedly JPG´s contain the same metadata as the original NEF or as less this the option marked for both softwares.

These are te data I propagate:

Please how could I make it working without much hassle ?[/list]

[Mario]

1. I have not seen the images, so I cannot tell you anything.
Please upload and provide a link

Are you sue that DxO does not damage the maker notes? Does it copy the maker notes from the NEF into EXIF? Again, see 1. above.

Make sure you are not propagating unsafe EXIF data from the NEF to the version.
Your screen shot is very tiny on my 4K screen and it does not show all the options you have enabled.
Do you propagate EXIF data from the master to the version? If so, have you made sure that this is safe? EXIF data is not meant to be copied between images, at least not most of it.

[jarraun]

I´m sending you files to you support email.

With metadata I´m sure of nothing, this is beacause I´m posting and have purchased IMatch. As I wrote before DXO don´t touch my NEF´s, just export the TIF genarated by NX Studio to JPG, I don´t know if it copies maker notes to exif in JPG´s.

As you can see in Untitled attached file I´m not propagating exif data.

[Mario]

The warning issued by ExifTool:
Truncated NikonScanIFD directory
and the error:
Error: Error reading StripOffsets data in IFD1

indicates that the file data is somehow corrupted or in a format not understood by ExifTool.
What did the DxO developers tell you?

[thrinn]

Javier,
if I understand your workflow correctly, you have 3 variants of a picture: NEF > TIF > JPG. You wrote:

Supposedly JPG´s contain the same metadata as the original NEF or as less this the option marked for both softwares.

This looks as if you propagate metadata from NEF to TIF using Nikon NX, then propagate metadata from the TIF to the JPG by DXO. I do not have these programs, but surely you can somehow configure what is propagated? Maybe it would be less error prone to switch off propagation for your JPG export in DXO, or at least minimize what is propagated. If you manage all three variants (NEF, TIF, JPG) in IMatch, why not do the propagation from NEF to JPG directly and completely in IMatch, using versioning? This should minimize the chance that something gets messed up in the process, and it should be much easier to fine-tune what is propagated with IMatch.

[jarraun]

Sorry Mario, no time here to contact DXO developers, all these metadata issues take to much time of the life of a hobbyist photographer, Id like software to simplify life instead of complicate it even more.

Thank you Thorsten, you have understand may workflow perfectly, I´m folowing your advice and unmark the options to fill TIFF/JPEG files with metadata and propagate them from NEF masters to JPEG versions in IMatch. Let´s see if it will optimize my work.

[Mario]

I have downloaded your files and had a look.

The NEF

The NEW contains some XMP data, written by the camera:

[XMP-xmp]       Creator Tool                    : NIKON Z 6 Ver.03.31
[XMP-xmp]       Create Date                     : 2021:10:16 12:12:26.63
[XMP-xmp]       Rating                          : 0

Rudimentary, with the Nikon-typical hard-coded "rating = none". For whatever purposes.
The file also contains a subset of the proprietary Adobe Camera RAW default XMP fields.

The file also contains a GPS record, which only consists of the standard GPS version number. Why Nikon bothers to write a partial GPS record, I don't know.
ExifTool reports two minor warnings:

Non-standard SubIFD1 tag 0xc7d5 NikonNEFInfo
Unknown SubIFD1 tag 0xc7d6

Looks like Nikon is stretching EXIF limits again, or made an error writing the sub IFD.

The JPG

Contains EXIF, XMP and the partial GPS data from the NEF.
The XMP record was written by ExifTool 12.31.
The Metadata Analyst reports the following warnings issued by ExifTool (I skipped the GPS warnings because the GPS record consists only of the version number, same as in the NEF)

Warning: [XMP] [ExifIFD]:UserComment not mapped to [XMP-dc]:Description (embedded).
Warning: [Detailed Validation] [minor] Non-standard IFD0 tag 0x4749 RatingPercent
Warning: [Detailed Validation] [minor] IPTC Province-State too long (36 bytes; should be 32 max)
Warning: [Detailed Validation] [minor] IPTC Country-PrimaryLocationCode too short (2 bytes; should be 3)
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashFired should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashFunction should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashRedEyeMode should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-xmpRights:Marked should be capitalized
Warning: [Detailed Validation] Missing required JPEG ExifIFD tag 0x9101 ComponentsConfiguration
Warning: [Detailed Validation] Missing required JPEG IFD0 tag 0x0213 YCbCrPositioning

Nothing really harmful, except maybe for the non-standard RatingPercent tag, which was used only by Microsoft (?) at some time.

I made a copy of the image and changed rating/label, headline, description, copyright, author and added GPS coordinates and location data.
ExifTool reported no problems during write-back.

Metadata Analyst now shows these warnings, basically the same as before:

Warning: [Detailed Validation] [minor] Non-standard IFD0 tag 0x4749 RatingPercent
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashFired should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashFunction should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-exif:FlashRedEyeMode should be capitalized
Warning: [Detailed Validation] [minor] Boolean value for XMP-xmpRights:Marked should be capitalized
Warning: [Detailed Validation] Missing required JPEG ExifIFD tag 0x9101 ComponentsConfiguration
Warning: [Detailed Validation] Missing required JPEG IFD0 tag 0x0213 YCbCrPositioning

Except for the typicial minor metadata issues, I see no problems in the file and no problems with reading or writing the metadata.

[jarraun]

Thank you Mario to expend your time with this problem, I´ll will follow the Thorsten way and let you know if the errors finish.