# Re-read/update XMP metadata from PDF

**URL:** https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240
**Category:** Help
**Created:** [March 28, 2024, 5:22pm UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240 "2024-03-28T17:22:51Z")
**Posts on this page:** 5
**Page:** 1

<div class="post-metadata">

### Author: ![skreisel](https://avatars.discourse-cdn.com/v4/letter/s/dc4da7/32.png) [@skreisel](https://discourse.jabref.org/u/skreisel)
#### Post date: [March 28, 2024, 5:22pm UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240/1 "2024-03-28T17:22:51Z")

</div>

By 1.) dragging a PDF with XMP metadata onto JabRef or 2.) `Lookup -> Search for unlinked local (PDF) files` JabRef extracts XMP metadata and adds it to the corresponding entry.  
Query: is there a way to re-read/update XMP metadata **from** the PDF later on? Or in other words: we can `Write BibTexEntry metadata to PDF` but can we also read metadata from the PDF a any timepoint?  
Cheers!

---

<div class="post-metadata">

### Author: ![ThiloteE](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.jabref.org/thilotee/32/1008_2.png) [@ThiloteE](https://discourse.jabref.org/u/ThiloteE)
#### Post date: [March 28, 2024, 6:06pm UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240/2 "2024-03-28T18:06:14Z")

</div>

Yes, there is a the option to “parse metadata from pdf”, which opens a metadata merge dialogue. You need to attach a pdf to an entry, then open the entry editor, go to the “General” tab and then click the “Parse metadata from pdf” button. See here in this screenshot:

 ![image](https://canada1.discourse-cdn.com/flex030/uploads/jabref/original/2X/5/582a45f285a625f492ab6746f9ea71d8b8c4a28a.png)

---

<div class="post-metadata">

### Author: ![skreisel](https://avatars.discourse-cdn.com/v4/letter/s/dc4da7/32.png) [@skreisel](https://discourse.jabref.org/u/skreisel)
#### Post date: [March 29, 2024, 8:53am UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240/3 "2024-03-29T08:53:35Z")

</div>

Thanks for your prompt reply!  
I see, I overlooked that option.  
Perhaps a query-after-the-query: any chance we could do this over more than one file at a time, i.e. select the entries and have their PDF’s metadata re-read?  
Thanks and regards!

---

<div class="post-metadata">

### Author: ![ThiloteE](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.jabref.org/thilotee/32/1008_2.png) [@ThiloteE](https://discourse.jabref.org/u/ThiloteE)
#### Post date: [March 29, 2024, 11:50am UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240/4 "2024-03-29T11:50:43Z")

</div>

I think you would need to remove and then re-add or re-link those files to JabRef.  
Not sure, if a more elaborate effort to implement this feature is justified.

I don’t know exactly about your workflow, but there are also more specialised tools out there that can read multiple files at the same time. I can recommend [https://exiftool.org/](https://exiftool.org/). The following is mentioned in its feature list: “Ability to read tags from multiple files simultaneously, allowing metadata to be compared and combined between files”. There is also a GUI here: [ExifToolGUI](https://exiftool.org/gui/)

---

<div class="post-metadata">

### Author: ![skreisel](https://avatars.discourse-cdn.com/v4/letter/s/dc4da7/32.png) [@skreisel](https://discourse.jabref.org/u/skreisel)
#### Post date: [March 29, 2024, 8:47pm UTC](https://discourse.jabref.org/t/re-read-update-xmp-metadata-from-pdf/4240/5 "2024-03-29T20:47:05Z")

</div>

Thanks! I’m using JabRefs excellent Groups feature - that sorts entries by keyword, including nesting. The attached PDFs at times get their keywords updated in an process outside of JabRef.  
So yes, I could rebuild the library from time to time; would have been nicer if it could have been done differently.  
Thanks!