Ok, just fetched the latest changes and everything still works fine. I parsed and saved a PDF/A created with Acrobat regenerating the document level xmp metadata, the process completed without issues and the resulting xmp stream is valid with everything in place. Thanks Andrea
Il giorno ven 28 nov 2025 alle ore 11:15 Tilman Hausherr <[email protected]> ha scritto: > > Hi, > > Thanks, but please update, build and test again, because I just removed > the change from 2 days ago, it is no longer needed. The only difference > I found is a different sequence of namespace declarations, and that > unused namespace declarations are lost. > > Tilman > > Am 28.11.2025 um 11:06 schrieb Andrea Vacondio: > > Ok, everything seems to be working fine, thanks! > > Andrea > > > > Il giorno gio 27 nov 2025 alle ore 16:42 Andrea Vacondio > > <[email protected]> ha scritto: > >> Great, thanks, I'll give it a try tomorrow. > >> Andrea > >> > >> Il giorno gio 27 nov 2025 alle ore 16:38 Tilman Hausherr > >> <[email protected]> ha scritto: > >>> Hi, > >>> > >>> I've fixed that as well, hopefully. Our Jenkins CI is down so if you'd > >>> like to test this you'll have to build yourself. It's sad that nobody, > >>> including me when I introduced the tests in PDFBOX-6026, had the idea to > >>> parse the serialized XMP content 🤦♂️ > >>> > >>> Tilman > >>> > >>> Am 27.11.2025 um 09:02 schrieb Andrea Vacondio: > >>>> It doesn't seem to have fixed the issue. I can share the xmp stream > >>>> but you can also use one of those already there, like > >>>> validxmp/Notepad++_A1b.xmp > >>>> If you parse it, serialize it and re-parse it, it will fail > >>>> complaining that the stEvt prefix is not associated... I think it's > >>>> because the > >>>> xmlns:stEvt="http://ns.adobe.com/xap/1.0/sType/ResourceEvent#"; > >>>> is gone in the resulting file. > >>>> Andrea > >>>> > >>>> Il giorno mer 26 nov 2025 alle ore 21:16 Tilman Hausherr > >>>> <[email protected]> ha scritto: > >>>>> fixed > >>>>> > >>>>> https://repository.apache.org/content/groups/snapshots/org/apache/pdfbox/xmpbox/3.0.7-SNAPSHOT/ > >>>>> > >>>>> Please check that it works now. If not, please share your xmp file. > >>>>> > >>>>> Tilman > >>>>> > >>>>> Am 26.11.2025 um 11:40 schrieb Tilman Hausherr: > >>>>>> Hi, > >>>>>> > >>>>>> This sounds like https://issues.apache.org/jira/browse/PDFBOX-2378 > >>>>>> > >>>>>> Tilman > >>>>>> > >>>>>> Am 26.11.2025 um 11:23 schrieb Andrea Vacondio: > >>>>>>> Hello, > >>>>>>> I've been playing with XMP lately and it seems, unless I'm doing > >>>>>>> something wrong, that XMPBox serialization is not working properly. > >>>>>>> You can modify the TestXMPWithDefinedSchemas to write down the xmp and > >>>>>>> parse it again and they all fail. > >>>>>>> Like this: > >>>>>>> > >>>>>>> @ParameterizedTest > >>>>>>> @MethodSource("initializeParameters") > >>>>>>> void main(String path) throws XmpParsingException, IOException, > >>>>>>> TransformerException > >>>>>>> { > >>>>>>> try (InputStream is = > >>>>>>> this.getClass().getResourceAsStream(path)) > >>>>>>> { > >>>>>>> DomXmpParser builder = new DomXmpParser(); > >>>>>>> XMPMetadata rxmp = builder.parse(is); > >>>>>>> // ensure basic parsing was OK > >>>>>>> assertFalse(rxmp.getAllSchemas().isEmpty()); > >>>>>>> var tmp = Files.createTempFile("prefix", "suffix"); > >>>>>>> new XmpSerializer().serialize(rxmp, > >>>>>>> Files.newOutputStream(tmp), true); > >>>>>>> new DomXmpParser().parse(Files.newInputStream(tmp)); > >>>>>>> } > >>>>>>> } > >>>>>>> > >>>>>>> To give some context, I'm trying to modify xmp metadata created by > >>>>>>> Acrobat (I can provide the xml file if needed) but the metadata stream > >>>>>>> serialized by xmpbox is missing a couple of namespaces defined in the > >>>>>>> original one. > >>>>>>> Andrea > >>>>>>> > >>>>>>> --------------------------------------------------------------------- > >>>>>>> To unsubscribe, e-mail: [email protected] > >>>>>>> For additional commands, e-mail: [email protected] > >>>>>>> > >>>>>> --------------------------------------------------------------------- > >>>>>> To unsubscribe, e-mail: [email protected] > >>>>>> For additional commands, e-mail: [email protected] > >>>>>> > >>>>> --------------------------------------------------------------------- > >>>>> To unsubscribe, e-mail: [email protected] > >>>>> For additional commands, e-mail: [email protected] > >>>>> > >>>> --------------------------------------------------------------------- > >>>> To unsubscribe, e-mail: [email protected] > >>>> For additional commands, e-mail: [email protected] > >>>> > >>> > >>> --------------------------------------------------------------------- > >>> To unsubscribe, e-mail: [email protected] > >>> For additional commands, e-mail: [email protected] > >>> > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [email protected] > > For additional commands, e-mail: [email protected] > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
