[Archivesspace_Users_Group] ArchivesSpace bug report - use of &

[Emma Jolley]

Hi

We have had a persistent issue with the export of the '&' character in EAD XML - we can get around it by using the encoding term "&amp" instead of &.

The issues is currently causing an problem with the synchronisation between ArchivesSpace and Preservica and is further explained in the email below.

It would be great if this issue could be fixed in the next release?

Please let me know if you'd like me to log this in formal system (JIRA?)

Many thanks

Emma

Emma Jolley| Curator of Digital Archives, Pictures and Manuscripts Branch|National Library of Australia Canberra ACT 2600
e: emma.jolley at nla.gov.au<mailto:emma.jolley at nla.gov.au>|t: 02 6262 1456| www.nla.gov.au/ms<http://www.nla.gov.au/ms>
http://www.nla.gov.au/support-us/make-a-collection-offer-pictures-and-manuscripts

From: Ninh Nguyen (IT)
Sent: Friday, 14 July 2017 10:51 AM
To: Emma Jolley <EJOLLEY at nla.gov.au<mailto:EJOLLEY at nla.gov.au>>
Cc: Rachael Eddowes <reddowes at nla.gov.au<mailto:reddowes at nla.gov.au>>; Ninh Nguyen (IT) <nnguyen at nla.gov.au<mailto:nnguyen at nla.gov.au>>; Kavinga Hubert <khubert at nla.gov.au<mailto:khubert at nla.gov.au>>; Shaun Coen <scoen at nla.gov.au<mailto:scoen at nla.gov.au>>
Subject: ArchivesSpace bug report

Hi Emma,

During the Preservica ArchicesSpace synchronisation testing, I think we've found a bug with the ArchivesSpace export to EAD module. I believe you are the contact person for ArchivesSpace? Can I pass this bug report to you to pass on to the ArchivesSpace people? Or should I contact ArchivesSpace directly by some means?

The bug was discovered when we tested the sync, but it doesn't relate to the Preservica sync. Any user can create the bug by editing some fields in ArchivesSpace.

And the bug is:

ArchivesSpace allows some fields to store html code. For example, following a sync, the following html code is store in the Sub Notes field of an 'Other Finding Aids' Notes field.
http://tallis.nla.gov.au:8080/resources/3644#tree::resource_3644
<a href="https://preservica-test.nla.gov.au/explorer/explorer.html#prop:4&0a5b4e7f-5fbe-4f99-832a-d69e0b5302b7">Preservica Internal</a>

When the collection is exported to EAD in xml, ArchivesSpace escaped characters inside the href to make the xml well-formed. The above will turn into a correct form (0a5b4e7f-5fbe-4f99-832a-d69e0b5302b7 is the Preservica UUID)
<a href="https://preservica-test.nla.gov.au/explorer/explorer.html#prop:4&0a5b4e7f-5fbe-4f99-832a-d69e0b5302b7">Preservica Internal</a>

However, if the UUID starts with character 'a', for some reason the export won't escape the ampersand, eg.
http://tallis.nla.gov.au:8080/resources/3646#tree::resource_3646
<a href="https://preservica-test.nla.gov.au/explorer/explorer.html#prop:4&aba6ac48-a17e-4096-ae9a-90a432b45227">
won't turn & to & and it makes the EAD xml not well-formed.

You can test it by creating a new record, and adding some html code (with &) in some fields before exporting it to EAD.

I've tested it in ArchivesSpace 1.5.4 as well as the latest version 2.0.0 and the results are the same.

Could you pass it to the ArchivesSpace people, or let me know how to report it?

Thanks,
Ninh




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lyralists.lyrasis.org/pipermail/archivesspace_users_group/attachments/20170714/9d653d16/attachment.html>