Archivematica | API integration

[STI]

Hello, I'm currently studying the possibilities of integrating Archivematica with our custom framework, so we can access archived files and automatize the process of transfer and ingest. I was wondering if Archivematica possess such API or functionality, so we can know the file's uuid after archived and the final path of the file for displaying it. And can you guys indicate me to something that further elaborates the service callbacks? Is that what I'm looking for? Thank you.

[Sarah Romkey]

Hello,

There is a forum thread describing currently available APIs here:

https://groups.google.com/forum/#!searchin/archivematica/API|sort:relevance/archivematica/9IYRUJrHrbw/w9QyKhJmAQAJ

But from the sounds of your use case, you will also be interested to know about some current development in Archivematica's Storage Service to make a generic search REST API, which will allow you to query your archival storage from third party applications. This is being developed in partnership with University of York and University of Hull as part of their research data preservation work (funded by Jisc). You can find a list of the kinds of queries we anticipate being possible through this API here:

https://wiki.archivematica.org/Research_data_management#Generic_search_REST_API

We have not merged the code yet but if you'd like to explore it, it's in this branch in github:

https://github.com/artefactual/archivematica-storage-service/tree/dev/issue-8895-search-api

Hope this helps,

Cheers,

Sarah

Sarah Romkey, MAS,MLIS

Systems Archivist
Artefactual Systems
604-527-2056

@archivematica / @accesstomemory

On Wed, Nov 18, 2015 at 7:05 AM, STI <s...@in.cscm-lx.pt> wrote:

Hello, I'm currently studying the possibilities of integrating Archivematica with our custom framework, so we can access archived files and automatize the process of transfer and ingest. I was wondering if Archivematica possess such API or functionality, so we can know the file's uuid after archived and the final path of the file for displaying it. And can you guys indicate me to something that further elaborates the service callbacks? Is that what I'm looking for? Thank you.

--
You received this message because you are subscribed to the Google Groups "archivematica" group.
To unsubscribe from this group and stop receiving emails from it, send an email to archivematic...@googlegroups.com.
To post to this group, send email to archiv...@googlegroups.com.
Visit this group at http://groups.google.com/group/archivematica.
For more options, visit https://groups.google.com/d/optout.

[STI]

Hello,

Thank you for the answer Sarah, I'll be looking into it. 

I just didn't understand one thing, in order for me as a framework, to be able to use the API to get the record's path, and point to wherever my file is archived (storage final path), need to know what uuid Archivematica has assigned to the record. 

There's any way to know this uuid automaticaly after Archivematica has done assigning it? Or the only way is to manually, as a person, go and check it?

Thank you again.

[Sarah Romkey]

Hello,

I just realized that a direct link to the preliminary API documentation may have been helpful: https://github.com/artefactual/archivematica-storage-service/blob/dev/issue-8895-search-api/docs/search.rst

To search for a specific package, yes, you would need the UUID which is hard to determine without either checking Archivematica manually or doing a get request for all packages of a certain type in that space, etc. You can however search for files using the filename. I'm thinking it would be helpful to be able to search for packages using a part of the package name (e.g. not the UUID part). Do you agree?

Cheers,

Sarah

Sarah Romkey, MAS,MLIS

Archivematica Program Manager
Artefactual Systems
604-527-2056

@archivematica / @accesstomemory

[Matthew Gettemy]

Sorry if I missed something, but I am curious if there are plans to make baggit metadata searchable through an API - similar to using the "Transfer Metadata (other)" key/value option.

[STI]

Hello Sarah,

The API documentation you've just sent me, seen to be a little off, cause when I use the addresses given, for example:

http://<storage service URL>/api/v2/search/file/

 I get a 404 error - page not found . I've started experimenting and got more lucky with the following:

http://<storage service URL>/api/v2/file/?format=json

But still, when I search for a partial filename, it always returns me everything, no matter what name I give, instead of a filtered result:

http://<storage service URL>/api/v2/file/?name=test2&format=json

Does it allows me to search for the filename this way? I only need to get the info of the single file I'm searching for, because on hundreds of files archived, I can't have the whole list returned to the framework, it's not cost effective.

Thank you

[Sarah Romkey]

Hi,

Just a reminder that this is a feature still under development so at this point your mileage may vary! Did you install from the branch I first linked to? If not I wouldn't expect it to work at all. We're still in the testing and QA phase.

Matthew, to answer your question about the bagit metadata, that's not currently part of the requirements.

Cheers,

Sarah

Sarah Romkey, MAS,MLIS

Archivematica Program Manager
Artefactual Systems
604-527-2056

@archivematica / @accesstomemory