# Central API submissions request attachment filenames and identifiers

**URL:** <https://forum.getodk.org/t/central-api-submissions-request-attachment-filenames-and-identifiers/37306>\
**Category:** Support\
**Tags:** odk-central\
**Created:** [April 9, 2022, 3:13am UTC](https://forum.getodk.org/t/central-api-submissions-request-attachment-filenames-and-identifiers/37306 "2022-04-09T03:13:53Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![akt01](https://getodk.b-cdn.net/letter_avatar_proxy/v4/letter/a/f07891/32.png) [@akt01](https://forum.getodk.org/u/akt01)\
**Post date:** [April 9, 2022, 3:13am UTC](https://forum.getodk.org/t/central-api-submissions-request-attachment-filenames-and-identifiers/37306/1 "2022-04-09T03:13:53Z")

</div>

**1. What is the issue? Please be detailed.**

When a form contains media attachment fields and submission files, an API request for [form submissions](https://odkcentral.docs.apiary.io/#reference/submissions/submissions/exporting-form-submissions-to-csv) return the media files (e.g., image files) inside the /media/ subfolder. The default behavior seems to attach a "-hh\_mm\_ss" timestamp postfix to the original filename, which is also recorded in the rows in the csv file.

I may be missing something, but doesn't this mean there may be filename collisions? Hypothetically if there are 2 image files with identical original filenames uploaded at the exact same time (hour:minute:second), how would it be handled?

A follow-up to the above. Is it possible to include the UUID+fieldID identifier as a prefix (or postfix) of the filename, which will map each uploaded file uniquely? Alternatively (I think is preferable/cleaner), to completely replace the filename with just the UUID+fieldID, and keep the original filename recorded in the CSV table so the information is not lost.

Thank you!