Once provisioned, your tenant has

How your collections are stored

Collections land in your tenant's Azure Blob storage. One container is created per BLOB location (named with that location's id), and items are organized under content-type prefix folders. Exchange content is grouped into fragments — 24-hour windows of a single owner's items — while OneDrive files are stored one blob per file.

# A collection writes to one Azure Blob container per BLOB location — the container is named
# with the location's id. Inside, items are grouped under content-type prefix "folders":

<collection-location-id>/                         # one container per BLOB ("Backup") location
├── files/<itemId>                            # OneDrive — one blob per file (not fragmented)
├── email/<owner>_<startUnix>-<endUnix>       # Exchange mail — one blob per 24h fragment
├── calendars/<owner>_<startUnix>-<endUnix>   # Exchange calendar fragments
├── tasks/<owner>_<startUnix>-<endUnix>       # Exchange tasks fragments
└── notes/<owner>_<startUnix>-<endUnix>       # Exchange notes fragments

# A "fragment" is a 24-hour window of one owner's items, stored as a single blob holding a
# serialized array of those items plus per-item component metadata (content hash + version).
# That is why retrieving one email needs both its ItemId and the FragmentId it lives in.

Browsing locations, users & collection data

The Discovery API lets you walk the hierarchy: locations → users / shared mailboxes → items → a single downloadable message. Listing email items requires a date range, and each item comes back with the FragmentId you need to download it. Discovery pages with PageSize + Marker (capped at 100) and responses are PascalCase.

# 0 · Your key goes in the X-API-KEY header on every call. Set a couple of shell vars first:
API_KEY="<your-tenant-api-key>"
TENANT_ID="<your-tenant-id>"
BASE="https://datatap-dev-api.azurewebsites.net"

# 1 · List the locations seeded for your tenant (Exchange Source, OneDrive Source, Backup).
#     Use a source location's id below as LOCATION_ID.
curl -H "X-API-KEY: $API_KEY" "$BASE/api/$TENANT_ID/locations"

# 2 · Browse users *and* shared mailboxes in that source location. Drop filter.type to list
#     regular user mailboxes only. Pick a mailbox/user id from the response to use as SOURCE_ID.
curl -H "X-API-KEY: $API_KEY" \
  "$BASE/api/$TENANT_ID/discovery/$LOCATION_ID/users?filter.type=USERS_AND_SHAREDMAILBOXES&PageSize=50"

# 3 · List collected items under a node of that mailbox/drive (SOURCE_ID). For email you must
#     pass a date range; "-" is the root node. Each MessageItem carries an ItemId *and* a FragmentId.
curl -H "X-API-KEY: $API_KEY" \
  "$BASE/api/$TENANT_ID/discovery/$LOCATION_ID/source/$SOURCE_ID/items/-?types=MessageItem&fromDate=2024-01-01T00:00:00Z&toDate=2024-03-31T23:59:59Z&PageSize=50"

# 4 · Download a single message as .eml — emails need both the ItemId and its FragmentId.
curl -H "X-API-KEY: $API_KEY" -o message.eml \
  "$BASE/api/$TENANT_ID/discovery/$LOCATION_ID/source/$SOURCE_ID/items/$ITEM_ID/download?dataType=email&fragmentId=$FRAGMENT_ID&format=eml"

Creating an Exchange or OneDrive collection job

Your trial ships with the two collection jobs above already wired (and disabled). Enable one, or create your own by pairing a source location with the Backup destination. The full lifecycle — schedules, monitoring and restore — is covered in the workflows guide.

# The trial already seeds two BACKUP jobs (Exchange→Blob, OneDrive→Blob), created *disabled*.
# Enable a seeded job, or create your own. SOURCE_ID = the Exchange or OneDrive source location;
# DEST_ID = the "Backup" (BLOB) location — read both ids from GET /api/$TENANT_ID/locations.
curl -X POST -H "X-API-KEY: $API_KEY" -H "Content-Type: application/json" \
  "$BASE/api/$TENANT_ID/jobs" -d '{
    "name": "Exchange Backup",
    "type": "BACKUP",
    "priority": "Medium",
    "schedule": { "type": "NOW" },
    "sourceId": "$SOURCE_ID",
    "destinationId": "$DEST_ID",
    "indexFiles": false
  }'
# For a OneDrive collection, send the same body with the OneDrive source location as sourceId.

# Start it and track the async task (STARTING -> RUNNING -> READY | FAILED):
curl -X POST -H "X-API-KEY: $API_KEY" "$BASE/api/$TENANT_ID/jobs/$JOB_ID/start"
curl -H "X-API-KEY: $API_KEY" "$BASE/api/$TENANT_ID/jobs/$JOB_ID/status/$TASK_ID"