CSV Import¶
Import charging sessions in bulk from CSV files. The import flow is available from the CSV Import tab on the Settings page.
Import Flow¶
The import uses a three-step flow:
- Upload -- Select a CSV file, choose a timezone, and upload
- Preview -- Review parsed rows, fix errors inline, handle duplicates
- Summary -- See counts of added, updated, skipped, and failed rows
CSV Template¶
Download the template CSV from the upload screen. It contains headers for all mappable fields -- fill in what you have and leave the rest blank.
The template includes columns for:
| Field | Example | Required |
|---|---|---|
| session_start_utc | 2025-06-15 14:30:00 |
Soft (flagged if missing) |
| energy_kwh | 32.5 |
Soft (flagged if missing) |
| cost | 8.12 |
No |
| location_name | Home |
No |
| charge_type | AC or DC |
No |
| duration_minutes | 45 |
No |
| soc_start / soc_end | 20 / 80 |
No |
| miles_added | 95.2 |
No |
| network_name | Electrify America |
No |
| evse_voltage | 480 |
No |
| evse_kw | 150 |
No |
The full template has ~24 columns covering all session and EVSE fields.
Auto-Detection¶
You don't have to use the template. The importer auto-detects common column name patterns:
- Exact matches against known column names
- Normalized matching (lowercase, stripped punctuation)
- Keyword hints (e.g., "kwh" maps to energy, "soc" maps to state of charge)
An info banner shows which columns were matched and which were skipped.
Timezone Handling¶
The upload form includes a timezone selector, defaulting to your app timezone setting.
- Naive timestamps (no timezone info in the CSV) are interpreted as the selected timezone and converted to UTC for storage
- Timezone-aware timestamps (with explicit offset or zone) are respected as-is
- All stored timestamps are UTC; display uses your configured timezone
Preview Table¶
The preview shows all parsed rows (no row cap) in a scrollable table with columns for date, location, energy, cost, type, network, duration, and status.
Row Status¶
| Status | Meaning |
|---|---|
| New | No matching session found -- will be imported |
| Duplicate | Matches an existing session |
| Error | Row has a parsing issue |
Inline Editing¶
Click an error or duplicate row to expand an inline editor below it. Edit problematic fields directly -- the row re-verifies automatically when you move focus away from a field (blur triggers server-side validation via HTMX).
Duplicate Handling¶
Duplicate rows offer three actions:
- Skip -- Don't import this row
- Insert anyway -- Import as a new session regardless
- Update existing -- Overwrite the existing session with CSV values
Duplicate detection uses exact match on session_id and fuzzy match on start time (within 1 hour), location, and energy (within 10%).
Import Results¶
After confirming, the summary shows:
- Added -- New sessions inserted
- Updated -- Existing sessions updated
- Skipped -- Rows you deselected or marked as skip
- Failed -- Rows that couldn't be inserted (shown with error details)
Each row is imported independently -- a failed row doesn't affect others.
Large Imports
For initial bulk imports (hundreds of sessions), consider using the seed script instead. It runs server-side with direct database access.