Notes / fixture set 01
Plain files.
Predictable edges.
A few conventions to help you import the fixtures without guessing.
CSV
UTF-8 text, commas between fields, a header on the first row and LF line endings. There is no byte-order mark. Values in this release do not contain embedded delimiters or newlines. Numeric fields contain base-10 integers.
NDJSON
One complete JSON object on each line, encoded as UTF-8. This is a stream of objects, not a single JSON array. Field names match the CSV header. Each file ends with a newline.
Event fields
| Field | Meaning |
|---|---|
event_id | Stable synthetic record identifier; string. |
recorded_at | UTC timestamp in ISO 8601 format. |
region | Synthetic deployment region label; string. |
service | Synthetic service label; string. |
method | Sample HTTP method; string. |
status_code | Sample HTTP response status; integer. |
duration_ms | Synthetic request duration in milliseconds; integer. |
bytes_sent | Synthetic response size in bytes; integer. |
cache_status | One of hit, miss, bypass. |
Small and large files
The CSVs are generated from the same fixed seed. The small CSV is a prefix of the standard CSV; the standard file is a prefix of the large one. All rows are synthetic. Timestamps describe the fixture timeline, not real events.
Download and verify
curl -O https://mertinlow.net/daily/2026-11-06/downloads/events-small-v1.csv sha256sum events-small-v1.csv
Compare the result with the checksum on the download shelf or SHA256SUMS.txt. On macOS, use shasum -a 256.
Resume a transfer
curl -C - -O https://mertinlow.net/daily/2026-11-06/downloads/events-large-v1.csv
The server supports standard HTTP byte ranges. A ranged response uses status 206; a request for the complete file uses 200. Check the full-file hash after resuming.
Archive contents
The ZIP contains the small and standard CSVs, the standard NDJSON, schema-v1.json and README.txt. It does not contain the large CSV.