Skip to main content
Stream every rejected row from an import job to stdout as CSV. The failedRows field on the job record only carries a small inline sample. This command returns the full list, ready to redirect to a file or pipe into another tool. The output is always CSV. The global --format and --json flags emit a stderr warning and are otherwise ignored.

​
Synopsis

talkvalue path import failed-export <id> > failed.csv

​
Arguments

ArgumentTypeDescription
<id>integerThe importJobId returned by import create.

​
Examples

​
1. Save failures to a file

talkvalue path import failed-export 4218 > failed-4218.csv
Captures the rejected rows. The CSV header is rowNum,errorCode,errorMessage,rawValue so you can sort and filter without re-parsing.

​
2. Count failures grouped by error code

talkvalue path import failed-export 4218 \
  | tail -n +2 \
  | awk -F',' '{print $2}' \
  | sort | uniq -c | sort -rn
Skips the header, pulls errorCode, and prints the rank-ordered failure types. Useful when triaging which fix unlocks the most rows.

​
3. Only run when the job had failures

job_id=4218
failed=$(talkvalue path import get "$job_id" --json | jq -r '.data.failedCount')
if [ "$failed" -gt 0 ]; then
  talkvalue path import failed-export "$job_id" > "failed-$job_id.csv"
fi
Pairs import get with failed-export so the CSV file only appears when there is something to fix.

​
Response

The command writes raw CSV to stdout, not the standard JSON envelope. 
rowNum,errorCode,errorMessage,rawValue
47,INVALID_EMAIL,Email could not be parsed,alice@@acme.com
214,MISSING_REQUIRED,Email column is empty,
If TalkValue returns a non-200 status the command fails with a usage error carrying the HTTP status. No CSV is written.

​
See also