Get an import's status and its row errors
Administrators only. Poll this while an import runs: the counts move as rows are written. `errors` holds the first 100 rows that could not be written (in row order) and `errorCount` the true total. An import in another workspace is a 404. `errors` also holds the rows the plan left out, which the worker saves when it starts (the upload is deleted when the job ends, so this is the only record of which rows and why). Notes, such as a value that was dropped from an issue that was still imported, are not rows left out and are not kept. At most 500 are saved, and when the plan had more, a last entry with code `more_problems` says how many are not listed. `errors=all` returns up to 10,000 instead of 100, for downloading them.
View as MarkdownAuthorization
bearer In: header
Path Parameters
uuidQuery Parameters
all returns every saved row problem (up to 10,000) instead of the first 100.
Value in
- "all"
Response Body
application/json
application/json
application/json
application/json
curl -X GET "https://example.com/v1/imports/497f6eca-6276-4993-bfeb-53cbbbba6f08"{ "import": { "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08", "teamId": "a4ede8ba-7c0a-4485-8763-cbd9b282fbec", "source": "csv", "namespace": "string", "status": "uploaded", "fileName": "string", "fileSize": 0, "totalRows": 0, "createdCount": 0, "skippedCount": 0, "failedCount": 0, "cancelRequested": true, "createdBy": "25a02396-1048-48f9-bf93-102d2fb7895e", "createdAt": "2019-08-24T14:15:22Z", "startedAt": "2019-08-24T14:15:22Z", "finishedAt": "2019-08-24T14:15:22Z", "retryAfter": "2019-08-24T14:15:22Z", "undoneAt": "2019-08-24T14:15:22Z", "undoableUntil": "2019-08-24T14:15:22Z", "commentCount": 0, "mapping": { "v": 1, "columns": { "property1": "string", "property2": "string" }, "idColumn": "string", "idAlias": "string", "people": { "property1": "string", "property2": "string" }, "strict": true, "sprintDates": { "property1": { "start": "2019-08-24", "end": "2019-08-24" }, "property2": { "start": "2019-08-24", "end": "2019-08-24" } }, "linkColumns": [ "string" ], "markup": "none", "commentFormat": "plain", "completedFallback": "string", "dateFormat": "iso", "timezone": "Asia/Kolkata", "labelSeparator": "str", "status": { "property1": { "stateId": "871bb69c-5872-42b2-88b0-72e9da663aaa", "create": true, "type": "backlog" }, "property2": { "stateId": "871bb69c-5872-42b2-88b0-72e9da663aaa", "create": true, "type": "backlog" } }, "priority": { "property1": 0, "property2": 0 } } }, "errors": [ { "row": 0, "field": "string", "code": "string", "message": "string" } ], "errorCount": 0}Upload a file to import
Step 1 of 3 (upload, preview, commit). Administrators only (`org.settings.manage`). multipart/form-data with the CSV in `file` (at most 10 MB and 5,000 rows) and the target `teamId`. The file is read at once, so one that cannot be used (empty, not a CSV, a stray quote, too many rows) is refused with 422 and nothing is stored. UTF-8, UTF-16 and Windows-1252 are recognised, and the separator is detected. Nothing is written to the team. The response carries the file's headers, a few sample rows, and `suggestedMapping`, a starting point guessed from the header names for the caller to review. `namespace` names where the file's row ids come from: rows already imported under the same namespace are recognised on a re-run and skipped. It defaults to one derived from the file name, or to `linear` or `jira` for an export of that tool, so overlapping exports share it. `detected` is set when the file is recognised as another tool's export. `previousImport` is set when this exact file was imported into the team before. Uploads that are never committed are closed after 24 hours and their file deleted. A team holds at most 10 unfinished imports (409).
Say exactly what an import would do
Step 2 of 3. Administrators only. Takes the column mapping and returns the plan: how many rows would be created, skipped (already imported) or refused, which statuses, labels and projects would be created, which people in the file are not members, the problems found, and the first few rows as they would be written. Nothing is written to the team. Can be repeated, and each call replaces the last. `plan.fingerprint` identifies the plan: commit must carry the fingerprint of the most recent preview. A mapping that cannot be used is a 422 listing every problem together. Only an import that has not been started can be previewed (409 otherwise).