New to Corsair? See API access, authentication, and error handling.
Documents
fetch
documents.fetch
Import a document from a public URL into a parser. Uses parsing credits (one per 5 pages). Parsing takes a few minutes; check documents.status, then read results.get.
Risk: destructive
POST /v2/document/fetch/{parserId} as form data.
Output
reintegrate
documents.reintegrate
Send documents’ parsed data to the parser’s integrations and webhooks again.
Risk: destructive
POST /v1/document/reintegrate/{parserId} with document_ids[].
Output
reparse
documents.reparse
Parse documents again with the parser’s current rules, replacing their parsed data. Uses no extra credits.
Risk: destructive
POST /v1/document/reparse/{parserId} with document_ids[].
Output
status
documents.status
Check whether a document has been imported and parsed (done), and list any failed jobs.
Risk: read
GET /v2/document/status/{parserId}/{documentId}. done is true once import and parsing finished.
Output
status full type
status full type
upload
documents.upload
Upload a base64-encoded document into a parser. Uses parsing credits (one per 5 pages). Parsing takes a few minutes; check documents.status, then read results.get.
Risk: destructive
POST /v1/document/upload/{parserId} with file_content (base64) as form data.
Output
Models
list
models.list
List the model layouts of one parser.
Risk: read
GET /v1/parser/models/{parserId}. The stored id is parser id plus layout id.
Output
models full type
models full type
Parsers
list
parsers.list
List document parsers (id and label).
Risk: read
GET /v1/parsers. Stores id and label.
Output
parsers full type
parsers full type
Ping
ping
ping
Check that the API key is accepted.
Risk: read
GET /v1/ping.
Output
ping full type
ping full type
Results
get
results.get
Get the parsed data of one document.
Risk: read
GET /v1/results/{parserId}/{documentId}. Returns a list with one row (more with includeChildren).
Output
results full type
results full type
list
results.list
Get the parsed data of a parser’s documents, newest first (100 by default). Filter by upload or processing date or remote id. Each row holds the fields the parser extracted.
Risk: read
GET /v1/results/{parserId}. Stores id, document id, file name, page count, and timestamps.
Output
results full type
results full type