What this category's data looks like
During the organization of vocational college skills competitions, three types of materials are the core basis: competition regulations, registration forms, and training plans. Competition regulations come from the organizer and are updated annually by competition event; their fields include event name, group, date, and signature field. Registration forms are submitted by participating institutions and are updated by registration batch; their fields include contestant name, ID number, competition event, quantity, date, and receipt. Training plans are prepared by instructors and are updated by training cycle; their fields include training stage, number of class hours, start and end dates, and signature. After these three types of materials enter the document parsing and chunking function, each is stored as an independent file record. Once parsing and chunking are complete, a visible data row is immediately generated and bound to a unique data identifier, providing the data foundation for subsequent retrieval and discrepancy localization.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), outputting data rows and index status (parsed, indexing, indexed). The "discrepancy list" is then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Competition regulations | The event name, group, date, and signature field must be consistent with the corresponding fields in the registration form and training plan | Verify the file record and data row separately, check the chunkSize and chunkSplitter settings, and then confirm that the data row status is parsed or indexed |
| Registration form | Contestant name, competition event, quantity, date, and receipt must be consistent with the competition regulations and training plan | Check whether the data row status has transitioned to indexed, and check whether the processing mode corresponding to trainingType covers this material |
| Training plan | Training stage, number of class hours, start and end dates, and signature must be consistent with the competition regulations and registration form | Check whether the data rows covered by the rebuild task include only the indexed status, and confirm that data pending indexing does not participate in the rebuild |
| Data row status | Is in one of the three states parsed, indexing, indexed and is consistent with the interface display | Open the data row to verify the status, and confirm that data pending indexing does not participate in recall |
Three common errors
- Symptom: The quantity and date for the same contestant do not match between the registration form and the training plan. Cause: The versions of the two materials were not synchronized; after parsing and chunking, they were stored separately, and the discrepancy was not localized.
- Symptom: Competition regulations missing a signature or receipt are still treated as valid materials. Cause: During material verification, the signature field and receipt field were not checked item by item.
- Symptom: Newly uploaded materials cannot be found. Cause: The data row is in parsed or indexing status, and data pending indexing is naturally isolated at the retrieval layer and is not recalled.
How to verify the setup
This feature's acceptance: Upload one material and observe the data row status transitions and the coverage scope of the rebuild task. Confirm that after parsing and chunking are complete, it is immediately stored as a visible data row and bound to a unique data identifier, with the status passing through parsed, indexing, and indexed in sequence; when changing models, only indexed data triggers a rebuild, and data pending indexing does not participate in the rebuild chain. You can use chunkSize, chunkSplitter, and trainingType to verify the chunking and training mode.
Business test: Use one real record to find fields inconsistent with the ledger. Take one registration record, separately verify the name, quantity, date, and signature in the competition regulations, registration form, and training plan, localize the discrepant fields and form a discrepancy list, and after confirmation by the reviewer, write it back.
- Open the data row to verify the status, and confirm that it is in one of the three states parsed, indexing, indexed and is consistent with the interface display
- Verify the file record and data row separately, and confirm that the identities of the file record and data row are kept distinct