What this category's data looks like
This page addresses the adult college entrance examination registration and admission management scenario, with materials including registration forms, preferential scoring materials, and admission rosters. Registration forms carry candidates' basic information and application preferences, preferential scoring materials record bonus items and their basis, and admission rosters summarize admission results and signed receipts. These materials enter in batches by admission cycle: centralized submission during the registration stage and gradual archiving during the admission stage. Fields involve name, ID number, application level, bonus item, admitted institution, date, and signature. After materials enter the document parsing and chunking function, each forms data rows. Registration forms and bonus materials are used for qualification verification, while admission rosters are used for result traceability. When the name, quantity, or date of the same record does not match across two materials, the version is not synchronized, or a signature or receipt is missing, these are recurring verification points when maintaining the version ledger and change records.
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 version ledger and change records are then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Registration form | Fields, dates, and signatures must be consistent with the preferential scoring materials, avoiding mismatches in the name, quantity, or date of the same record | Check whether the data row status is indexed, and compare the name, ID number, date, and signature fields between the registration form and the bonus materials |
| Preferential scoring materials | Bonus items and their basis must correspond to the registration form; a missing signature or receipt is deemed inconsistent | Check whether each data row after chunking is bound to a unique data identifier, and confirm that references between the bonus materials and the registration form are synchronized |
| Admission roster | The name, quantity, and date of admission results must match the registration form and bonus materials, and old versions must be traceable | Check the data row status of the admission roster, and see whether the coverage of the rebuild task includes only data with completed indexing |
| Data row | In one of the three states and consistent with the interface display; data pending indexing does not participate in recall | Open the data row to check its status, confirm consistency with the interface display, and verify the chunking results corresponding to trainingType, chunkSize, and chunkSplitter |
Three common errors
- Symptom: The name or date of the same candidate does not match between the registration form and the bonus materials. Cause: The two materials are parsed and stored separately, field comparison is not performed for the same record, and the version ledger does not record the discrepancy.
- Symptom: After the admission roster is updated, the old version is still referenced. Cause: The version is not synchronized; the rebuild covers only data with completed indexing, and data pending indexing is not included in the reference update chain.
- Symptom: Bonus materials missing a signature or receipt still enter the index. Cause: Successful parsing immediately stores them as data rows, and the signature and receipt fields are not treated as mandatory verification items in the version ledger.
How to verify the setup
Acceptance for this feature: upload one material and observe the data row status transitions and the coverage of the rebuild task. The status should be the three states parsed, indexing, and indexed, and the rebuild task should cover only data with completed indexing.
Business test: change one piece of content and see whether the version number and references are updated synchronously. After the change, check the data row status and reference text; record the array length separately, and verify the tokens of the reference text separately.
- Open the data row to check its status, confirm consistency with the interface display, and separately verify the identifiers of the file record and the data row.
- Check the data rows covered by the rebuild task, confirm that only data with completed indexing is covered, and that data pending indexing does not participate in the rebuild chain.