What this category's data looks like
Student status change and graduation eligibility review materials come from three types of forms: leave-of-absence/resumption-of-study applications, major transfer applications, and graduation review checklists, containing multilingual names, course translated names, dates, and signature fields. The update cadence follows semester batches, and the field characteristic is that the same record recurs across materials. When entering this feature, each of the three form types is stored as an independent file record, and after parsing and chunking, data rows are generated for subsequent retrieval and reference draft generation.
Configuration and implementation checks
This stage is executed by parsing and training tasks (server-side tasks), outputting data rows and index statuses (parsed, indexing, indexed). The unified terminology table and reference draft are then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Leave-of-absence/resumption-of-study application | Name, student ID, date, and signature field must have the same name and value as in the major transfer application | After upload, check whether the data row status transitions through parsed, indexing, and indexed, and check whether chunkSize chunking preserves the signature field |
| Major transfer application | Course name, credit count, and effective date must be consistent with the graduation review checklist | Separately check the file record and data row identifiers, confirm that trainingType corresponds to the processing mode, and sample three terms to check their spelling throughout the document |
| Graduation review checklist | Graduation conclusion, signature, and receipt field must be complete and version-synchronized | Check the data rows covered by the rebuild task, confirm that only indexed data participates in the rebuild, and that chunkSplitter chunking does not truncate the receipt |
| Unified terminology table and reference draft | The same term is consistent throughout the document, and the translated name can refer back to the original text | Sample three terms to check whether their spelling is consistent throughout the document, and verify that the cited text tokens and array item counts are recorded separately |
Three common errors
- Symptom: newly uploaded materials cannot be found. Cause: the data row is in parsed or indexing status, and data pending indexing does not participate in recall, so it can only be retrieved after indexed.
- Symptom: the name, quantity, and date of the same record do not match across two materials. Cause: versions are not synchronized, and the major transfer application and graduation review checklist are parsed separately without terminology cross-referencing.
- Symptom: empty data rows remain after training fails. Cause: successful parsing does not equal retrieval hits, failed data has not been edited or deleted, and the index status is not completed.
How to verify the setup
This feature's acceptance: upload one material, observe the data row status transitions and the coverage scope of the rebuild task, and confirm that changing models only rebuilds data whose indexing is already completed.
Business test: sample three terms to check whether their spelling is consistent throughout the document, and verify that the translated names can refer back to the original text.
- Open the data row and check whether the status is one of the three states and consistent with the interface display
- Separately check the file record and data row, and confirm that the file identifier and data row identifier are distinguishable