What this category's data looks like
Degree conferral and certificate information reporting involves three types of materials: degree application forms, thesis review forms, and defense resolutions. The sources are training units and degree management departments, and the update cadence advances round by round with batch applications, review and defense, and resolution archiving. When the three types of materials enter the document parsing and chunking stage, their roles differ: the degree application form carries personal basic information and course grade fields, the thesis review form carries review opinions, scores, and reviewer signatures, and the defense resolution carries the defense conclusion, voting results, and chair signature. Field characteristics are concentrated in name, student ID, major, date, quantity, and signature fields. The same record often appears repeatedly in two materials, so after parsing and chunking, fields and signature consistency must be checked by data row, providing a comparable material draft for the version ledger and change records.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), which outputs data rows and index status (parsed, indexing, indexed). The version ledger and change records are then completed by retrieval and subsequent processing. Successful parsing does not mean retrieval will necessarily hit; the index status also matters.
| Object | Check | Verification action |
|---|---|---|
| Degree application form | Fields, dates, and signatures must correspond item by item with the defense resolution; the name, quantity, and date of the same record must not be inconsistent in two places | Open the data rows of the two materials separately, check whether the name, student ID, major, date, and signature fields are consistent, and check whether the fields fall in the same data row after splitting by chunkSize and chunkSplitter |
| Thesis review form | Review opinions, scores, and reviewer signatures must be complete; missing signatures or receipts are regarded as incomplete materials | Check the reviewer signature and receipt fields, confirm that the material's data row status is parsed or indexed, and check whether the fields are correctly parsed under the processing mode corresponding to trainingType |
| Defense resolution | The defense conclusion, voting results, and chair signature must be synchronized with the date of the degree application form; if the version is not synchronized, it must not enter reporting | Check the resolution date and the application form date, confirm that old-version data rows are traceable and the current version is unique, and that the rebuild task only covers data rows whose indexing has been completed |
| Data rows and file records | The identities of file records and data rows are distinguished separately; data rows are in one of the three states and consistent with the interface display | Check the file record and data row identifiers separately, confirm that the data row status is parsed, indexing, or indexed, and that data pending indexing does not participate in recall |
Three common errors
- Symptom: the name and date of the same record do not match between the degree application form and the defense resolution. Cause: after the two materials are parsed and chunked separately, fields are not checked item by item by data row, and the versions are not synchronized.
- Symptom: the thesis review form lacks a reviewer signature or receipt, and the material is returned. Cause: parsing and chunking only stores visible data rows, signature and receipt fields are not included in the verification action, and missing items are not discovered.
- Symptom: a newly uploaded material cannot be found, and rebuilding after changing the model takes abnormally long. Cause: data pending indexing does not participate in recall, rebuilding only targets data whose indexing has been completed, and the scope of initial training and rebuilding has been swapped.
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. Check whether data rows transition through the three states parsed, indexing, and indexed and are consistent with the interface display, confirm the actual values of chunkSize, chunkSplitter, and trainingType in parsing/chunking and the training task, and that the rebuild task only covers data rows whose indexing has been completed.
Business test: change one piece of content and see whether the version number and references are updated synchronously.
- Open the data row to check the status, and confirm that it is in one of the three states parsed, indexing, or indexed, and consistent with the interface display.
- Check the file record and data row identifiers separately, and confirm that the identities of file records and data rows are distinguished separately.