What this category's data looks like
In the business of regular higher education institution admissions and enrollment filing, enrollment plans, filing rules, and admission rosters constitute the core materials. Enrollment plans include institution codes, major groups, planned numbers, selected subjects, batches, and dates; filing rules include filing ratios, ranking scores, tie-breaking handling, withdrawal conditions, and signature fields; admission rosters include candidate numbers, names, scores, admitted majors, receipts, and seals. These materials are updated by batch and year, with many fields and rapid version iterations. It is common for the same record to have inconsistent names, quantities, or dates across two materials, unsynchronized versions, or missing signatures or receipts. When materials enter the parsing and chunking function, each assumes the role of a basis source: enrollment plans and filing rules provide rule entries, while admission rosters provide result entries. All must be stored as basis entries that can be traced back to the original text, for answering parents or writing into proposals.
Configuration and implementation checks
This step is executed by parsing and training tasks (server-side tasks), outputting data rows and index statuses (parsed, indexing, indexed). Basis entries that can be traced back to the original text are completed by retrieval and subsequent processing. After parsing is completed, parsed chunks are immediately stored as visible data rows and bound to a unique data identifier. The index transitions through three states: parsed, indexing, and indexed. Data pending indexing does not participate in recall.
| Object | Check | Verification action |
|---|---|---|
| Enrollment plan | Fields include institution codes, major groups, planned numbers, selected subjects, batches, and dates, and must be aligned item by item with filing rules and admission rosters | Verify file records and data rows separately, confirm that planned numbers, dates, and signature fields are consistent, and check whether chunkSize and chunkSplitter are set according to the table structure |
| Filing rules | Fields include filing ratios, ranking scores, tie-breaking handling, withdrawal conditions, and signatures, and must be consistent with the current version | Verify whether rule entries can be traced back to the original text, check version dates and signature receipts, and confirm that trainingType and the index statuses parsed, indexing, and indexed transition normally |
| Admission roster | Fields include candidate numbers, names, scores, admitted majors, receipts, and seals, and must match the names, quantities, and dates in the enrollment plan | Verify roster and plan names, quantities, and dates item by item, check for missing signatures or receipt items, and confirm that data row statuses are consistent with the interface display |
| Data row | Is in one of the three states parsed, indexing, or indexed and is consistent with the interface display; data pending indexing does not participate in recall | Open the data row to verify its status, confirm that the unique data identifier is distinguishable from file records, and that rebuild tasks only cover indexed data |
Three common errors
- Symptom: Newly uploaded materials cannot be found. Cause: The data row is still in parsed or indexing and has not completed indexing; data pending indexing is naturally isolated at the retrieval layer and is not recalled.
- Symptom: The same record has inconsistent names, quantities, or dates in the enrollment plan and admission roster. Cause: The two materials' versions are not synchronized, or field boundaries are misaligned during parsing and chunking, and
chunkSizeandchunkSplitterare not set according to the table structure. - Symptom: Empty data rows remain after a training failure. Cause: Training interruption leaves orphan data, and the failed data is not directly edited or deleted, causing deviations in subsequent retrieval and rebuild scope.
How to verify the setup
Acceptance for this function: 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, whether the rebuild task only covers indexed data, and whether the initial training covers data pending indexing and data being indexed.
Business test: Ask a cross-material basis question and see whether the answer provides a traceable source. For example, ask whether the planned number and admitted number for the same major in the enrollment plan and admission roster match, and check whether the answer can trace back to the original text entries, whether the version is current, and whether signatures and receipts are complete.
- Open the data row to verify its status, confirm it is consistent with the interface display, and separately verify the identity identifiers of file records and data rows.
- View the data rows covered by the rebuild task, confirm that only indexed data is covered, and that data pending indexing does not participate in the rebuild chain.