What this category's data looks like
At the signing stage of research project task books, materials come from the project task book, the topic task breakdown table, and the budget book, and the update cadence advances with project approval review, budget adjustments, and version countersigning. Field characteristics center on topic number, undertaking unit, principal investigator, start and end dates, budget subjects, amounts, and signature fields. The task book defines the overall objectives and assessment indicators, the breakdown table breaks the objectives down to topics and milestones, and the budget book provides the funding basis and annual allocation. When these three materials enter document parsing and chunking, they respectively play the roles of overall outline, task mapping, and funding constraints, and after parsing they form data rows and index statuses for retrieval and subsequent processing by the unified terminology table and comparison draft.
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). The "unified terminology table and comparison draft" is then completed by retrieval and subsequent processing. After parsing is completed, parsing and chunking are immediately persisted as visible data rows and bound to a unique data identifier; the index transitions through the three states parsed, indexing, and indexed; data pending indexing does not participate in recall; and changing models only rebuilds data that has already completed indexing.
| Object | Check | Verification action |
|---|---|---|
| Project task book | The topic name, principal investigator, start and end dates, and signature fields are consistent in both materials | Separately verify the topic name, dates, and signatures in the task book and the breakdown table, and confirm that the data row status is parsed / indexing / indexed |
| Topic task breakdown table | The name, quantity, and dates of the same record match the budget book, and versions are synchronized | Trace back by topic number to the name, quantity, and dates in the breakdown table and budget book, and verify the chunking results of chunkSize and chunkSplitter |
| Budget book | Budget subjects, amounts, and annual allocation can be traced back to the original text of the task book, and receipts are complete | Sample three terms to check whether their spelling is consistent throughout the document, and verify trainingType and the data rows covered by the rebuild task |
| Data row | It is in one of the three states and consistent with the interface display, and the file record and the data row identity are distinguishable | Open the data row to verify the status, and confirm that the rebuild only covers data that has already completed indexing |
Three common errors
- Symptom: The same topic name is written differently in the task book and the breakdown table. Cause: versions are not synchronized, the two materials are updated separately, and terminology is not unified.
- Symptom: The amount in the budget book does not match the quantity and dates in the task book. Cause: the field definitions for the same record are inconsistent in the two materials, and receipts are missing.
- Symptom: Materials that were just uploaded cannot be found. Cause: the data row is in a pending-indexing state, and successful parsing does not mean retrieval will necessarily hit.
How to verify the setup
Acceptance for this feature: upload one material, observe the status transitions of the data row and the coverage scope of the rebuild task, and confirm the parsing and training outputs corresponding to chunkSize, chunkSplitter, and trainingType; only parsed / indexing / indexed are recognized as statuses.
Business test: sample three terms to check whether their spelling is consistent throughout the document, and verify whether the unified terminology table and comparison draft can trace back to the original text.
- Open the data row to verify the status, confirm that the identifiers of the file record and the data row are distinguishable, and that the rebuild task only covers data that has already completed indexing.
- Compare the name, quantity, dates, and signature receipts of the task book, breakdown table, and budget book by topic number, and record inconsistent items.