What this category's data looks like
The materials in the initial review stage of academic journal submissions are mainly manuscripts, author information forms, and plagiarism check reports. Sources include exports from the editorial system, email attachments, and scanned documents, and the update cadence changes on a rolling basis with review rounds and returns from revision. Manuscripts carry the main text and figures/tables, author information forms carry authorship, affiliations, funding, and date signatures, and plagiarism check reports carry similarity rates and comparison sources. After these three types of materials enter the document parsing and chunking function, they are first split by content structure into retrievable data rows, and then enter the three-state index flow. In terms of field characteristics, manuscripts focus on titles, sections, figure/table numbers, and paragraph hierarchy; author information forms focus on names, order, affiliations, dates, and signature fields; plagiarism check reports focus on numbers, dates, and comparison entries. The material roles differ, but all must remain version-traceable and citation-synchronizable within the same data row system.
Configuration and implementation checks
Parsing and training tasks (server-side tasks) only output data rows and index states (parsed, indexing, indexed). The version ledger and change records must be handled next 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 flows through the three states parsed, indexing, and indexed. Data pending indexing does not participate in recall. When changing models, only data that has already completed indexing is rebuilt.
| Object | Check | Verification action |
|---|---|---|
| Manuscript | In the version ledger and change records, the title, section number, figure/table count, receipt date, and revision return date of the same manuscript should be consistent across the manuscript, author information form, and plagiarism check report, and the current version should be unique | Open the data row and check whether the state is one of parsed, indexing, or indexed; separately check the file record and the data row identifier; check the chunking and training records corresponding to chunkSize, chunkSplitter, and trainingType |
| Author information form | Author names, order, affiliations, funding information, and signature dates should match the manuscript authorship and the submission information in the plagiarism check report; missing signatures or receipts are regarded as inconsistent | Compare the fields and dates of the author information form with the manuscript and plagiarism check report item by item; check whether the signature fields and receipts are complete; check whether the data row state is consistent with the interface display |
| Plagiarism check report | The plagiarism check number, submission date, comparison sources, and manuscript version should correspond to the same round, and old-version reports must not be mixed with the current manuscript | Check the plagiarism check report date against the manuscript version date; see whether the data rows covered by the rebuild task include only data that has completed indexing; confirm that data pending indexing has not been recalled |
Three common errors
- Symptom: a manuscript that was just uploaded cannot be found in retrieval. Cause: the data row is still in parsed or indexing, and data pending indexing is naturally isolated at the retrieval layer; successful parsing does not equal a retrieval hit.
- Symptom: the name, count, and date of the same record do not match between the manuscript and the author information form. Cause: versions are not synchronized, old-version data rows are still referenced, and references are not updated with the version number.
- Symptom: after a training failure, empty data rows are left behind, or rebuilding after a model change takes abnormally long. Cause: failed data is not edited or deleted in time, and the rebuild scope mistakenly covers data pending indexing.
How to verify the setup
Acceptance for this feature: upload one material, observe the data row state transitions and the coverage scope of the rebuild task, and confirm that the state transitions correctly among parsed, indexing, and indexed, and that rebuilding covers only data that has completed indexing.
Business test: change one piece of content and see whether the version number and references are updated synchronously; check whether the records for chunkSize, chunkSplitter, and trainingType correspond to the data row identifier.
- Open the data row and check whether the state is consistent with the interface display, and separately check the file record and the data row identifier.
- Check whether the signatures and receipts in the author information form are complete, and compare the fields, dates, and counts of the manuscript, author information form, and plagiarism check report.