What the data for this category looks like
This category of data comes from the national petition information system, local petition systems, the 12345 citizen hotline platform, grassroots governance platforms, and historical petition archives. The update frequency is daily. Commonly processed documents include petition registration forms, acceptance notices, transfer and assignment letters, handling opinions, review and re-examination applications, resolution plans, and more. Data fields cover basic petitioner information, request matters, handling process nodes, and policy explanation content. Some documents include scanned attachments or multi-level nested attachment structures.
What constraints these characteristics impose on the "document parsing and chunking" step
The data sources for this category are diverse and have mixed formats, including structured forms, unstructured documents, and scanned attachments. This places requirements on the parsing system's ability to adapt to different formats. The daily updated batch data demand requires the parsing process to have stable concurrent processing capabilities. Documents contain legally required process node content such as petition acceptance, handling, review and re-examination. Chunking must retain process relevance and avoid splitting key handling information. Some transfer documents have inconsistent formats, and historical archives use outdated file formats, so custom parsing rules are needed to adapt to different scenarios.
How to set the configuration
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
chunkSize | 800–1200 characters | Covers complete semantic units for petition-related documents that include requests and handling opinions, avoiding splitting policy explanations or process nodes |
chunkSplitter | Split by paragraph + heading | Petition documents often contain hierarchical handling process headings; splitting by paragraph preserves the integrity of handling logic |
UPLOAD_FILE_MAX_SIZE | 1000 MB | Adapts to batch-imported petition archive packages, preventing large file parsing interruptions |
PARSE_FILE_TIMEOUT_SECONDS | 600 seconds | Reserves sufficient parsing time for complex petition archives containing multi-page scanned items or nested attachments |
customPdfParse | Enable custom parsing | Adapts to the fixed format headers of petition-specific PDFs to extract structured fields |
chunkSettingMode | Manual configuration | Adjust chunking rules manually to fit business requirements for the fixed structure of petition documents |
Three common mistakes
- File upload parsing returns a 413 status code. Cause: The
UPLOAD_FILE_MAX_SIZEconfiguration has not been adjusted, and the file size exceeds the platform's default limit. - Imported CSV-format petition registration forms only display the first two columns of content. Cause: Custom CSV parsing configuration has not been enabled; the platform defaults to reading only the first two column fields.
- No key setting entry is found when configuring an external pdf-maker parser. Cause: The custom parsing module has not been entered, and the key required for the external parser has not been bound in the corresponding configuration item.
How to confirm the configuration is correct
- Upload a single standard petition acceptance notice, check that the parsed text fully covers the notice content and request acceptance scope, and verify that the chunking rules fit business requirements
- Import a batch of CSV-format petition registration forms, confirm that all preset fields are correctly extracted, and verify that the custom CSV parsing configuration is effective
- Configure an external pdf-maker parser and bind the key, upload a test PDF petition document, check that the parsing result has no abnormalities, and verify that the key configuration is correct
- Upload a Feishu document package with multi-level directories, check that all subdirectory content, embedded images, and doc text are covered, and verify that recursive parsing configuration is enabled
The values provided are common starting points and should be measured against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.