### Summary
Aligns Go and Python error codes/messages so both backends honor the
same RESTful API contract, removing implementation-specific error leaks
(MySQL errors, `ValueError`, `AttributeError`, Gin validator format) in
favor of clean business error codes.
**Chat list** — invalid `orderby` now returns code 101 (was: raw Python
`AttributeError` code 100); invalid `page`/`page_size` values fall back
to defaults (was: raw `ValueError`/`ProgrammingError` code 100).
**Dataset create/update/delete** — adds UUID validation (101),
extra-field rejection (101), duplicate-id detection (101), content-type
/ JSON-syntax / object-shape checks (101), and "lacks permission" for
nonexistent datasets (IDOR). Create auto-deduplicates dataset names.
Pagerank updates tolerate a missing ES index. List response includes
`parser_config` and `pagerank`.
**Session list/update** — adds filtering, sorting, and pagination
support. Empty payloads are valid no-ops. Authorization errors map to
code 109.
**Chunk list** — doc object uses Python key names (`chunk_count`,
`dataset_id`, `chunk_method`, run text status). Add validates list
element types.
**Document update** — adds `chunk_method` alias, pydantic-style Field
error messages, metadata index auto-create with refresh, and "These
documents do not belong to dataset" messages. List validates
`metadata_condition` and reports ownership errors for unmatched name/id
filters.
**Search completion** — `kb_ids` ownership failure returns code 102
instead of 109.
Released 31 contract tests from `GO_ONLY_SKIPS` (all verified passing on
both Go and Python backends with real LLM keys).
## Summary
Align the Go implementations of these APIs with the Python behavior:
- `POST /api/v1/datasets/:dataset_id/metadata/update`
- `PATCH /api/v1/datasets/:dataset_id/documents/metadatas`
- `POST /api/v1/documents/upload`
## What changed
- Added the Go routes and handlers for the 3 APIs.
- Aligned batch document metadata updates with Python semantics:
- support `match` in update items
- support list append / replace behavior
- support deleting specific list values
- remove metadata entirely when it becomes empty
- create metadata for documents that previously had none when updates
apply
- count `updated` only when a document actually changes
- Aligned `documents/upload` file uploads with Python-style
`upload_info` behavior:
- store upload-info blobs in the per-user downloads bucket
- return lightweight upload descriptors instead of normal
file-management responses
- Improved URL upload behavior:
- SSRF-guarded fetch with redirect validation
- redirect limit aligned to Python behavior
- normalize filename and MIME type
- add `.pdf` when the fetched content is PDF
- normalize HTML content into readable text instead of storing raw HTML
shells
## Validation
### Unit tests
Passed:
- `go test ./internal/service`
- `go test ./internal/handler`
Also verified targeted cases for:
- batch metadata update semantics
- upload_info URL handling
- upload_info download bucket behavior
### curl checks
Verified the new Go endpoints with `curl` and compared the response
shape and behavior with Python for:
- `POST /api/v1/datasets/{dataset_id}/metadata/update`
- `PATCH /api/v1/datasets/{dataset_id}/documents/metadatas`
- `POST /api/v1/documents/upload`
The Go responses were checked against Python for:
- argument validation
- success response shape
- metadata update results
- upload_info result structure
- file vs URL input handling
### What problem does this PR solve?
Implement OpenAI chat completions in GO
POST /api/v1/openai/<chat_id>/chat/completions
OpenAI chat cli: internal/development.md
### Type of change
- [x] Refactoring
## Summary
Add `EnrichChunksWithDocMetadata` as a method on `MetadataService` that
attaches document metadata to retrieval chunks in-place. Equivalent to
Python's `enrich_chunks_with_document_metadata()` from
`api/utils/reference_metadata_utils.py`.
### Usage
```go
metadataSvc.EnrichChunksWithDocMetadata(chunks, tenantID, metadataFields)
```
### Changes
- **`service/metadata.go`**: Added `EnrichChunksWithDocMetadata` method
- **`service/enrich_metadata_test.go`** (new): 7 test cases
### Algorithm
1. Collect unique `(kb_id, doc_id)` pairs from chunks
2. Fetch metadata from ES via `SearchMetadata(kbID, tenantID, docIDs)`
3. Attach `document_metadata` field to each matching chunk
4. Optionally filter to specified `metadataFields`
### Testing
All 7 tests pass:
```
=== RUN TestEnrichChunksWithDocMetadata_NoChunks --- PASS
=== RUN TestEnrichChunksWithDocMetadata_EmptyChunks --- PASS
=== RUN TestEnrichChunksWithDocMetadata_EmptyDocID --- PASS
=== RUN TestEnrichChunksWithDocMetadata_DuplicateDocIDs --- PASS
=== RUN TestEnrichChunksWithDocMetadata_MultipleKBs --- PASS
=== RUN TestEnrichChunksWithDocMetadata_WithMetadataFields --- PASS
=== RUN TestEnrichChunksWithDocMetadata_MixedFields --- PASS
```
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
## Summary
Change `GetFlattedMetaByKBs` return type from `map[string]interface{}`
to strongly-typed `common.MetaData`.
**Depends on**: #15648 (provides `MetaData`, `MetaValueDocs` types)
### Changes
- `service/metadata.go`: Changed return type, removed type assertions
- `service/metadata_filter.go`: Updated all metadata function signatures
- `service/metadata_filter_test.go` (new): 12 test cases
### Bug fix
`applySingleCondition` used `.([]interface{})` assertions on `[]string`
data, silently breaking operators like `!=`, `contains`, `start with`,
etc.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
### What problem does this PR solve?
Add cli
LIST DOCUMENTS OF DATASET quoted_string ";"
LIST METADATA OF DATASETS quoted_string ("," quoted_string)* ";"
LIST METADATA SUMMARY OF DATASET quoted_string (DOCUMENTS quoted_string
("," quoted_string)*)? ";"
### Type of change
- [x] Refactoring
