Files
ragflow/internal/deepdoc/parser/pdf/pagespdfiumtest/pages_pdfium_e2e_test.go
Jack d12fd3b79d feat: parser pages range and parse type validation for dataset/document (#17293)
## Summary

Adds page-range parsing support to the Go-native pipeline path and
introduces strict `parse_type` validation for both dataset and document
update endpoints.

## What changed

### Pages range parsing
- **`internal/utility/pdf_pages.go`** — `NormalizePDFPages`: normalizes
raw page ranges (list of `[from,to]` 1-indexed inclusive ranges) into
sorted, merged, deduplicated `[][]int`. Invalid ranges are dropped.
- **`internal/ingestion/pipeline/pdf_pages.go`** —
`NormalizeParserConfigPages`: walks any parser_config map and normalizes
`"pages"` values under every component → filetype setup, so the
persisted config always carries clean, merged ranges.
- **`internal/deepdoc/parser/pdf/parser.go`** — integrates
`resolvePagesToProcess` to filter parsed PDF pages by the configured
ranges.
- Pipeline integration (parser pages):
`internal/parser/parser/pdf_parser_common.go`, `chunk_process.go`, plus
associated e2e and unit tests.

### Parse type validation (shared logic)
- **`internal/service/parser_mode.go`** (new) — `ValidateParseTypeMode`:
shared function that validates `parse_type` (1=BuiltIn/parser_id,
2=Pipeline/pipeline_id) and ensures the corresponding field is present.
Used by both dataset and document update endpoints.
- **`internal/service/dataset/crud.go`** / `update.go` — replaces inline
`isPipelineMode`/`isBuiltinMode` computation with the shared
`service.ValidateParseTypeMode`.
- **`internal/service/document/document_dataset_update.go`** — adds
strict `parse_type` validation in `validateDatasetDocumentUpdate`,
simplifies the reparse logic to a two-way switch (isBuiltin/isPipeline)
now that parse_type is always valid.
- **`internal/service/document/document.go`** — adds `ParseType` field
to `UpdateDatasetDocumentRequest`.
- **`internal/service/document/document_dataset_update.go`** —
`updateDocumentParserConfig` fallback path when DSL loading fails.
- **`internal/service/parser_mode_test.go`** (new) — test coverage for
nil, invalid, and missing-field scenarios.

### Frontend
- **`web/src/interfaces/request/document.ts`** — adds `parseType` to
`IChangeParserRequestBody`.
- **`web/src/hooks/use-document-request.ts`** —
`useSetDocumentPipelineParser` sends `parse_type` in the PATCH payload.
- **`web/src/pages/dataset/dataset/use-change-document-parser.ts`** —
Go/Python branching for the document parser config dialog.
-
**`web/src/components/document-pipeline-dialog/use-document-pipeline-form.ts`**
— `buildSubmitData` returns `parseType` (bugfix: was dropped from the
return value).

### Test changes
- **Removed**: 2 tests that verified the old "mutually exclusive" error
(replaced by `ValidateParseTypeMode` coverage).
- **Modified**: 6 tests across document and dataset packages to include
`ParseType` in request structs.
- **Added**: new e2e tests for pages parsing (`pages_e2e_test.go`,
`pdf_parser_pages_e2e_test.go`) and unit tests for `NormalizePDFPages`,
`NormalizeParserConfigPages`, `resolvePagesToProcess`.

## Backward compatibility
- The `parse_type` field is **required** when `parser_id` or
`pipeline_id` is sent. This changes the contract for both dataset and
document PATCH endpoints, but aligns the Go backend with the existing
frontend behavior (the frontend already sends `parse_type`). Callers
that omit `parse_type` when updating parser/pipeline selections will
receive a clear error message.
- Existing callers that only update fields like `name`, `enabled`, or
`meta_fields` are unaffected.
- Test updates ensure all known call sites are compliant.
2026-07-23 19:57:27 +08:00

133 lines
4.2 KiB
Go

//go:build cgo && manual
// Package pagespdfiumtest exercises Config.Pages against a real PDF via the
// pdfium CGO engine. It lives in a separate subpackage so that it does not
// share the build of the parent package's manual-tag test files (some of
// which do not currently compile on main).
package pagespdfiumtest
import (
"context"
"image"
"os"
"path/filepath"
"reflect"
"testing"
deepdocpdf "ragflow/internal/deepdoc/parser/pdf"
deepdoctype "ragflow/internal/deepdoc/parser/pdf/type"
)
// noopDocAnalyzer reports unhealthy and returns empty results, forcing the
// parser onto the charsToBoxes path so no DeepDoc model service is required.
type noopDocAnalyzer struct{}
func (noopDocAnalyzer) DLA(context.Context, image.Image) ([]deepdoctype.DLARegion, error) {
return nil, nil
}
func (noopDocAnalyzer) TSR(context.Context, image.Image) ([]deepdoctype.TSRCell, error) {
return nil, nil
}
func (noopDocAnalyzer) OCRDetect(context.Context, image.Image) ([]deepdoctype.OCRBox, error) {
return nil, nil
}
func (noopDocAnalyzer) OCRRecognize(context.Context, image.Image) ([]deepdoctype.OCRText, error) {
return nil, nil
}
func (noopDocAnalyzer) Health() bool { return false }
// pageKeys extracts the set of processed page numbers from PageHeight.
func pageKeys(m map[int]float64) map[int]struct{} {
out := make(map[int]struct{}, len(m))
for k := range m {
out[k] = struct{}{}
}
return out
}
// TestPagesRealPdf_PdfiumFilter verifies Config.Pages end-to-end against a
// real multi-page PDF through the pdfium CGO engine (NewEngine ->
// PageCount/ExtractChars/RenderPage call the pdfium C library). DeepDoc
// DLA/TSR/OCR are stubbed by noopDocAnalyzer.
//
// Run with:
//
// ./build.sh --test -tags manual -v -run TestPagesRealPdf_PdfiumFilter ./internal/deepdoc/parser/pdf/pagespdfiumtest/
func TestPagesRealPdf_PdfiumFilter(t *testing.T) {
t.Setenv("BATCH_SKIP_DEEPDOC", "1")
pdfPath := filepath.Join("..", "testdata", "pdfs", "03_multipage.pdf")
data, err := os.ReadFile(pdfPath)
if err != nil {
t.Fatalf("read %s: %v", pdfPath, err)
}
// NewEngine calls into pdfium (CGO). Failure here means the real C path
// was reached but pdfium could not open the document.
eng, err := deepdocpdf.NewEngine(data)
if err != nil {
t.Fatalf("NewEngine (pdfium CGO): %v", err)
}
defer eng.Close()
total, err := eng.PageCount()
if err != nil {
t.Fatalf("eng.PageCount: %v", err)
}
t.Logf("03_multipage.pdf: pdfium PageCount() = %d (real CGO call)", total)
if total < 2 {
t.Skipf("need >=2 pages to verify filtering, got %d", total)
}
mock := noopDocAnalyzer{}
// Group 1: restrict to page 1 (1-indexed) -> only 0-based page 0.
t.Run("filter to page 1 only", func(t *testing.T) {
cfg := deepdoctype.DefaultParserConfig()
cfg.Pages = [][]int{{1, 1}}
p := deepdocpdf.NewParser(cfg)
result, err := p.ParseRaw(context.Background(), eng, mock)
if err != nil {
t.Fatalf("ParseRaw: %v", err)
}
if len(result.PageHeight) != 1 {
t.Errorf("expected 1 page parsed, got %d (keys=%v)",
len(result.PageHeight), pageKeys(result.PageHeight))
}
if _, ok := result.PageHeight[0]; !ok {
t.Errorf("expected page 0 in PageHeight, got keys %v",
pageKeys(result.PageHeight))
}
})
// Group 2: control — no Pages restriction -> all pages.
t.Run("no pages -> all pages (control)", func(t *testing.T) {
p := deepdocpdf.NewParser(deepdoctype.DefaultParserConfig())
result, err := p.ParseRaw(context.Background(), eng, mock)
if err != nil {
t.Fatalf("ParseRaw: %v", err)
}
if len(result.PageHeight) != total {
t.Errorf("expected %d pages (all), got %d (keys=%v)",
total, len(result.PageHeight), pageKeys(result.PageHeight))
}
})
// Group 3: multi-range — first and last page only.
if total >= 3 {
t.Run("multi-range first and last page", func(t *testing.T) {
cfg := deepdoctype.DefaultParserConfig()
cfg.Pages = [][]int{{1, 1}, {total, total}}
p := deepdocpdf.NewParser(cfg)
result, err := p.ParseRaw(context.Background(), eng, mock)
if err != nil {
t.Fatalf("ParseRaw: %v", err)
}
want := map[int]struct{}{0: {}, total - 1: {}}
if got := pageKeys(result.PageHeight); !reflect.DeepEqual(got, want) {
t.Errorf("PageHeight keys = %v, want %v", got, want)
}
})
}
}