Benjamin_Boenisch/breakpilot-lehrer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: independent sessions for page-split + URL-based pipeline navigation

Browse Source 
Page-split now creates independent sessions (no parent_session_id),
parent marked as status='split' and hidden from list. Navigation uses
useSearchParams for URL-based step tracking (browser back/forward works).
page.tsx reduced from 684 to 443 lines via usePipelineNavigation hook.

Box sub-sessions (column detection) remain unchanged.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-03-24 17:05:33 +01:00
parent f34340de9c
commit f931091b57
9 changed files with 368 additions and 391 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
admin-lehrer/app/(admin)/ai/ocr-overlay/page.tsx
View File

@@ -383,7 +383,7 @@ export default function OcrOverlayPage() {
if (mode === 'paddle-direct' || mode === 'kombi') { if (mode === 'paddle-direct' || mode === 'kombi') {
switch (currentStep) { switch (currentStep) {
case 0: case 0:
return <StepOrientation key={sessionId} sessionId={sessionId} onNext={handleOrientationComplete} onSubSessionsCreated={handleBoxSessionsCreated} /> return <StepOrientation key={sessionId} sessionId={sessionId} onNext={handleOrientationComplete} onSessionList={() => { loadSessions(); setSessionId(null) }} />
case 1: case 1:
return <StepDeskew key={sessionId} sessionId={sessionId} onNext={handleNext} /> return <StepDeskew key={sessionId} sessionId={sessionId} onNext={handleNext} />
case 2: case 2:
@@ -421,7 +421,7 @@ export default function OcrOverlayPage() {
} }
switch (currentStep) { switch (currentStep) {
case 0: case 0:
return <StepOrientation key={sessionId} sessionId={sessionId} onNext={handleOrientationComplete} onSubSessionsCreated={handleBoxSessionsCreated} /> return <StepOrientation key={sessionId} sessionId={sessionId} onNext={handleOrientationComplete} onSessionList={() => { loadSessions(); setSessionId(null) }} />
case 1: case 1:
return <StepDeskew key={sessionId} sessionId={sessionId} onNext={handleNext} /> return <StepDeskew key={sessionId} sessionId={sessionId} onNext={handleNext} />
case 2: case 2:

463
admin-lehrer/app/(admin)/ai/ocr-pipeline/page.tsx
View File

@@ -1,6 +1,6 @@
'use client' 'use client'
import { useCallback, useEffect, useState } from 'react' import { Suspense, useCallback, useEffect, useState } from 'react'
import { PagePurpose } from '@/components/common/PagePurpose' import { PagePurpose } from '@/components/common/PagePurpose'
import { PipelineStepper } from '@/components/ocr-pipeline/PipelineStepper' import { PipelineStepper } from '@/components/ocr-pipeline/PipelineStepper'
import { StepOrientation } from '@/components/ocr-pipeline/StepOrientation' import { StepOrientation } from '@/components/ocr-pipeline/StepOrientation'
@@ -14,37 +14,28 @@ import { StepWordRecognition } from '@/components/ocr-pipeline/StepWordRecogniti
import { StepLlmReview } from '@/components/ocr-pipeline/StepLlmReview' import { StepLlmReview } from '@/components/ocr-pipeline/StepLlmReview'
import { StepReconstruction } from '@/components/ocr-pipeline/StepReconstruction' import { StepReconstruction } from '@/components/ocr-pipeline/StepReconstruction'
import { StepGroundTruth } from '@/components/ocr-pipeline/StepGroundTruth' import { StepGroundTruth } from '@/components/ocr-pipeline/StepGroundTruth'
import { BoxSessionTabs } from '@/components/ocr-pipeline/BoxSessionTabs' import { DOCUMENT_CATEGORIES, type SessionListItem, type DocumentTypeResult, type DocumentCategory, type SubSession } from './types'
import { PIPELINE_STEPS, DOCUMENT_CATEGORIES, type PipelineStep, type SessionListItem, type DocumentTypeResult, type DocumentCategory, type SubSession } from './types' import { usePipelineNavigation } from './usePipelineNavigation'
const KLAUSUR_API = '/klausur-api' const KLAUSUR_API = '/klausur-api'
export default function OcrPipelinePage() { const STEP_NAMES: Record<number, string> = {
const [currentStep, setCurrentStep] = useState(0) 1: 'Orientierung', 2: 'Begradigung', 3: 'Entzerrung', 4: 'Zuschneiden',
const [sessionId, setSessionId] = useState<string | null>(null) 5: 'Spalten', 6: 'Zeilen', 7: 'Woerter', 8: 'Struktur',
const [sessionName, setSessionName] = useState<string>('') 9: 'Korrektur', 10: 'Rekonstruktion', 11: 'Validierung',
}
function OcrPipelineContent() {
const nav = usePipelineNavigation()
const [sessions, setSessions] = useState<SessionListItem[]>([]) const [sessions, setSessions] = useState<SessionListItem[]>([])
const [loadingSessions, setLoadingSessions] = useState(true) const [loadingSessions, setLoadingSessions] = useState(true)
const [editingName, setEditingName] = useState<string | null>(null) const [editingName, setEditingName] = useState<string | null>(null)
const [editNameValue, setEditNameValue] = useState('') const [editNameValue, setEditNameValue] = useState('')
const [editingCategory, setEditingCategory] = useState<string | null>(null) const [editingCategory, setEditingCategory] = useState<string | null>(null)
const [docTypeResult, setDocTypeResult] = useState<DocumentTypeResult | null>(null) const [sessionName, setSessionName] = useState('')
const [activeCategory, setActiveCategory] = useState<DocumentCategory | undefined>(undefined) const [activeCategory, setActiveCategory] = useState<DocumentCategory | undefined>(undefined)
const [subSessions, setSubSessions] = useState<SubSession[]>([])
const [parentSessionId, setParentSessionId] = useState<string | null>(null)
const [steps, setSteps] = useState<PipelineStep[]>(
PIPELINE_STEPS.map((s, i) => ({
...s,
status: i === 0 ? 'active' : 'pending',
})),
)
// Load session list on mount const loadSessions = useCallback(async () => {
useEffect(() => {
loadSessions()
}, [])
const loadSessions = async () => {
setLoadingSessions(true) setLoadingSessions(true)
try { try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions`) const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions`)
@@ -57,103 +48,42 @@ export default function OcrPipelinePage() {
} finally { } finally {
setLoadingSessions(false) setLoadingSessions(false)
} }
}
const openSession = useCallback(async (sid: string, keepSubSessions?: boolean) => {
try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sid}`)
if (!res.ok) return
const data = await res.json()
setSessionId(sid)
setSessionName(data.name || data.filename || '')
setActiveCategory(data.document_category || undefined)
// Sub-session handling
if (data.sub_sessions && data.sub_sessions.length > 0) {
setSubSessions(data.sub_sessions)
setParentSessionId(sid)
// Parent has sub-sessions — open the first incomplete one (or most advanced if all done)
const incomplete = data.sub_sessions.find(
(s: SubSession) => !s.current_step || s.current_step < 10,
)
const target = incomplete || [...data.sub_sessions].sort(
(a: SubSession, b: SubSession) => (b.current_step || 0) - (a.current_step || 0),
)[0]
if (target) {
openSession(target.id, true)
return
}
} else if (data.parent_session_id) {
// This is a sub-session — keep parent info but don't reset sub-session list
setParentSessionId(data.parent_session_id)
} else if (!keepSubSessions) {
setSubSessions([])
setParentSessionId(null)
}
// Restore doc type result if available
const savedDocType: DocumentTypeResult | null = data.doc_type_result || null
setDocTypeResult(savedDocType)
// Determine which step to jump to based on current_step
const dbStep = data.current_step || 1
// DB steps: 1=start, 2=orientation, 3=deskew, 4=dewarp, 5=crop, 6=columns, ...
// UI steps are 0-indexed: 0=orientation, 1=deskew, 2=dewarp, 3=crop, 4=columns, ...
let uiStep = Math.max(0, dbStep - 1)
const skipSteps = [...(savedDocType?.skip_steps || [])]
// Sub-session handling depends on how they were created:
// - Crop-based (current_step >= 5): image already cropped, skip all pre-processing
// - Page-split (current_step 2): orientation done on parent, skip only orientation
// - Page-split from original (current_step 1): needs full pipeline
const isSubSession = !!data.parent_session_id
if (isSubSession) {
if (dbStep >= 5) {
// Crop-based sub-sessions: image already cropped
const SUB_SESSION_SKIP = ['orientation', 'deskew', 'dewarp', 'crop']
for (const s of SUB_SESSION_SKIP) {
if (!skipSteps.includes(s)) skipSteps.push(s)
}
if (uiStep < 4) uiStep = 4 // columns step (index 4)
} else if (dbStep >= 2) {
// Page-split sub-session: parent orientation applied, skip only orientation
if (!skipSteps.includes('orientation')) skipSteps.push('orientation')
if (uiStep < 1) uiStep = 1 // advance past skipped orientation to deskew
}
// dbStep === 1: page-split from original image, needs full pipeline
}
setSteps(
PIPELINE_STEPS.map((s, i) => ({
...s,
status: skipSteps.includes(s.id)
? 'skipped'
: i < uiStep ? 'completed' : i === uiStep ? 'active' : 'pending',
})),
)
setCurrentStep(uiStep)
} catch (e) {
console.error('Failed to open session:', e)
}
}, []) }, [])
useEffect(() => { loadSessions() }, [loadSessions])
// Sync session name when nav.sessionId changes
useEffect(() => {
if (!nav.sessionId) {
setSessionName('')
setActiveCategory(undefined)
return
}
const load = async () => {
try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${nav.sessionId}`)
if (!res.ok) return
const data = await res.json()
setSessionName(data.name || data.filename || '')
setActiveCategory(data.document_category || undefined)
} catch { /* ignore */ }
}
load()
}, [nav.sessionId])
const openSession = useCallback((sid: string) => {
nav.goToSession(sid)
}, [nav])
const deleteSession = useCallback(async (sid: string) => { const deleteSession = useCallback(async (sid: string) => {
try { try {
await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sid}`, { method: 'DELETE' }) await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sid}`, { method: 'DELETE' })
setSessions((prev) => prev.filter((s) => s.id !== sid)) setSessions(prev => prev.filter(s => s.id !== sid))
if (sessionId === sid) { if (nav.sessionId === sid) nav.goToSessionList()
setSessionId(null)
setCurrentStep(0)
setDocTypeResult(null)
setSubSessions([])
setParentSessionId(null)
setSteps(PIPELINE_STEPS.map((s, i) => ({ ...s, status: i === 0 ? 'active' : 'pending' })))
}
} catch (e) { } catch (e) {
console.error('Failed to delete session:', e) console.error('Failed to delete session:', e)
} }
}, [sessionId]) }, [nav])
const renameSession = useCallback(async (sid: string, newName: string) => { const renameSession = useCallback(async (sid: string, newName: string) => {
try { try {
@@ -162,13 +92,13 @@ export default function OcrPipelinePage() {
headers: { 'Content-Type': 'application/json' }, headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ name: newName }), body: JSON.stringify({ name: newName }),
}) })
setSessions((prev) => prev.map((s) => (s.id === sid ? { ...s, name: newName } : s))) setSessions(prev => prev.map(s => (s.id === sid ? { ...s, name: newName } : s)))
if (sessionId === sid) setSessionName(newName) if (nav.sessionId === sid) setSessionName(newName)
} catch (e) { } catch (e) {
console.error('Failed to rename session:', e) console.error('Failed to rename session:', e)
} }
setEditingName(null) setEditingName(null)
}, [sessionId]) }, [nav.sessionId])
const updateCategory = useCallback(async (sid: string, category: DocumentCategory) => { const updateCategory = useCallback(async (sid: string, category: DocumentCategory) => {
try { try {
@@ -177,275 +107,107 @@ export default function OcrPipelinePage() {
headers: { 'Content-Type': 'application/json' }, headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ document_category: category }), body: JSON.stringify({ document_category: category }),
}) })
setSessions((prev) => prev.map((s) => (s.id === sid ? { ...s, document_category: category } : s))) setSessions(prev => prev.map(s => (s.id === sid ? { ...s, document_category: category } : s)))
if (sessionId === sid) setActiveCategory(category) if (nav.sessionId === sid) setActiveCategory(category)
} catch (e) { } catch (e) {
console.error('Failed to update category:', e) console.error('Failed to update category:', e)
} }
setEditingCategory(null) setEditingCategory(null)
}, [sessionId]) }, [nav.sessionId])
const deleteAllSessions = useCallback(async () => { const deleteAllSessions = useCallback(async () => {
if (!confirm('Alle Sessions loeschen? Dies kann nicht rueckgaengig gemacht werden.')) return if (!confirm('Alle Sessions loeschen? Dies kann nicht rueckgaengig gemacht werden.')) return
try { try {
await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions`, { method: 'DELETE' }) await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions`, { method: 'DELETE' })
setSessions([]) setSessions([])
setSessionId(null) nav.goToSessionList()
setCurrentStep(0)
setDocTypeResult(null)
setActiveCategory(undefined)
setSubSessions([])
setParentSessionId(null)
setSteps(PIPELINE_STEPS.map((s, i) => ({ ...s, status: i === 0 ? 'active' : 'pending' })))
} catch (e) { } catch (e) {
console.error('Failed to delete all sessions:', e) console.error('Failed to delete all sessions:', e)
} }
}, []) }, [nav])
const handleStepClick = (index: number) => { const handleStepClick = (index: number) => {
if (index <= currentStep || steps[index].status === 'completed') { if (index <= nav.currentStepIndex || nav.steps[index].status === 'completed') {
setCurrentStep(index) nav.goToStep(index)
} }
} }
const goToStep = (step: number) => { // Orientation: after upload, navigate to session at deskew step
setCurrentStep(step) const handleOrientationComplete = useCallback(async (sid: string) => {
setSteps((prev) =>
prev.map((s, i) => ({
...s,
status: i < step ? 'completed' : i === step ? 'active' : 'pending',
})),
)
}
const handleNext = () => {
if (currentStep >= steps.length - 1) {
// Last step completed
if (parentSessionId && sessionId !== parentSessionId) {
// Sub-session completed — mark it and find next incomplete one
const updatedSubs = subSessions.map((s) =>
s.id === sessionId ? { ...s, status: 'completed' as const, current_step: 10 } : s,
)
setSubSessions(updatedSubs)
// Find next incomplete sub-session
const nextIncomplete = updatedSubs.find(
(s) => s.id !== sessionId && (!s.current_step || s.current_step < 10),
)
if (nextIncomplete) {
// Open next incomplete sub-session
openSession(nextIncomplete.id, true)
} else {
// All sub-sessions done — return to session list
setSteps(PIPELINE_STEPS.map((s, i) => ({ ...s, status: i === 0 ? 'active' : 'pending' })))
setCurrentStep(0)
setSessionId(null)
setSubSessions([])
setParentSessionId(null)
loadSessions()
}
return
}
// Main session: return to session list
setSteps(PIPELINE_STEPS.map((s, i) => ({ ...s, status: i === 0 ? 'active' : 'pending' })))
setCurrentStep(0)
setSessionId(null)
setSubSessions([])
setParentSessionId(null)
loadSessions()
return
}
// Find the next non-skipped step
const skipSteps = docTypeResult?.skip_steps || []
let nextStep = currentStep + 1
while (nextStep < steps.length && skipSteps.includes(PIPELINE_STEPS[nextStep]?.id)) {
nextStep++
}
if (nextStep >= steps.length) nextStep = steps.length - 1
setSteps((prev) =>
prev.map((s, i) => {
if (i === currentStep) return { ...s, status: 'completed' }
if (i === nextStep) return { ...s, status: 'active' }
// Mark skipped steps between current and next
if (i > currentStep && i < nextStep && skipSteps.includes(PIPELINE_STEPS[i]?.id)) {
return { ...s, status: 'skipped' }
}
return s
}),
)
setCurrentStep(nextStep)
}
const handleOrientationComplete = async (sid: string) => {
setSessionId(sid)
loadSessions() loadSessions()
// Navigate directly to deskew step (index 1) for this session
nav.goToSession(sid)
}, [nav, loadSessions])
// Check for page-split sub-sessions directly from API // Crop: detect doc type then advance
// (React state may not be committed yet due to batching) const handleCropNext = useCallback(async () => {
try { if (nav.sessionId) {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sid}`)
if (res.ok) {
const data = await res.json()
if (data.sub_sessions?.length > 0) {
const subs: SubSession[] = data.sub_sessions.map((s: SubSession) => ({
id: s.id,
name: s.name,
box_index: s.box_index,
current_step: s.current_step,
}))
setSubSessions(subs)
setParentSessionId(sid)
openSession(subs[0].id, true)
return
}
}
} catch (e) {
console.error('Failed to check for sub-sessions:', e)
}
handleNext()
}
const handleCropNext = async () => {
// Auto-detect document type after crop (last image-processing step), then advance
if (sessionId) {
try { try {
const res = await fetch( const res = await fetch(
`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sessionId}/detect-type`, `${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${nav.sessionId}/detect-type`,
{ method: 'POST' }, { method: 'POST' },
) )
if (res.ok) { if (res.ok) {
const data: DocumentTypeResult = await res.json() const data: DocumentTypeResult = await res.json()
setDocTypeResult(data) nav.setDocType(data)
// Mark skipped steps immediately
const skipSteps = data.skip_steps || []
if (skipSteps.length > 0) {
setSteps((prev) =>
prev.map((s) =>
skipSteps.includes(s.id) ? { ...s, status: 'skipped' } : s,
),
)
}
} }
} catch (e) { } catch (e) {
console.error('Doc type detection failed:', e) console.error('Doc type detection failed:', e)
// Not critical — continue without it
} }
} }
handleNext() nav.goToNextStep()
} }, [nav])
const handleDocTypeChange = (newDocType: DocumentTypeResult['doc_type']) => { const handleDocTypeChange = (newDocType: DocumentTypeResult['doc_type']) => {
if (!docTypeResult) return if (!nav.docTypeResult) return
// Build new skip_steps based on doc type
let skipSteps: string[] = [] let skipSteps: string[] = []
if (newDocType === 'full_text') { if (newDocType === 'full_text') skipSteps = ['columns', 'rows']
skipSteps = ['columns', 'rows']
}
// vocab_table and generic_table: no skips
const updated: DocumentTypeResult = { nav.setDocType({
...docTypeResult, ...nav.docTypeResult,
doc_type: newDocType, doc_type: newDocType,
skip_steps: skipSteps, skip_steps: skipSteps,
pipeline: newDocType === 'full_text' ? 'full_page' : 'cell_first', pipeline: newDocType === 'full_text' ? 'full_page' : 'cell_first',
} })
setDocTypeResult(updated)
// Update step statuses
setSteps((prev) =>
prev.map((s) => {
if (skipSteps.includes(s.id)) return { ...s, status: 'skipped' as const }
if (s.status === 'skipped') return { ...s, status: 'pending' as const }
return s
}),
)
} }
const handleNewSession = () => { // Box sub-sessions (column detection) — still supported
setSessionId(null) const handleBoxSessionsCreated = useCallback((_subs: SubSession[]) => {
setSessionName('') // Box sub-sessions are tracked by the backend; no client-side state needed anymore
setCurrentStep(0) }, [])
setDocTypeResult(null)
setSubSessions([])
setParentSessionId(null)
setSteps(PIPELINE_STEPS.map((s, i) => ({ ...s, status: i === 0 ? 'active' : 'pending' })))
}
const handleSessionChange = useCallback((newSessionId: string) => {
openSession(newSessionId, true)
}, [openSession])
const handleBoxSessionsCreated = useCallback((subs: SubSession[]) => {
setSubSessions(subs)
if (sessionId) setParentSessionId(sessionId)
}, [sessionId])
const stepNames: Record<number, string> = {
1: 'Orientierung',
2: 'Begradigung',
3: 'Entzerrung',
4: 'Zuschneiden',
5: 'Spalten',
6: 'Zeilen',
7: 'Woerter',
8: 'Struktur',
9: 'Korrektur',
10: 'Rekonstruktion',
11: 'Validierung',
}
const reprocessFromStep = useCallback(async (uiStep: number) => {
if (!sessionId) return
const dbStep = uiStep + 1 // UI is 0-indexed, DB is 1-indexed
if (!confirm(`Ab Schritt ${dbStep} (${stepNames[dbStep] || '?'}) neu verarbeiten? Nachfolgende Daten werden geloescht.`)) return
try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sessionId}/reprocess`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ from_step: dbStep }),
})
if (!res.ok) {
const data = await res.json().catch(() => ({}))
console.error('Reprocess failed:', data.detail || res.status)
return
}
// Reset UI steps
goToStep(uiStep)
} catch (e) {
console.error('Reprocess error:', e)
}
// eslint-disable-next-line react-hooks/exhaustive-deps
}, [sessionId, goToStep])
const renderStep = () => { const renderStep = () => {
switch (currentStep) { const sid = nav.sessionId
switch (nav.currentStepIndex) {
case 0: case 0:
return <StepOrientation key={sessionId} sessionId={sessionId} onNext={handleOrientationComplete} onSubSessionsCreated={handleBoxSessionsCreated} /> return (
<StepOrientation
key={sid}
sessionId={sid}
onNext={handleOrientationComplete}
onSessionList={() => { loadSessions(); nav.goToSessionList() }}
/>
)
case 1: case 1:
return <StepDeskew key={sessionId} sessionId={sessionId} onNext={handleNext} /> return <StepDeskew key={sid} sessionId={sid} onNext={nav.goToNextStep} />
case 2: case 2:
return <StepDewarp key={sessionId} sessionId={sessionId} onNext={handleNext} /> return <StepDewarp key={sid} sessionId={sid} onNext={nav.goToNextStep} />
case 3: case 3:
return <StepCrop key={sessionId} sessionId={sessionId} onNext={handleCropNext} /> return <StepCrop key={sid} sessionId={sid} onNext={handleCropNext} />
case 4: case 4:
return <StepColumnDetection sessionId={sessionId} onNext={handleNext} onBoxSessionsCreated={handleBoxSessionsCreated} /> return <StepColumnDetection sessionId={sid} onNext={nav.goToNextStep} onBoxSessionsCreated={handleBoxSessionsCreated} />
case 5: case 5:
return <StepRowDetection sessionId={sessionId} onNext={handleNext} /> return <StepRowDetection sessionId={sid} onNext={nav.goToNextStep} />
case 6: case 6:
return <StepWordRecognition sessionId={sessionId} onNext={handleNext} goToStep={goToStep} /> return <StepWordRecognition sessionId={sid} onNext={nav.goToNextStep} goToStep={nav.goToStep} />
case 7: case 7:
return <StepStructureDetection sessionId={sessionId} onNext={handleNext} /> return <StepStructureDetection sessionId={sid} onNext={nav.goToNextStep} />
case 8: case 8:
return <StepLlmReview sessionId={sessionId} onNext={handleNext} /> return <StepLlmReview sessionId={sid} onNext={nav.goToNextStep} />
case 9: case 9:
return <StepReconstruction sessionId={sessionId} onNext={handleNext} /> return <StepReconstruction sessionId={sid} onNext={nav.goToNextStep} />
case 10: case 10:
return <StepGroundTruth sessionId={sessionId} onNext={handleNext} /> return <StepGroundTruth sessionId={sid} onNext={nav.goToNextStep} />
default: default:
return null return null
} }
@@ -485,7 +247,7 @@ export default function OcrPipelinePage() {
</button> </button>
)} )}
<button <button
onClick={handleNewSession} onClick={() => nav.goToSessionList()}
className="text-xs px-3 py-1.5 bg-teal-600 text-white rounded-lg hover:bg-teal-700 transition-colors" className="text-xs px-3 py-1.5 bg-teal-600 text-white rounded-lg hover:bg-teal-700 transition-colors"
> >
+ Neue Session + Neue Session
@@ -505,7 +267,7 @@ export default function OcrPipelinePage() {
<div <div
key={s.id} key={s.id}
className={`relative flex items-start gap-3 px-3 py-2.5 rounded-lg text-sm transition-colors cursor-pointer ${ className={`relative flex items-start gap-3 px-3 py-2.5 rounded-lg text-sm transition-colors cursor-pointer ${
sessionId === s.id nav.sessionId === s.id
? 'bg-teal-50 dark:bg-teal-900/30 border border-teal-200 dark:border-teal-700' ? 'bg-teal-50 dark:bg-teal-900/30 border border-teal-200 dark:border-teal-700'
: 'hover:bg-gray-50 dark:hover:bg-gray-700/50' : 'hover:bg-gray-50 dark:hover:bg-gray-700/50'
}`} }`}
@@ -561,13 +323,12 @@ export default function OcrPipelinePage() {
</button> </button>
<div className="text-xs text-gray-400 flex gap-2 mt-0.5"> <div className="text-xs text-gray-400 flex gap-2 mt-0.5">
<span>{new Date(s.created_at).toLocaleDateString('de-DE', { day: '2-digit', month: '2-digit', year: '2-digit', hour: '2-digit', minute: '2-digit' })}</span> <span>{new Date(s.created_at).toLocaleDateString('de-DE', { day: '2-digit', month: '2-digit', year: '2-digit', hour: '2-digit', minute: '2-digit' })}</span>
<span>Schritt {s.current_step}: {stepNames[s.current_step] || '?'}</span> <span>Schritt {s.current_step}: {STEP_NAMES[s.current_step] || '?'}</span>
</div> </div>
</div> </div>
{/* Badges */} {/* Badges */}
<div className="flex flex-col gap-1 items-end flex-shrink-0" onClick={(e) => e.stopPropagation()}> <div className="flex flex-col gap-1 items-end flex-shrink-0" onClick={(e) => e.stopPropagation()}>
{/* Category Badge */}
<button <button
onClick={() => setEditingCategory(editingCategory === s.id ? null : s.id)} onClick={() => setEditingCategory(editingCategory === s.id ? null : s.id)}
className={`text-[10px] px-1.5 py-0.5 rounded-full border transition-colors ${ className={`text-[10px] px-1.5 py-0.5 rounded-full border transition-colors ${
@@ -579,7 +340,6 @@ export default function OcrPipelinePage() {
> >
{catInfo ? `${catInfo.icon} ${catInfo.label}` : '+ Kategorie'} {catInfo ? `${catInfo.icon} ${catInfo.label}` : '+ Kategorie'}
</button> </button>
{/* Doc Type Badge (read-only) */}
{s.doc_type && ( {s.doc_type && (
<span className="text-[10px] px-1.5 py-0.5 rounded-full bg-gray-100 dark:bg-gray-700 text-gray-500 dark:text-gray-400 border border-gray-200 dark:border-gray-600"> <span className="text-[10px] px-1.5 py-0.5 rounded-full bg-gray-100 dark:bg-gray-700 text-gray-500 dark:text-gray-400 border border-gray-200 dark:border-gray-600">
{s.doc_type} {s.doc_type}
@@ -616,7 +376,7 @@ export default function OcrPipelinePage() {
</button> </button>
</div> </div>
{/* Category dropdown (inline) */} {/* Category dropdown */}
{editingCategory === s.id && ( {editingCategory === s.id && (
<div <div
className="absolute right-0 top-full mt-1 z-20 bg-white dark:bg-gray-800 border border-gray-200 dark:border-gray-700 rounded-lg shadow-lg p-2 grid grid-cols-2 gap-1 w-64" className="absolute right-0 top-full mt-1 z-20 bg-white dark:bg-gray-800 border border-gray-200 dark:border-gray-700 rounded-lg shadow-lg p-2 grid grid-cols-2 gap-1 w-64"
@@ -645,40 +405,39 @@ export default function OcrPipelinePage() {
</div> </div>
{/* Active session info */} {/* Active session info */}
{sessionId && sessionName && ( {nav.sessionId && sessionName && (
<div className="flex items-center gap-3 text-sm text-gray-500 dark:text-gray-400"> <div className="flex items-center gap-3 text-sm text-gray-500 dark:text-gray-400">
<span>Aktive Session: <span className="font-medium text-gray-700 dark:text-gray-300">{sessionName}</span></span> <span>Aktive Session: <span className="font-medium text-gray-700 dark:text-gray-300">{sessionName}</span></span>
{activeCategory && (() => { {activeCategory && (() => {
const cat = DOCUMENT_CATEGORIES.find(c => c.value === activeCategory) const cat = DOCUMENT_CATEGORIES.find(c => c.value === activeCategory)
return cat ? <span className="text-xs px-2 py-0.5 rounded-full bg-teal-50 dark:bg-teal-900/30 border border-teal-200 dark:border-teal-700 text-teal-700 dark:text-teal-300">{cat.icon} {cat.label}</span> : null return cat ? <span className="text-xs px-2 py-0.5 rounded-full bg-teal-50 dark:bg-teal-900/30 border border-teal-200 dark:border-teal-700 text-teal-700 dark:text-teal-300">{cat.icon} {cat.label}</span> : null
})()} })()}
{docTypeResult && ( {nav.docTypeResult && (
<span className="text-xs px-2 py-0.5 rounded-full bg-gray-100 dark:bg-gray-700 text-gray-500 dark:text-gray-400 border border-gray-200 dark:border-gray-600"> <span className="text-xs px-2 py-0.5 rounded-full bg-gray-100 dark:bg-gray-700 text-gray-500 dark:text-gray-400 border border-gray-200 dark:border-gray-600">
{docTypeResult.doc_type} {nav.docTypeResult.doc_type}
</span> </span>
)} )}
</div> </div>
)} )}
<PipelineStepper <PipelineStepper
steps={steps} steps={nav.steps}
currentStep={currentStep} currentStep={nav.currentStepIndex}
onStepClick={handleStepClick} onStepClick={handleStepClick}
onReprocess={sessionId ? reprocessFromStep : undefined} onReprocess={nav.sessionId ? nav.reprocessFromStep : undefined}
docTypeResult={docTypeResult} docTypeResult={nav.docTypeResult}
onDocTypeChange={handleDocTypeChange} onDocTypeChange={handleDocTypeChange}
/> />
{subSessions.length > 0 && parentSessionId && sessionId && (
<BoxSessionTabs
parentSessionId={parentSessionId}
subSessions={subSessions}
activeSessionId={sessionId}
onSessionChange={handleSessionChange}
/>
)}
<div className="min-h-[400px]">{renderStep()}</div> <div className="min-h-[400px]">{renderStep()}</div>
</div> </div>
) )
} }
export default function OcrPipelinePage() {
return (
<Suspense fallback={<div className="p-8 text-gray-400">Lade Pipeline...</div>}>
<OcrPipelineContent />
</Suspense>
)
}

3
admin-lehrer/app/(admin)/ai/ocr-pipeline/types.ts
View File

@@ -35,10 +35,9 @@ export interface SessionListItem {
doc_type?: string doc_type?: string
created_at: string created_at: string
updated_at?: string updated_at?: string
parent_session_id?: string | null
box_index?: number | null
} }
/** Box sub-session (from column detection zone_type='box') */
export interface SubSession { export interface SubSession {
id: string id: string
name: string name: string

225
admin-lehrer/app/(admin)/ai/ocr-pipeline/usePipelineNavigation.ts Normal file
View File

@@ -0,0 +1,225 @@
'use client'
import { useCallback, useEffect, useState } from 'react'
import { useRouter, useSearchParams } from 'next/navigation'
import { PIPELINE_STEPS, type PipelineStep, type PipelineStepStatus, type DocumentTypeResult } from './types'
const KLAUSUR_API = '/klausur-api'
export interface PipelineNav {
sessionId: string | null
currentStepIndex: number
currentStepId: string
steps: PipelineStep[]
docTypeResult: DocumentTypeResult | null
goToNextStep: () => void
goToStep: (index: number) => void
goToSession: (sessionId: string) => void
goToSessionList: () => void
setDocType: (result: DocumentTypeResult) => void
reprocessFromStep: (uiStep: number) => Promise<void>
}
const STEP_NAMES: Record<number, string> = {
1: 'Orientierung', 2: 'Begradigung', 3: 'Entzerrung', 4: 'Zuschneiden',
5: 'Spalten', 6: 'Zeilen', 7: 'Woerter', 8: 'Struktur',
9: 'Korrektur', 10: 'Rekonstruktion', 11: 'Validierung',
}
function buildSteps(uiStep: number, skipSteps: string[]): PipelineStep[] {
return PIPELINE_STEPS.map((s, i) => ({
...s,
status: (
skipSteps.includes(s.id) ? 'skipped'
: i < uiStep ? 'completed'
: i === uiStep ? 'active'
: 'pending'
) as PipelineStepStatus,
}))
}
export function usePipelineNavigation(): PipelineNav {
const router = useRouter()
const searchParams = useSearchParams()
const paramSession = searchParams.get('session')
const paramStep = searchParams.get('step')
const [sessionId, setSessionId] = useState<string | null>(paramSession)
const [currentStepIndex, setCurrentStepIndex] = useState(0)
const [docTypeResult, setDocTypeResult] = useState<DocumentTypeResult | null>(null)
const [steps, setSteps] = useState<PipelineStep[]>(buildSteps(0, []))
const [loaded, setLoaded] = useState(false)
// Load session info when session param changes
useEffect(() => {
if (!paramSession) {
setSessionId(null)
setCurrentStepIndex(0)
setDocTypeResult(null)
setSteps(buildSteps(0, []))
setLoaded(true)
return
}
const load = async () => {
try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${paramSession}`)
if (!res.ok) return
const data = await res.json()
setSessionId(paramSession)
const savedDocType: DocumentTypeResult | null = data.doc_type_result || null
setDocTypeResult(savedDocType)
const dbStep = data.current_step || 1
let uiStep = Math.max(0, dbStep - 1)
const skipSteps = [...(savedDocType?.skip_steps || [])]
// Box sub-sessions (from column detection) skip pre-processing
const isBoxSubSession = !!data.parent_session_id
if (isBoxSubSession && dbStep >= 5) {
const SUB_SESSION_SKIP = ['orientation', 'deskew', 'dewarp', 'crop']
for (const s of SUB_SESSION_SKIP) {
if (!skipSteps.includes(s)) skipSteps.push(s)
}
if (uiStep < 4) uiStep = 4
}
// If URL has a step param, use that instead
if (paramStep) {
const stepIdx = PIPELINE_STEPS.findIndex(s => s.id === paramStep)
if (stepIdx >= 0) uiStep = stepIdx
}
setCurrentStepIndex(uiStep)
setSteps(buildSteps(uiStep, skipSteps))
} catch (e) {
console.error('Failed to load session:', e)
} finally {
setLoaded(true)
}
}
load()
}, [paramSession, paramStep])
const updateUrl = useCallback((sid: string | null, stepIdx?: number) => {
if (!sid) {
router.push('/ai/ocr-pipeline')
return
}
const stepId = stepIdx !== undefined ? PIPELINE_STEPS[stepIdx]?.id : undefined
const params = new URLSearchParams()
params.set('session', sid)
if (stepId) params.set('step', stepId)
router.push(`/ai/ocr-pipeline?${params.toString()}`)
}, [router])
const goToNextStep = useCallback(() => {
if (currentStepIndex >= steps.length - 1) {
// Last step — return to session list
setSessionId(null)
setCurrentStepIndex(0)
setDocTypeResult(null)
setSteps(buildSteps(0, []))
router.push('/ai/ocr-pipeline')
return
}
const skipSteps = docTypeResult?.skip_steps || []
let nextStep = currentStepIndex + 1
while (nextStep < steps.length && skipSteps.includes(PIPELINE_STEPS[nextStep]?.id)) {
nextStep++
}
if (nextStep >= steps.length) nextStep = steps.length - 1
setSteps(prev =>
prev.map((s, i) => {
if (i === currentStepIndex) return { ...s, status: 'completed' as PipelineStepStatus }
if (i === nextStep) return { ...s, status: 'active' as PipelineStepStatus }
if (i > currentStepIndex && i < nextStep && skipSteps.includes(PIPELINE_STEPS[i]?.id)) {
return { ...s, status: 'skipped' as PipelineStepStatus }
}
return s
}),
)
setCurrentStepIndex(nextStep)
if (sessionId) updateUrl(sessionId, nextStep)
}, [currentStepIndex, steps.length, docTypeResult, sessionId, updateUrl, router])
const goToStep = useCallback((index: number) => {
setCurrentStepIndex(index)
setSteps(prev =>
prev.map((s, i) => ({
...s,
status: s.status === 'skipped' ? 'skipped'
: i < index ? 'completed'
: i === index ? 'active'
: 'pending' as PipelineStepStatus,
})),
)
if (sessionId) updateUrl(sessionId, index)
}, [sessionId, updateUrl])
const goToSession = useCallback((sid: string) => {
updateUrl(sid)
}, [updateUrl])
const goToSessionList = useCallback(() => {
setSessionId(null)
setCurrentStepIndex(0)
setDocTypeResult(null)
setSteps(buildSteps(0, []))
router.push('/ai/ocr-pipeline')
}, [router])
const setDocType = useCallback((result: DocumentTypeResult) => {
setDocTypeResult(result)
const skipSteps = result.skip_steps || []
if (skipSteps.length > 0) {
setSteps(prev =>
prev.map(s =>
skipSteps.includes(s.id) ? { ...s, status: 'skipped' as PipelineStepStatus } : s,
),
)
}
}, [])
const reprocessFromStep = useCallback(async (uiStep: number) => {
if (!sessionId) return
const dbStep = uiStep + 1
if (!confirm(`Ab Schritt ${dbStep} (${STEP_NAMES[dbStep] || '?'}) neu verarbeiten? Nachfolgende Daten werden geloescht.`)) return
try {
const res = await fetch(`${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sessionId}/reprocess`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ from_step: dbStep }),
})
if (!res.ok) {
const data = await res.json().catch(() => ({}))
console.error('Reprocess failed:', data.detail || res.status)
return
}
goToStep(uiStep)
} catch (e) {
console.error('Reprocess error:', e)
}
}, [sessionId, goToStep])
return {
sessionId,
currentStepIndex,
currentStepId: PIPELINE_STEPS[currentStepIndex]?.id || 'orientation',
steps,
docTypeResult,
goToNextStep,
goToStep,
goToSession,
goToSessionList,
setDocType,
reprocessFromStep,
}
}

5
admin-lehrer/components/ocr-pipeline/BoxSessionTabs.tsx
View File

@@ -21,6 +21,7 @@ function getStatusIcon(sub: SubSession): string {
return STATUS_ICONS.pending return STATUS_ICONS.pending
} }
/** Tabs for box sub-sessions (from column detection zone_type='box'). */
export function BoxSessionTabs({ parentSessionId, subSessions, activeSessionId, onSessionChange }: BoxSessionTabsProps) { export function BoxSessionTabs({ parentSessionId, subSessions, activeSessionId, onSessionChange }: BoxSessionTabsProps) {
if (subSessions.length === 0) return null if (subSessions.length === 0) return null
@@ -28,7 +29,6 @@ export function BoxSessionTabs({ parentSessionId, subSessions, activeSessionId,
return ( return (
<div className="flex items-center gap-1.5 px-1 py-1.5 bg-gray-50 dark:bg-gray-800/50 rounded-xl border border-gray-200 dark:border-gray-700"> <div className="flex items-center gap-1.5 px-1 py-1.5 bg-gray-50 dark:bg-gray-800/50 rounded-xl border border-gray-200 dark:border-gray-700">
{/* Main session tab */}
<button <button
onClick={() => onSessionChange(parentSessionId)} onClick={() => onSessionChange(parentSessionId)}
className={`px-3 py-1.5 rounded-lg text-xs font-medium transition-colors ${ className={`px-3 py-1.5 rounded-lg text-xs font-medium transition-colors ${
@@ -42,7 +42,6 @@ export function BoxSessionTabs({ parentSessionId, subSessions, activeSessionId,
<div className="w-px h-5 bg-gray-200 dark:bg-gray-700" /> <div className="w-px h-5 bg-gray-200 dark:bg-gray-700" />
{/* Sub-session tabs */}
{subSessions.map((sub) => { {subSessions.map((sub) => {
const isActive = activeSessionId === sub.id const isActive = activeSessionId === sub.id
const icon = getStatusIcon(sub) const icon = getStatusIcon(sub)
@@ -59,7 +58,7 @@ export function BoxSessionTabs({ parentSessionId, subSessions, activeSessionId,
title={sub.name} title={sub.name}
> >
<span className="mr-1">{icon}</span> <span className="mr-1">{icon}</span>
Seite {sub.box_index + 1} Box {sub.box_index + 1}
</button> </button>
) )
})} })}

43
admin-lehrer/components/ocr-pipeline/StepOrientation.tsx
View File

@@ -1,7 +1,7 @@
'use client' 'use client'
import { useCallback, useEffect, useState } from 'react' import { useCallback, useEffect, useState } from 'react'
import type { OrientationResult, SessionInfo, SubSession } from '@/app/(admin)/ai/ocr-pipeline/types' import type { OrientationResult, SessionInfo } from '@/app/(admin)/ai/ocr-pipeline/types'
import { ImageCompareView } from './ImageCompareView' import { ImageCompareView } from './ImageCompareView'
const KLAUSUR_API = '/klausur-api' const KLAUSUR_API = '/klausur-api'
@@ -17,10 +17,10 @@ interface PageSplitResult {
interface StepOrientationProps { interface StepOrientationProps {
sessionId?: string | null sessionId?: string | null
onNext: (sessionId: string) => void onNext: (sessionId: string) => void
onSubSessionsCreated?: (subs: SubSession[]) => void onSessionList?: () => void
} }
export function StepOrientation({ sessionId: existingSessionId, onNext, onSubSessionsCreated }: StepOrientationProps) { export function StepOrientation({ sessionId: existingSessionId, onNext, onSessionList }: StepOrientationProps) {
const [session, setSession] = useState<SessionInfo | null>(null) const [session, setSession] = useState<SessionInfo | null>(null)
const [orientationResult, setOrientationResult] = useState<OrientationResult | null>(null) const [orientationResult, setOrientationResult] = useState<OrientationResult | null>(null)
const [pageSplitResult, setPageSplitResult] = useState<PageSplitResult | null>(null) const [pageSplitResult, setPageSplitResult] = useState<PageSplitResult | null>(null)
@@ -112,16 +112,6 @@ export function StepOrientation({ sessionId: existingSessionId, onNext, onSubSes
if (splitRes.ok) { if (splitRes.ok) {
const splitData: PageSplitResult = await splitRes.json() const splitData: PageSplitResult = await splitRes.json()
setPageSplitResult(splitData) setPageSplitResult(splitData)
if (splitData.multi_page && splitData.sub_sessions && onSubSessionsCreated) {
onSubSessionsCreated(
splitData.sub_sessions.map((s) => ({
id: s.id,
name: s.name,
box_index: s.page_index,
current_step: splitData.used_original ? 1 : 2,
}))
)
}
} }
} catch (e) { } catch (e) {
console.error('Page-split detection failed:', e) console.error('Page-split detection failed:', e)
@@ -133,7 +123,7 @@ export function StepOrientation({ sessionId: existingSessionId, onNext, onSubSes
setUploading(false) setUploading(false)
setDetecting(false) setDetecting(false)
} }
}, [sessionName, onSubSessionsCreated]) }, [sessionName])
const handleDrop = useCallback((e: React.DragEvent) => { const handleDrop = useCallback((e: React.DragEvent) => {
e.preventDefault() e.preventDefault()
@@ -264,10 +254,10 @@ export function StepOrientation({ sessionId: existingSessionId, onNext, onSubSes
{pageSplitResult?.multi_page && ( {pageSplitResult?.multi_page && (
<div className="bg-blue-50 dark:bg-blue-900/20 rounded-lg border border-blue-200 dark:border-blue-700 p-4"> <div className="bg-blue-50 dark:bg-blue-900/20 rounded-lg border border-blue-200 dark:border-blue-700 p-4">
<div className="text-sm font-medium text-blue-700 dark:text-blue-300"> <div className="text-sm font-medium text-blue-700 dark:text-blue-300">
Doppelseite erkannt — {pageSplitResult.page_count} Seiten Doppelseite erkannt — {pageSplitResult.page_count} unabhaengige Sessions erstellt
</div> </div>
<p className="text-xs text-blue-600 dark:text-blue-400 mt-1"> <p className="text-xs text-blue-600 dark:text-blue-400 mt-1">
Jede Seite wird einzeln durch die Pipeline (Begradigung, Entzerrung, Zuschnitt, ...) verarbeitet. Jede Seite wird als eigene Session durch die Pipeline verarbeitet.
{pageSplitResult.used_original && ' (Seitentrennung auf dem Originalbild, da die Orientierung die Doppelseite gedreht hat.)'} {pageSplitResult.used_original && ' (Seitentrennung auf dem Originalbild, da die Orientierung die Doppelseite gedreht hat.)'}
</p> </p>
<div className="flex gap-2 mt-2"> <div className="flex gap-2 mt-2">
@@ -286,12 +276,21 @@ export function StepOrientation({ sessionId: existingSessionId, onNext, onSubSes
{/* Next button */} {/* Next button */}
{orientationResult && ( {orientationResult && (
<div className="flex justify-end"> <div className="flex justify-end">
<button {pageSplitResult?.multi_page ? (
onClick={() => onNext(session.session_id)} <button
className="px-6 py-2 bg-teal-600 text-white rounded-lg hover:bg-teal-700 font-medium transition-colors" onClick={() => onSessionList?.()}
> className="px-6 py-2 bg-teal-600 text-white rounded-lg hover:bg-teal-700 font-medium transition-colors"
{pageSplitResult?.multi_page ? 'Seiten verarbeiten' : 'Weiter'} &rarr; >
</button> Zur Session-Liste &rarr;
</button>
) : (
<button
onClick={() => onNext(session.session_id)}
className="px-6 py-2 bg-teal-600 text-white rounded-lg hover:bg-teal-700 font-medium transition-colors"
>
Weiter &rarr;
</button>
)}
</div> </div>
)} )}

2
klausur-service/backend/ocr_pipeline_session_store.py
View File

@@ -238,7 +238,7 @@ async def list_sessions_db(
""" """
pool = await get_pool() pool = await get_pool()
async with pool.acquire() as conn: async with pool.acquire() as conn:
where = "" if include_sub_sessions else "WHERE parent_session_id IS NULL" where = "" if include_sub_sessions else "WHERE parent_session_id IS NULL AND (status IS NULL OR status != 'split')"
rows = await conn.fetch(f""" rows = await conn.fetch(f"""
SELECT id, name, filename, status, current_step, SELECT id, name, filename, status, current_step,
document_category, doc_type, document_category, doc_type,

4
klausur-service/backend/ocr_pipeline_sessions.py
View File

@@ -191,12 +191,12 @@ async def get_session_info(session_id: str):
if session.get("ground_truth"): if session.get("ground_truth"):
result["ground_truth"] = session["ground_truth"] result["ground_truth"] = session["ground_truth"]
# Sub-session info # Box sub-session info (zone_type='box' from column detection — NOT page-split)
if session.get("parent_session_id"): if session.get("parent_session_id"):
result["parent_session_id"] = session["parent_session_id"] result["parent_session_id"] = session["parent_session_id"]
result["box_index"] = session.get("box_index") result["box_index"] = session.get("box_index")
else: else:
# Check for sub-sessions # Check for box sub-sessions (column detection creates these)
subs = await get_sub_sessions(session_id) subs = await get_sub_sessions(session_id)
if subs: if subs:
result["sub_sessions"] = [ result["sub_sessions"] = [

10
klausur-service/backend/orientation_crop_api.py
View File

@@ -238,8 +238,8 @@ async def detect_page_split(session_id: str):
"duration_seconds": round(duration, 2), "duration_seconds": round(duration, 2),
} }
# Mark parent session as split (store info in crop_result for backward compat) # Mark parent session as split and hidden from session list
await update_session_db(session_id, crop_result=split_info) await update_session_db(session_id, crop_result=split_info, status='split')
cached["crop_result"] = split_info cached["crop_result"] = split_info
await _append_pipeline_log(session_id, "page_split", { await _append_pipeline_log(session_id, "page_split", {
@@ -346,6 +346,7 @@ async def auto_crop(session_id: str):
cropped_png=png_buf.tobytes() if ok else b"", cropped_png=png_buf.tobytes() if ok else b"",
crop_result=crop_info, crop_result=crop_info,
current_step=5, current_step=5,
status='split',
) )
logger.info( logger.info(
@@ -461,8 +462,6 @@ async def _create_page_sub_sessions(
name=sub_name, name=sub_name,
filename=parent_filename, filename=parent_filename,
original_png=page_png, original_png=page_png,
parent_session_id=parent_session_id,
box_index=pi,
) )
# Pre-populate: set cropped = original (already cropped) # Pre-populate: set cropped = original (already cropped)
@@ -540,8 +539,6 @@ async def _create_page_sub_sessions_full(
name=sub_name, name=sub_name,
filename=parent_filename, filename=parent_filename,
original_png=page_png, original_png=page_png,
parent_session_id=parent_session_id,
box_index=pi,
) )
# start_step=2 → ready for deskew (orientation already done on spread) # start_step=2 → ready for deskew (orientation already done on spread)
@@ -553,7 +550,6 @@ async def _create_page_sub_sessions_full(
"id": sub_id, "id": sub_id,
"filename": parent_filename, "filename": parent_filename,
"name": sub_name, "name": sub_name,
"parent_session_id": parent_session_id,
"original_bgr": page_bgr, "original_bgr": page_bgr,
"oriented_bgr": None, "oriented_bgr": None,
"cropped_bgr": None, "cropped_bgr": None,