feat: reimport pdf

data/.gitignore

@@ -3,5 +3,8 @@
 *.sqlite-journal
 *.db
 
+# Ignore stored documents (PDFs and JSONs)
+documents/
+
 # Keep this .gitignore
 !.gitignore

frontend/src/App.vue

@@ -18,6 +18,14 @@
             Accueil
           </router-link>
           
+          <router-link 
+            to="/documents" 
+            class="px-3 py-1.5 text-sm rounded transition-colors"
+            :class="$route.path === '/documents' ? 'bg-gray-700 text-white' : 'text-gray-400 hover:text-white'"
+          >
+            Documents
+          </router-link>
+          
           <router-link 
             to="/revenus" 
             class="px-3 py-1.5 text-sm rounded transition-colors"

frontend/src/components/JsonViewer.vue

@@ -327,13 +327,32 @@ function updateOperationsGroup(categorie, updatedGroupOperations) {
   if (!updated.data) updated.data = {}
   if (!updated.data.recapitulatif_operations) updated.data.recapitulatif_operations = []
   
-  // Reconstruire la liste complete des operations
-  // 1. Filtrer toutes les operations de cette categorie
-  const otherOperations = updated.data.recapitulatif_operations.filter(op => op.categorie !== categorie)
+  // Reconstruire la liste en preservant l'ordre original
+  const newOperations = []
+  let updatedIndex = 0
   
-  // 2. Combiner avec les operations mises a jour
-  updated.data.recapitulatif_operations = [...otherOperations, ...updatedGroupOperations]
+  // Parcourir les operations originales et remplacer celles de la categorie modifiee
+  for (const op of updated.data.recapitulatif_operations) {
+    if (op.categorie === categorie) {
+      // Remplacer par les operations mises a jour (dans l'ordre)
+      if (updatedIndex < updatedGroupOperations.length) {
+        newOperations.push(updatedGroupOperations[updatedIndex])
+        updatedIndex++
+      }
+      // Si on a supprime des operations, on ne les ajoute pas
+    } else {
+      // Garder les operations des autres categories telles quelles
+      newOperations.push(op)
+    }
+  }
   
+  // Ajouter les nouvelles operations si on en a ajoute
+  while (updatedIndex < updatedGroupOperations.length) {
+    newOperations.push(updatedGroupOperations[updatedIndex])
+    updatedIndex++
+  }
+  
+  updated.data.recapitulatif_operations = newOperations
   emit('update:data', updated)
 }
 

frontend/src/components/PdfPreview.vue

@@ -113,6 +113,10 @@ const props = defineProps({
     type: File,
     default: null
   },
+  url: {
+    type: String,
+    default: null
+  },
   fileName: {
     type: String,
     default: 'document.pdf'
@@ -131,14 +135,26 @@ const baseScale = ref(1)
 let pdfDoc = null
 let renderTask = null
 
-async function loadPdf(file) {
-  if (!file) return
+async function loadPdf(source) {
+  if (!source) return
   
   isLoading.value = true
   error.value = null
   
   try {
-    const arrayBuffer = await file.arrayBuffer()
+    let arrayBuffer
+    
+    // Support URL
+    if (typeof source === 'string') {
+      const response = await fetch(source)
+      if (!response.ok) throw new Error('Erreur lors du chargement du PDF')
+      arrayBuffer = await response.arrayBuffer()
+    } 
+    // Support File
+    else {
+      arrayBuffer = await source.arrayBuffer()
+    }
+    
     const loadingTask = pdfjsLib.getDocument({ data: arrayBuffer })
     pdfDoc = await loadingTask.promise
     
@@ -251,9 +267,10 @@ function onWheel(e) {
   }
 }
 
-watch(() => props.file, async (newFile) => {
-  if (newFile) {
-    await loadPdf(newFile)
+watch([() => props.file, () => props.url], async ([newFile, newUrl]) => {
+  const source = newUrl || newFile
+  if (source) {
+    await loadPdf(source)
   } else {
     pdfDoc = null
     totalPages.value = 0

frontend/src/pages/DocumentsPage.vue

@@ -0,0 +1,234 @@
+<template>
+  <div class="flex-1 overflow-auto p-6 bg-gray-800">
+    <div class="max-w-6xl mx-auto space-y-6">
+      
+      <!-- Header -->
+      <div class="flex items-center justify-between">
+        <div>
+          <h1 class="text-xl font-semibold text-white">Documents importes</h1>
+          <p class="text-sm text-gray-400 mt-1">
+            {{ documents.length }} document{{ documents.length > 1 ? 's' : '' }} en base
+          </p>
+        </div>
+        
+        <!-- Import button -->
+        <label class="px-4 py-2 bg-blue-600 text-white rounded-lg hover:bg-blue-700 transition-colors cursor-pointer flex items-center gap-2">
+          <svg class="w-5 h-5" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+            <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16v1a3 3 0 003 3h10a3 3 0 003-3v-1m-4-8l-4-4m0 0L8 8m4-4v12" />
+          </svg>
+          Importer un PDF
+          <input
+            type="file"
+            accept="application/pdf,.pdf"
+            class="hidden"
+            @change="onFileSelect"
+          />
+        </label>
+      </div>
+
+      <!-- Loading state -->
+      <div v-if="isLoading" class="bg-gray-900 rounded-lg p-12 text-center">
+        <div class="inline-block animate-spin rounded-full h-8 w-8 border-2 border-gray-600 border-t-blue-400"></div>
+        <p class="text-gray-400 mt-4">Chargement des documents...</p>
+      </div>
+
+      <!-- Empty state -->
+      <div v-else-if="documents.length === 0" class="bg-gray-900 rounded-lg p-12 text-center">
+        <svg class="mx-auto h-16 w-16 text-gray-600 mb-4" fill="none" viewBox="0 0 24 24" stroke="currentColor">
+          <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z" />
+        </svg>
+        <p class="text-gray-400 text-lg">Aucun document importe</p>
+        <p class="text-gray-500 text-sm mt-2">Importez votre premier PDF pour commencer</p>
+      </div>
+
+      <!-- Documents table -->
+      <div v-else class="bg-gray-900 rounded-lg overflow-hidden border border-gray-700">
+        <table class="w-full">
+          <thead class="bg-gray-800 border-b border-gray-700">
+            <tr>
+              <th class="px-4 py-3 text-left text-xs font-medium text-gray-400 uppercase tracking-wider">Document</th>
+              <th class="px-4 py-3 text-left text-xs font-medium text-gray-400 uppercase tracking-wider">Immeuble</th>
+              <th class="px-4 py-3 text-left text-xs font-medium text-gray-400 uppercase tracking-wider">Date</th>
+              <th class="px-4 py-3 text-right text-xs font-medium text-gray-400 uppercase tracking-wider">Solde</th>
+              <th class="px-4 py-3 text-center text-xs font-medium text-gray-400 uppercase tracking-wider">Fichiers</th>
+              <th class="px-4 py-3 text-right text-xs font-medium text-gray-400 uppercase tracking-wider">Actions</th>
+            </tr>
+          </thead>
+          <tbody class="divide-y divide-gray-800">
+            <tr 
+              v-for="doc in documents" 
+              :key="doc.id"
+              class="hover:bg-gray-800/50 transition-colors"
+            >
+              <!-- Document info -->
+              <td class="px-4 py-4">
+                <div class="flex items-center gap-3">
+                  <div class="flex-shrink-0 w-10 h-10 bg-red-500/20 rounded-lg flex items-center justify-center">
+                    <svg class="w-5 h-5 text-red-400" fill="currentColor" viewBox="0 0 24 24">
+                      <path d="M14 2H6a2 2 0 0 0-2 2v16a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2V8l-6-6z"/>
+                      <path d="M14 2v6h6M16 13H8M16 17H8M10 9H8"/>
+                    </svg>
+                  </div>
+                  <div>
+                    <div class="text-sm font-medium text-white font-mono">{{ doc.reference }}</div>
+                    <div class="text-xs text-gray-500">{{ doc.source_file || 'Fichier non stocke' }}</div>
+                  </div>
+                </div>
+              </td>
+
+              <!-- Immeuble -->
+              <td class="px-4 py-4">
+                <div class="text-sm text-white">{{ doc.immeuble_adresse || '-' }}</div>
+                <div class="text-xs text-gray-500 font-mono">{{ doc.immeuble_code }}</div>
+              </td>
+
+              <!-- Date -->
+              <td class="px-4 py-4">
+                <div class="text-sm text-white">{{ formatDate(doc.date) }}</div>
+                <div class="text-xs text-gray-500">Importe le {{ formatDateTime(doc.created_at) }}</div>
+              </td>
+
+              <!-- Solde -->
+              <td class="px-4 py-4 text-right">
+                <span 
+                  :class="[
+                    'text-sm font-medium',
+                    doc.solde_type === 'crediteur' ? 'text-green-400' : 'text-red-400'
+                  ]"
+                >
+                  {{ formatAmount(doc.solde_montant) }}
+                </span>
+                <div class="text-xs text-gray-500">{{ doc.solde_type }}</div>
+              </td>
+
+              <!-- Files status -->
+              <td class="px-4 py-4">
+                <div class="flex items-center justify-center gap-2">
+                  <!-- PDF download -->
+                  <a 
+                    v-if="doc.has_pdf"
+                    :href="`/api/documents/${doc.id}/pdf`"
+                    target="_blank"
+                    class="p-1.5 rounded bg-red-500/20 text-red-400 hover:bg-red-500/30 transition-colors"
+                    title="Telecharger le PDF"
+                  >
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 10v6m0 0l-3-3m3 3l3-3m2 8H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z" />
+                    </svg>
+                  </a>
+                  <span v-else class="p-1.5 rounded bg-gray-700 text-gray-500" title="PDF non stocke">
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M6 18L18 6M6 6l12 12" />
+                    </svg>
+                  </span>
+
+                  <!-- JSON download -->
+                  <a 
+                    v-if="doc.has_json"
+                    :href="`/api/documents/${doc.id}/json`"
+                    target="_blank"
+                    class="p-1.5 rounded bg-yellow-500/20 text-yellow-400 hover:bg-yellow-500/30 transition-colors"
+                    title="Telecharger le JSON"
+                  >
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M4 16l4.586-4.586a2 2 0 012.828 0L16 16m-2-2l1.586-1.586a2 2 0 012.828 0L20 14m-6-6h.01M6 20h12a2 2 0 002-2V6a2 2 0 00-2-2H6a2 2 0 00-2 2v12a2 2 0 002 2z" />
+                    </svg>
+                  </a>
+                  <span v-else class="p-1.5 rounded bg-gray-700 text-gray-500" title="JSON non stocke">
+                    <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M6 18L18 6M6 6l12 12" />
+                    </svg>
+                  </span>
+                </div>
+              </td>
+
+              <!-- Actions -->
+              <td class="px-4 py-4 text-right">
+                <div class="flex items-center justify-end gap-2">
+                  <!-- Edit & reimport button -->
+                  <button
+                    @click="router.push(`/documents/${doc.id}/edit`)"
+                    class="px-3 py-1.5 text-xs rounded bg-blue-600/20 text-blue-400 hover:bg-blue-600 hover:text-white transition-colors flex items-center gap-1"
+                    title="Editer et reimporter"
+                  >
+                    <svg class="w-3.5 h-3.5" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+                      <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M11 5H6a2 1 0 00-2 2v11a2 2 0 002 2h11a2 2 0 002-2v-5m-1.414-9.414a2 2 0 112.828 2.828L11.828 15H9v-2.828l8.586-8.586z" />
+                    </svg>
+                    Editer
+                  </button>
+                </div>
+              </td>
+            </tr>
+          </tbody>
+        </table>
+      </div>
+
+    </div>
+
+  </div>
+</template>
+
+<script setup>
+import { ref, onMounted } from 'vue'
+import { useRouter } from 'vue-router'
+import { pendingFile } from '../store'
+
+const router = useRouter()
+
+// State
+const isLoading = ref(true)
+const documents = ref([])
+
+// Formatting helpers
+function formatDate(dateStr) {
+  if (!dateStr) return '-'
+  const [year, month, day] = dateStr.split('-')
+  return `${day}/${month}/${year}`
+}
+
+function formatDateTime(isoStr) {
+  if (!isoStr) return '-'
+  const date = new Date(isoStr)
+  return date.toLocaleDateString('fr-FR', {
+    day: '2-digit',
+    month: '2-digit', 
+    year: 'numeric',
+    hour: '2-digit',
+    minute: '2-digit'
+  })
+}
+
+function formatAmount(amount) {
+  if (amount == null) return '-'
+  return new Intl.NumberFormat('fr-FR', { style: 'currency', currency: 'EUR' }).format(amount)
+}
+
+// Load documents
+async function loadDocuments() {
+  isLoading.value = true
+  try {
+    const response = await fetch('/api/documents?limit=100')
+    if (response.ok) {
+      documents.value = await response.json()
+    }
+  } catch (err) {
+    console.error('Failed to load documents:', err)
+  } finally {
+    isLoading.value = false
+  }
+}
+
+// File selection for new import
+function onFileSelect(e) {
+  const file = e.target.files[0]
+  if (file && file.name.toLowerCase().endsWith('.pdf')) {
+    pendingFile.value = file
+    router.push('/extract')
+  }
+  e.target.value = ''
+}
+
+onMounted(() => {
+  loadDocuments()
+})
+</script>

frontend/src/pages/EditDocumentPage.vue

@@ -0,0 +1,217 @@
+<template>
+  <div class="flex-1 flex flex-col overflow-hidden">
+    <!-- Header avec breadcrumb et actions -->
+    <div class="flex-shrink-0 px-6 py-4 bg-gray-800 border-b border-gray-700">
+      <div class="flex items-center justify-between">
+        <div>
+          <div class="flex items-center gap-2 text-sm text-gray-400 mb-1">
+            <router-link to="/documents" class="hover:text-white transition-colors">Documents</router-link>
+            <span>›</span>
+            <span class="text-white">{{ documentData?.reference || 'Chargement...' }}</span>
+          </div>
+          <h1 class="text-xl font-semibold text-white">
+            Édition du document {{ documentData?.reference }}
+          </h1>
+          <p v-if="documentData" class="text-sm text-gray-400 mt-1">
+            {{ documentData.immeuble?.adresse || '-' }} - {{ formatDate(documentData.date) }}
+          </p>
+        </div>
+        
+        <div class="flex items-center gap-3">
+          <button
+            @click="cancel"
+            :disabled="isSaving"
+            class="px-4 py-2 text-sm text-gray-400 hover:text-white transition-colors disabled:opacity-50"
+          >
+            Annuler
+          </button>
+          <button 
+            v-if="!showTagging && extractedData"
+            @click="goToTagging"
+            :disabled="isSaving"
+            class="px-4 py-2 text-sm bg-blue-600 text-white rounded-lg hover:bg-blue-700 transition-colors disabled:opacity-50 flex items-center gap-2"
+          >
+            <svg class="w-4 h-4" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+              <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M7 7h.01M7 3h5c.512 0 1.024.195 1.414.586l7 7a2 2 0 010 2.828l-7 7a2 2 0 01-2.828 0l-7-7A1.994 1.994 0 013 12V7a4 4 0 014-4z" />
+            </svg>
+            Continuer vers le tagging
+          </button>
+        </div>
+      </div>
+    </div>
+
+    <!-- Loading state -->
+    <div v-if="isLoading" class="flex-1 flex items-center justify-center bg-gray-800">
+      <div class="text-center">
+        <div class="inline-block animate-spin rounded-full h-12 w-12 border-4 border-gray-600 border-t-blue-400 mb-4"></div>
+        <p class="text-gray-400">Chargement du document...</p>
+      </div>
+    </div>
+
+    <!-- Corps : Split view PDF + Données -->
+    <div v-else class="flex-1 flex overflow-hidden">
+      <!-- Gauche : PDF Preview -->
+      <div class="w-1/2 border-r border-gray-700 flex flex-col">
+        <div class="flex-shrink-0 px-4 py-2 bg-gray-700/50 border-b border-gray-700">
+          <span class="text-sm text-gray-300">Aperçu PDF</span>
+        </div>
+        <div class="flex-1 overflow-hidden">
+          <PdfPreview
+            v-if="documentData?.has_pdf"
+            :url="`/api/documents/${documentId}/pdf`"
+            :file-name="documentData?.source_file || 'document.pdf'"
+            class="h-full"
+          />
+          <div v-else class="h-full flex items-center justify-center bg-gray-800 text-gray-500">
+            <div class="text-center">
+              <svg class="mx-auto h-16 w-16 text-gray-600 mb-4" fill="none" viewBox="0 0 24 24" stroke="currentColor">
+                <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M9 12h6m-6 4h6m2 5H7a2 2 0 01-2-2V5a2 2 0 012-2h5.586a1 1 0 01.707.293l5.414 5.414a1 1 0 01.293.707V19a2 2 0 01-2 2z" />
+              </svg>
+              <p class="text-lg font-medium">PDF non disponible</p>
+              <p class="text-sm text-gray-600 mt-1">Ce document a été importé sans stockage du PDF</p>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <!-- Droite : Édition ou Tagging -->
+      <div class="w-1/2 flex flex-col bg-gray-50">
+        <!-- Vue 1 : JsonViewer (mode édition) -->
+        <JsonViewer 
+          v-if="!showTagging && extractedData"
+          :data="extractedData"
+          @update:data="extractedData = $event"
+          :is-loading="false"
+          class="flex-1"
+        />
+
+        <!-- Vue 2 : TaggingStep -->
+        <TaggingStep
+          v-if="showTagging && extractedData"
+          :depenses="extractedData.data.recapitulatif_operations"
+          :is-duplicate="true"
+          @save="handleSave"
+          @cancel="showTagging = false"
+          class="flex-1"
+        />
+        
+        <!-- Error message -->
+        <div v-if="saveError" class="flex-shrink-0 px-4 py-3 bg-red-500/10 border-t border-red-500/30">
+          <div class="flex items-center gap-2 text-sm text-red-400">
+            <svg class="w-5 h-5" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+              <path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 8v4m0 4h.01M21 12a9 9 0 11-18 0 9 9 0 0118 0z" />
+            </svg>
+            <span>{{ saveError }}</span>
+          </div>
+        </div>
+      </div>
+    </div>
+  </div>
+</template>
+
+<script setup>
+import { ref, onMounted } from 'vue'
+import { useRouter, useRoute } from 'vue-router'
+import PdfPreview from '../components/PdfPreview.vue'
+import JsonViewer from '../components/JsonViewer.vue'
+import TaggingStep from '../components/TaggingStep.vue'
+
+const router = useRouter()
+const route = useRoute()
+const documentId = route.params.id
+
+const documentData = ref(null)
+const extractedData = ref(null)
+const isLoading = ref(true)
+const showTagging = ref(false)
+const isSaving = ref(false)
+const saveError = ref(null)
+
+onMounted(async () => {
+  await loadDocument()
+})
+
+async function loadDocument() {
+  isLoading.value = true
+  try {
+    const response = await fetch(`/api/documents/${documentId}`)
+    if (!response.ok) {
+      throw new Error('Document non trouvé')
+    }
+    
+    documentData.value = await response.json()
+    
+    // Formatter les données pour JsonViewer (même structure que ExtractPage)
+    extractedData.value = {
+      source_file: documentData.value.source_file,
+      data: documentData.value.json_data
+    }
+  } catch (err) {
+    console.error('Error loading document:', err)
+    alert('Erreur lors du chargement du document: ' + err.message)
+    router.push('/documents')
+  } finally {
+    isLoading.value = false
+  }
+}
+
+function formatDate(dateStr) {
+  if (!dateStr) return '-'
+  const [year, month, day] = dateStr.split('-')
+  return `${day}/${month}/${year}`
+}
+
+function cancel() {
+  const hasChanges = extractedData.value !== null
+  if (hasChanges) {
+    const confirmed = confirm('Abandonner les modifications ?')
+    if (!confirmed) return
+  }
+  router.push('/documents')
+}
+
+function goToTagging() {
+  showTagging.value = true
+  saveError.value = null
+}
+
+async function handleSave(depensesTags, shouldOverwrite) {
+  isSaving.value = true
+  saveError.value = null
+  
+  try {
+    const response = await fetch('/api/save', {
+      method: 'POST',
+      headers: { 
+        'Content-Type': 'application/json' 
+      },
+      body: JSON.stringify({
+        source_file: extractedData.value.source_file,
+        data: extractedData.value.data,
+        depenses_tags: depensesTags,
+        overwrite: true  // Toujours true en mode édition
+      })
+    })
+    
+    const result = await response.json()
+    
+    if (!response.ok) {
+      throw new Error(result.detail || 'Erreur lors de la sauvegarde')
+    }
+    
+    if (result.success) {
+      // Succès - redirection vers la liste des documents
+      router.push('/documents')
+    } else {
+      throw new Error(result.message || 'Échec de la sauvegarde')
+    }
+  } catch (err) {
+    console.error('Save error:', err)
+    saveError.value = err.message || 'Une erreur est survenue lors de la sauvegarde'
+    // Scroll to top to show error
+    window.scrollTo({ top: 0, behavior: 'smooth' })
+  } finally {
+    isSaving.value = false
+  }
+}
+</script>

frontend/src/router.js

@@ -3,6 +3,8 @@ import HomePage from './pages/HomePage.vue'
 import ExtractPage from './pages/ExtractPage.vue'
 import AnalyticsPage from './pages/AnalyticsPage.vue'
 import RevenusPage from './pages/RevenusPage.vue'
+import DocumentsPage from './pages/DocumentsPage.vue'
+import EditDocumentPage from './pages/EditDocumentPage.vue'
 
 const routes = [
   {
@@ -24,6 +26,16 @@ const routes = [
     path: '/revenus',
     name: 'revenus',
     component: RevenusPage
+  },
+  {
+    path: '/documents',
+    name: 'documents',
+    component: DocumentsPage
+  },
+  {
+    path: '/documents/:id/edit',
+    name: 'edit-document',
+    component: EditDocumentPage
   }
 ]
 

src/plesna_gerance/api/routes/documents.py

@@ -3,13 +3,15 @@
 import json
 from datetime import datetime
 
-from fastapi import APIRouter, Depends, HTTPException
+from fastapi import APIRouter, Depends, HTTPException, File, UploadFile, Form
+from fastapi.responses import Response, JSONResponse
 from sqlalchemy.orm import Session
 from sqlalchemy import func, select
 
-from ...database import get_session, DatabaseService
+from ...database import get_session, DatabaseService, storage
 from ...database.models import Document, Immeuble, Lot, Locataire, Revenu, Depense
 from ...database.service import DuplicateDocumentError
+from ...extractor import extract_compte_rendu
 from ..schemas import SaveRequest, SaveResponse, DocumentSummary
 
 router = APIRouter(prefix="/api", tags=["documents"])
@@ -64,6 +66,80 @@ async def save_document(
         )
 
 
+@router.post("/save-with-pdf", response_model=SaveResponse)
+async def save_document_with_pdf(
+    pdf_file: UploadFile = File(..., description="Fichier PDF original"),
+    data: str = Form(..., description="Donnees JSON extraites"),
+    depenses_tags: str | None = Form(None, description="Tags JSON pour les depenses"),
+    overwrite: bool = Form(False, description="Ecraser si existant"),
+    session: Session = Depends(get_session),
+) -> SaveResponse:
+    """Sauvegarde les donnees extraites avec le fichier PDF original.
+
+    Cette version stocke le PDF et le JSON sur le disque pour tracabilite.
+
+    - **pdf_file**: Fichier PDF original (multipart)
+    - **data**: Donnees JSON extraites (string JSON)
+    - **depenses_tags**: Tags JSON pour les depenses (optionnel)
+    - **overwrite**: Si True, ecrase le document existant
+    """
+    # Parse JSON data
+    try:
+        parsed_data = json.loads(data)
+    except json.JSONDecodeError as e:
+        raise HTTPException(status_code=400, detail=f"JSON invalide pour data: {e}")
+
+    # Parse depenses_tags if provided
+    parsed_tags = None
+    if depenses_tags:
+        try:
+            parsed_tags = json.loads(depenses_tags)
+        except json.JSONDecodeError as e:
+            raise HTTPException(
+                status_code=400, detail=f"JSON invalide pour depenses_tags: {e}"
+            )
+
+    # Read PDF content
+    try:
+        pdf_content = await pdf_file.read()
+    except Exception as e:
+        raise HTTPException(status_code=400, detail=f"Erreur lecture du PDF: {str(e)}")
+
+    try:
+        db_service = DatabaseService(session)
+        document = db_service.save_document(
+            data=parsed_data,
+            source_file=pdf_file.filename,
+            pdf_content=pdf_content,
+            depenses_tags=parsed_tags,
+            overwrite=overwrite,
+        )
+
+        return SaveResponse(
+            success=True,
+            message="Document sauvegarde avec succes (PDF et JSON stockes)",
+            document_id=document.id,
+            reference=document.reference,
+            date=str(document.date),
+        )
+
+    except DuplicateDocumentError as e:
+        return SaveResponse(
+            success=False,
+            message=f"Document deja existant: reference={e.reference}, date={e.date}",
+            reference=e.reference,
+            date=str(e.date),
+        )
+
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+
+    except Exception as e:
+        raise HTTPException(
+            status_code=500, detail=f"Erreur lors de la sauvegarde: {str(e)}"
+        )
+
+
 @router.get("/stats")
 async def get_stats(
     session: Session = Depends(get_session),
@@ -110,6 +186,8 @@ async def list_documents(
             solde_montant=doc.solde_montant,
             solde_type=doc.solde_type,
             created_at=doc.created_at.isoformat() if doc.created_at else None,
+            has_pdf=doc.pdf_path is not None,
+            has_json=doc.json_path is not None,
         )
         for doc in documents
     ]
@@ -157,6 +235,8 @@ async def get_document(
         },
         "json_data": json.loads(document.json_data) if document.json_data else None,
         "created_at": document.created_at.isoformat() if document.created_at else None,
+        "has_pdf": document.pdf_path is not None,
+        "has_json": document.json_path is not None,
     }
 
 
@@ -189,3 +269,135 @@ async def check_duplicate(
         "reference": reference,
         "date": date,
     }
+
+
+@router.get("/documents/{document_id}/pdf")
+async def download_document_pdf(
+    document_id: int,
+    session: Session = Depends(get_session),
+) -> Response:
+    """Telecharge le fichier PDF original d'un document.
+
+    Retourne le fichier PDF si disponible, sinon erreur 404.
+    """
+    db_service = DatabaseService(session)
+    document = db_service.get_document_by_id(document_id)
+
+    if not document:
+        raise HTTPException(status_code=404, detail="Document non trouve")
+
+    if not document.pdf_path:
+        raise HTTPException(
+            status_code=404,
+            detail="Fichier PDF non disponible pour ce document",
+        )
+
+    try:
+        pdf_content = storage.read_pdf(document.pdf_path)
+        filename = document.source_file or f"{document.reference}.pdf"
+        return Response(
+            content=pdf_content,
+            media_type="application/pdf",
+            headers={"Content-Disposition": f'inline; filename="{filename}"'},
+        )
+    except FileNotFoundError:
+        raise HTTPException(
+            status_code=404,
+            detail="Fichier PDF introuvable sur le disque",
+        )
+
+
+@router.get("/documents/{document_id}/json")
+async def download_document_json(
+    document_id: int,
+    session: Session = Depends(get_session),
+) -> Response:
+    """Telecharge le fichier JSON extrait d'un document.
+
+    Retourne le fichier JSON si disponible sur disque,
+    sinon retourne le json_data de la base de donnees.
+    """
+    db_service = DatabaseService(session)
+    document = db_service.get_document_by_id(document_id)
+
+    if not document:
+        raise HTTPException(status_code=404, detail="Document non trouve")
+
+    filename = f"{document.reference}_{document.date}.json"
+
+    # Try to read from storage first
+    if document.json_path:
+        try:
+            json_data = storage.read_json(document.json_path)
+            return JSONResponse(
+                content=json_data,
+                headers={"Content-Disposition": f'attachment; filename="{filename}"'},
+            )
+        except FileNotFoundError:
+            pass  # Fall back to database
+
+    # Fall back to json_data in database
+    if document.json_data:
+        json_data = json.loads(document.json_data)
+        return JSONResponse(
+            content=json_data,
+            headers={"Content-Disposition": f'attachment; filename="{filename}"'},
+        )
+
+    raise HTTPException(
+        status_code=404,
+        detail="Donnees JSON non disponibles pour ce document",
+    )
+
+
+@router.post("/documents/{document_id}/re-extract")
+async def re_extract_document(
+    document_id: int,
+    session: Session = Depends(get_session),
+) -> dict:
+    """Re-extrait les donnees depuis le PDF stocke.
+
+    Utile pour corriger l'extraction apres amelioration des parsers.
+    Ne modifie pas automatiquement la base - retourne les nouvelles donnees
+    pour validation par l'utilisateur.
+
+    Retourne les donnees re-extraites du PDF.
+    """
+    db_service = DatabaseService(session)
+    document = db_service.get_document_by_id(document_id)
+
+    if not document:
+        raise HTTPException(status_code=404, detail="Document non trouve")
+
+    if not document.pdf_path:
+        raise HTTPException(
+            status_code=404,
+            detail="Fichier PDF non disponible pour ce document - re-extraction impossible",
+        )
+
+    # Get absolute path
+    pdf_full_path = storage.get_absolute_path(document.pdf_path)
+
+    if not pdf_full_path.exists():
+        raise HTTPException(
+            status_code=404,
+            detail="Fichier PDF introuvable sur le disque",
+        )
+
+    # Re-extract
+    try:
+        new_data = extract_compte_rendu(str(pdf_full_path))
+    except Exception as e:
+        raise HTTPException(
+            status_code=422,
+            detail=f"Erreur lors de la re-extraction: {str(e)}",
+        )
+
+    return {
+        "document_id": document_id,
+        "reference": document.reference,
+        "date": str(document.date),
+        "original_json_path": document.json_path,
+        "re_extracted_data": new_data,
+        "message": "Donnees re-extraites. Utilisez PUT /api/documents/{id} pour mettre a jour.",
+    }

src/plesna_gerance/api/schemas/models.py

@@ -54,6 +54,8 @@ class DocumentSummary(BaseModel):
     solde_montant: float | None
     solde_type: str | None
     created_at: str
+    has_pdf: bool = False
+    has_json: bool = False
 
 
 # ============================================================

src/plesna_gerance/database/__init__.py

@@ -3,6 +3,7 @@
 from .connection import get_engine, get_session, get_session_factory, init_db
 from .models import Base, Document, Immeuble, Lot, Locataire, Revenu, Depense
 from .service import DatabaseService, DuplicateDocumentError
+from . import storage
 
 __all__ = [
     "get_engine",
@@ -18,4 +19,5 @@ __all__ = [
     "Depense",
     "DatabaseService",
     "DuplicateDocumentError",
+    "storage",
 ]

src/plesna_gerance/database/models.py

@@ -144,6 +144,14 @@ class Document(Base):
 
     created_at = Column(DateTime, default=datetime.utcnow)
 
+    # Chemins vers les fichiers stockés (relatifs à PLESNA_STORAGE_PATH)
+    pdf_path = Column(
+        String(500), nullable=True
+    )  # ex: "2024/S_REF-01234_2024-01-15.pdf"
+    json_path = Column(
+        String(500), nullable=True
+    )  # ex: "2024/S_REF-01234_2024-01-15.json"
+
     # Clé unique: référence + date (évite les doublons)
     __table_args__ = (
         UniqueConstraint("reference", "date", name="uq_document_reference_date"),

src/plesna_gerance/database/service.py

@@ -9,6 +9,7 @@ from sqlalchemy.orm import Session
 from sqlalchemy.exc import IntegrityError
 
 from .models import Document, Immeuble, Lot, Locataire, Revenu, Depense, Tag
+from . import storage
 
 
 class DuplicateDocumentError(Exception):
@@ -98,6 +99,7 @@ class DatabaseService:
         self,
         data: dict[str, Any],
         source_file: str = None,
+        pdf_content: bytes = None,
         depenses_tags: list[dict] = None,
         overwrite: bool = False,
     ) -> Document:
@@ -107,6 +109,7 @@ class DatabaseService:
             data: The 'data' portion of the extracted JSON (contains metadata,
                   situation_locataires, recapitulatif_operations)
             source_file: Original PDF filename
+            pdf_content: Binary content of the PDF file (optional, for storage)
             depenses_tags: List of tags for expenses
             overwrite: If True, delete existing document and recreate it
 
@@ -129,11 +132,17 @@ class DatabaseService:
         if not reference or not doc_date:
             raise ValueError("Document must have reference and date")
 
-        # Check for duplicates
+        # Check for duplicates and preserve existing file paths if overwriting
+        existing_pdf_path = None
+        existing_json_path = None
         existing = self.check_duplicate(reference, doc_date)
         if existing:
             if overwrite:
+                # Preserve existing file paths for reuse
+                existing_pdf_path = existing.pdf_path
+                existing_json_path = existing.json_path
                 # Delete existing document (cascade will delete related data)
+                # But DON'T delete files - we'll reuse or update them
                 self.session.delete(existing)
                 self.session.flush()
             else:
@@ -147,6 +156,26 @@ class DatabaseService:
             code_postal=immeuble_info.get("code_postal"),
         )
 
+        # Compute storage paths for PDF and JSON
+        pdf_path = None
+        json_path = None
+        if pdf_content is not None:
+            # New PDF provided - compute new paths
+            pdf_path, json_path = storage.compute_document_paths(
+                reference=reference,
+                doc_date=doc_date,
+                immeuble_adresse=immeuble.adresse,
+            )
+            # Delete old files if paths are different
+            if existing_pdf_path and existing_pdf_path != pdf_path:
+                storage.delete_document_files(existing_pdf_path, None)
+            if existing_json_path and existing_json_path != json_path:
+                storage.delete_document_files(None, existing_json_path)
+        elif existing_json_path:
+            # No new PDF but we have existing paths - preserve them
+            pdf_path = existing_pdf_path
+            json_path = existing_json_path
+
         # Create document
         document = Document(
             reference=reference,
@@ -160,10 +189,21 @@ class DatabaseService:
             solde_montant=solde_info.get("montant"),
             solde_type=solde_info.get("type"),
             solde_date_arrete=self._parse_date(solde_info.get("date_arrete")),
+            pdf_path=pdf_path,
+            json_path=json_path,
         )
         self.session.add(document)
         self.session.flush()
 
+        # Save files to storage
+        if pdf_content is not None and pdf_path and json_path:
+            # New PDF provided - save both files
+            storage.save_pdf(pdf_content, pdf_path)
+            storage.save_json(data, json_path)
+        elif json_path:
+            # No new PDF but we have a json_path - update the JSON file
+            storage.save_json(data, json_path)
+
         # Process situation_locataires (revenus)
         for situation in data.get("situation_locataires", []):
             self._save_situation_locataire(document, immeuble, situation)

src/plesna_gerance/database/storage.py

@@ -0,0 +1,280 @@
+"""Service de stockage des fichiers PDF et JSON."""
+
+import json
+import os
+import re
+from datetime import date
+from pathlib import Path
+from typing import Any
+
+
+def _get_project_root() -> Path:
+    """Find project root by looking for pyproject.toml."""
+    current = Path(__file__).resolve()
+    for parent in current.parents:
+        if (parent / "pyproject.toml").exists():
+            return parent
+    return Path.home() / ".plesna_gerance"
+
+
+def get_storage_root() -> Path:
+    """Get storage root from environment or default.
+
+    Returns:
+        Path to the documents storage directory.
+    """
+    env_path = os.environ.get("PLESNA_STORAGE_PATH")
+    if env_path:
+        return Path(env_path)
+    return _get_project_root() / "data" / "documents"
+
+
+def extract_street_letter(adresse: str | None) -> str:
+    """Extract the first letter of the main street name.
+
+    Examples:
+        "4 RUE SERVIENT" -> "S"
+        "33 RUE MARC BLOCH" -> "M"
+        "1 RUE MARIETTON" -> "M"
+        "12 AVENUE JEAN JAURES" -> "J"
+        None -> "X"
+
+    Args:
+        adresse: The address string (e.g., "4 RUE SERVIENT")
+
+    Returns:
+        Single uppercase letter representing the street, or "X" if unknown.
+    """
+    if not adresse:
+        return "X"
+
+    # Normalize: uppercase, remove extra spaces
+    adresse = adresse.upper().strip()
+
+    # Remove common street type prefixes to get to the street name
+    # Pattern: [number] [street type] [street name]
+    street_types = [
+        r"^\d+\s+",  # Leading number
+        r"^RUE\s+",
+        r"^AVENUE\s+",
+        r"^BOULEVARD\s+",
+        r"^PLACE\s+",
+        r"^ALLEE\s+",
+        r"^IMPASSE\s+",
+        r"^CHEMIN\s+",
+        r"^COURS\s+",
+        r"^PASSAGE\s+",
+        r"^QUAI\s+",
+    ]
+
+    remaining = adresse
+    for pattern in street_types:
+        remaining = re.sub(pattern, "", remaining, flags=re.IGNORECASE)
+
+    # Get first letter of what remains
+    remaining = remaining.strip()
+    if remaining and remaining[0].isalpha():
+        return remaining[0].upper()
+
+    return "X"
+
+
+def sanitize_filename(name: str) -> str:
+    """Sanitize a string to be used as filename.
+
+    Args:
+        name: The string to sanitize.
+
+    Returns:
+        Sanitized string safe for filenames.
+    """
+    # Replace problematic characters
+    sanitized = re.sub(r'[<>:"/\\|?*]', "_", name)
+    # Remove leading/trailing spaces and dots
+    sanitized = sanitized.strip(". ")
+    return sanitized
+
+
+def compute_document_paths(
+    reference: str,
+    doc_date: date,
+    immeuble_adresse: str | None = None,
+) -> tuple[str, str]:
+    """Compute the relative paths for PDF and JSON storage.
+
+    Naming convention: {YYYY}/{L}_{reference}_{YYYY-MM-DD}.{ext}
+    Where L is the first letter of the street name.
+
+    Args:
+        reference: Document reference (e.g., "01234")
+        doc_date: Document date
+        immeuble_adresse: Address to extract street letter from
+
+    Returns:
+        Tuple of (pdf_relative_path, json_relative_path)
+    """
+    year = str(doc_date.year)
+    letter = extract_street_letter(immeuble_adresse)
+    date_str = doc_date.strftime("%Y-%m-%d")
+
+    # Sanitize reference for filename
+    safe_ref = sanitize_filename(reference)
+
+    # Build filename: L_reference_date
+    base_name = f"{letter}_{safe_ref}_{date_str}"
+
+    pdf_path = f"{year}/{base_name}.pdf"
+    json_path = f"{year}/{base_name}.json"
+
+    return pdf_path, json_path
+
+
+def save_pdf(
+    pdf_content: bytes,
+    relative_path: str,
+    storage_root: Path | None = None,
+) -> Path:
+    """Save PDF content to storage.
+
+    Args:
+        pdf_content: Binary content of the PDF file.
+        relative_path: Relative path (from compute_document_paths).
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Absolute path to the saved file.
+    """
+    if storage_root is None:
+        storage_root = get_storage_root()
+
+    full_path = storage_root / relative_path
+    full_path.parent.mkdir(parents=True, exist_ok=True)
+
+    full_path.write_bytes(pdf_content)
+    return full_path
+
+
+def save_json(
+    data: dict[str, Any],
+    relative_path: str,
+    storage_root: Path | None = None,
+) -> Path:
+    """Save JSON data to storage.
+
+    Args:
+        data: Dictionary to serialize as JSON.
+        relative_path: Relative path (from compute_document_paths).
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Absolute path to the saved file.
+    """
+    if storage_root is None:
+        storage_root = get_storage_root()
+
+    full_path = storage_root / relative_path
+    full_path.parent.mkdir(parents=True, exist_ok=True)
+
+    with open(full_path, "w", encoding="utf-8") as f:
+        json.dump(data, f, ensure_ascii=False, indent=2, default=str)
+
+    return full_path
+
+
+def get_absolute_path(relative_path: str, storage_root: Path | None = None) -> Path:
+    """Get absolute path from relative storage path.
+
+    Args:
+        relative_path: Relative path stored in database.
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Absolute path to the file.
+    """
+    if storage_root is None:
+        storage_root = get_storage_root()
+    return storage_root / relative_path
+
+
+def file_exists(relative_path: str, storage_root: Path | None = None) -> bool:
+    """Check if a file exists in storage.
+
+    Args:
+        relative_path: Relative path stored in database.
+        storage_root: Optional custom storage root.
+
+    Returns:
+        True if file exists.
+    """
+    return get_absolute_path(relative_path, storage_root).exists()
+
+
+def read_pdf(relative_path: str, storage_root: Path | None = None) -> bytes:
+    """Read PDF content from storage.
+
+    Args:
+        relative_path: Relative path stored in database.
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Binary content of the PDF.
+
+    Raises:
+        FileNotFoundError: If file doesn't exist.
+    """
+    full_path = get_absolute_path(relative_path, storage_root)
+    return full_path.read_bytes()
+
+
+def read_json(relative_path: str, storage_root: Path | None = None) -> dict[str, Any]:
+    """Read JSON data from storage.
+
+    Args:
+        relative_path: Relative path stored in database.
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Parsed JSON data.
+
+    Raises:
+        FileNotFoundError: If file doesn't exist.
+    """
+    full_path = get_absolute_path(relative_path, storage_root)
+    with open(full_path, "r", encoding="utf-8") as f:
+        return json.load(f)
+
+
+def delete_document_files(
+    pdf_path: str | None,
+    json_path: str | None,
+    storage_root: Path | None = None,
+) -> tuple[bool, bool]:
+    """Delete document files from storage.
+
+    Args:
+        pdf_path: Relative path to PDF (can be None).
+        json_path: Relative path to JSON (can be None).
+        storage_root: Optional custom storage root.
+
+    Returns:
+        Tuple of (pdf_deleted, json_deleted) booleans.
+    """
+    if storage_root is None:
+        storage_root = get_storage_root()
+
+    pdf_deleted = False
+    json_deleted = False
+
+    if pdf_path:
+        pdf_full = storage_root / pdf_path
+        if pdf_full.exists():
+            pdf_full.unlink()
+            pdf_deleted = True
+
+    if json_path:
+        json_full = storage_root / json_path
+        if json_full.exists():
+            json_full.unlink()
+            json_deleted = True
+
+    return pdf_deleted, json_deleted