feat(translation): quality pipeline overhaul + new features (audit 2026-08-29)

Translation quality & format preservation:
- Word: merge adjacent same-format runs into one unit (sentence-level
  coherence like inline-tag handling); translate comments/balloons;
  dedupe textbox collection (was translated twice); RTL no longer
  overrides center/justify alignment; CJK/Arabic font hints (eastAsia/cs)
- PPTX: chart translations now actually reach the output file
  (ChartPart.blob is read-only — rewrite chart XML in the saved ZIP);
  CJK typeface hints (a:ea)
- Excel: sheet renames no longer break references — rewrite cell
  formulas (3D/quoted), defined names, data validations, cond. formats
- PDF: bold/italic honored (hebo/heit/hebi); table cells never merge;
  unchanged blocks left untouched (typography preserved, fixes duplicate
  hyperlinks); attempted/changed stats + route gate now cover PDF;
  CJK font paths; scanned PDFs via Mistral OCR (detection + admin settings)

Features:
- formality param (formal/informal) + automatic regional-variant prompts
- output_mode=bilingual docx (source above translation)
- per-user translation memory on Redis (falls back to LRU), context-hashed
- QA report + 0-100 confidence score in job status; L0 on by default
- OpenAI-compatible providers: whole chunk in ONE numbered-JSON request
  (~15x fewer calls) with per-item fallback; base prompt always present
  (custom prompt no longer replaces translation instructions)

Infra & marketing alignment:
- plan-based engine gating + vision gating (closes paid-engine leak);
  /providers/available filtered per plan; 107 languages exposed
- zh-CN/zh-TW validation fixed; libmagic disabled on Windows (native crash)
- admin: Mistral OCR settings + engine status dashboard; httpx<0.28 pin
  (TestClient breakage); Prometheus test fixture fixed
- marketing docs aligned with code (PDF+OCR, retention, engines, pricing)
- security: .env.ionos/.env.production/provider_settings.json removed

Tests: 1173 passed / 0 failed (6 network tests deselected: free Google
endpoint temporarily blocked from this machine)

translators/excel_translator.py

@@ -107,6 +107,7 @@ class ExcelTranslator:
         self._provider = provider
         self.formula_pattern = re.compile(r"=.*")
         self._custom_prompt: Optional[str] = None
+        self._tm_scope = None  # set via set_tm_scope (per-user translation memory)
         self._translation_stats = {"attempted": 0, "changed": 0}
 
     def set_provider(self, provider: TranslationProvider) -> None:
@@ -116,6 +117,14 @@ class ExcelTranslator:
     def set_custom_prompt(self, prompt: Optional[str]) -> None:
         """Set custom system prompt for LLM providers."""
         self._custom_prompt = prompt
+    def set_tm_scope(self, user_id, prompt=None) -> None:
+        """Enable the per-user translation memory for this job."""
+        from services.translation_tm import TMScope
+
+        self._tm_scope = TMScope.from_prompt(
+            user_id, prompt or getattr(self, "_custom_prompt", None)
+        )
+
 
     def translate_file(
         self,
@@ -318,6 +327,13 @@ class ExcelTranslator:
                             new_name=new_name,
                         )
 
+                # openpyxl does NOT rewrite references on rename: cell
+                # formulas, defined names and chart refs pointing at the old
+                # sheet would break (#REF!/#NAME?). Charts are handled later
+                # via ZIP re-injection; fix cells + defined names here.
+                if sheet_name_mapping:
+                    self._rewrite_sheet_refs_in_workbook(workbook, sheet_name_mapping)
+
             if translate_images:
                 _log_info("excel_image_translation_start", sheets=len(workbook.sheetnames))
                 for sheet_name in workbook.sheetnames:
@@ -427,12 +443,30 @@ class ExcelTranslator:
         non_empty = [t for t in texts if t and t.strip()]
         self._translation_stats["attempted"] += len(non_empty)
 
+        from services.translation_tm import translate_with_tm
+
+        provider_name = (
+            self._provider.get_name() if hasattr(self._provider, "get_name")
+            else type(self._provider).__name__
+        ) if self._provider is not None else "legacy"
+
         if self._provider is not None:
-            translated = self._translate_with_provider(
-                texts, target_language, source_language
-            )
+            def _do_translate(miss_texts):
+                return self._translate_with_provider(
+                    miss_texts, target_language, source_language
+                )
         else:
-            translated = self._translate_with_legacy(texts, target_language, source_language)
+            def _do_translate(miss_texts):
+                return self._translate_with_legacy(
+                    miss_texts, target_language, source_language
+                )
+
+        # Translation memory: reuse this user's previous translations
+        # (identical context/prompt) before hitting the provider.
+        translated = translate_with_tm(
+            texts, target_language, source_language,
+            provider_name, getattr(self, "_tm_scope", None), _do_translate,
+        )
 
         changed = sum(1 for orig, trans in zip(texts, translated) if orig != trans and trans.strip())
         self._translation_stats["changed"] += changed
@@ -931,6 +965,140 @@ class ExcelTranslator:
                 rewritten += 1
         return rewritten
 
+    @staticmethod
+    def _rewrite_sheet_refs_in_formula(
+        formula: str, sheet_name_mapping: Dict[str, str]
+    ) -> str:
+        """Rewrite every sheet reference inside a formula string.
+
+        Handles quoted ('Ventes 2026'!), unquoted (Ventes!) and 3D refs
+        (Sheet1!A1:Sheet2!B2 — each end is rewritten independently).
+        Longest names are replaced first so a name that is a prefix of
+        another is not corrupted.
+        """
+        if not formula or not sheet_name_mapping or "!" not in formula:
+            return formula
+
+        result = formula
+        # longest first to avoid partial-name collisions
+        for old in sorted(sheet_name_mapping, key=len, reverse=True):
+            new = sheet_name_mapping[old]
+            if new == old:
+                continue
+            new_quoted = ExcelTranslator._quote_sheet_name_for_ref(new)
+            # Quoted form: inner apostrophes are escaped as ''
+            escaped = old.replace("'", "''")
+            result = re.sub(
+                rf"'{re.escape(escaped)}'!",
+                new_quoted + "!",
+                result,
+            )
+            # Unquoted form: only when the old name needs no quotes; guard
+            # against matching the tail of a longer name.
+            if re.fullmatch(r"[A-Za-z_][A-Za-z0-9_.]*", old):
+                result = re.sub(
+                    rf"(?<![A-Za-z0-9_.']){re.escape(old)}!",
+                    new_quoted + "!",
+                    result,
+                )
+        return result
+
+    @classmethod
+    def _rewrite_sheet_refs_in_workbook(
+        cls, workbook, sheet_name_mapping: Dict[str, str]
+    ) -> None:
+        """After a sheet rename, fix every reference that still points at the
+        old names: cell formulas, defined names, data validations and
+        conditional-formatting rules. openpyxl does none of this on rename.
+
+        Best-effort: individual failures are logged and skipped — a broken
+        rename must never lose the whole file.
+        """
+        if not sheet_name_mapping:
+            return
+
+        cells_fixed = names_fixed = validations_fixed = cf_fixed = 0
+
+        try:
+            for worksheet in workbook.worksheets:
+                # 1) Cell formulas
+                for row in worksheet.iter_rows():
+                    for cell in row:
+                        value = cell.value
+                        if isinstance(value, str) and value.startswith("=") and "!" in value:
+                            updated = cls._rewrite_sheet_refs_in_formula(
+                                value, sheet_name_mapping
+                            )
+                            if updated != value:
+                                cell.value = updated
+                                cells_fixed += 1
+
+                # 2) Data validations (list sources / custom formulas)
+                try:
+                    for dv in worksheet.data_validations.dataValidation:
+                        for attr in ("formula1", "formula2"):
+                            fx = getattr(dv, attr, None)
+                            if isinstance(fx, str) and "!" in fx:
+                                updated = cls._rewrite_sheet_refs_in_formula(
+                                    fx, sheet_name_mapping
+                                )
+                                if updated != fx:
+                                    setattr(dv, attr, updated)
+                                    validations_fixed += 1
+                except Exception as e:
+                    _log_warning(
+                        "excel_sheet_refs_validations_failed",
+                        sheet=worksheet.title,
+                        error=str(e),
+                    )
+
+                # 3) Conditional formatting rules
+                try:
+                    for cf in worksheet.conditional_formatting:
+                        for rule in cf.rules:
+                            formulas = getattr(rule, "formula", None) or []
+                            for idx, fx in enumerate(formulas):
+                                if isinstance(fx, str) and "!" in fx:
+                                    updated = cls._rewrite_sheet_refs_in_formula(
+                                        fx, sheet_name_mapping
+                                    )
+                                    if updated != fx:
+                                        formulas[idx] = updated
+                                        cf_fixed += 1
+                except Exception as e:
+                    _log_warning(
+                        "excel_sheet_refs_condfmt_failed",
+                        sheet=worksheet.title,
+                        error=str(e),
+                    )
+
+            # 4) Workbook-level defined names
+            try:
+                for name in list(workbook.defined_names.values()):
+                    attr_text = getattr(name, "attr_text", None)
+                    if isinstance(attr_text, str) and "!" in attr_text:
+                        updated = cls._rewrite_sheet_refs_in_formula(
+                            attr_text, sheet_name_mapping
+                        )
+                        if updated != attr_text:
+                            name.attr_text = updated
+                            names_fixed += 1
+            except Exception as e:
+                _log_warning("excel_sheet_refs_names_failed", error=str(e))
+
+        except Exception as e:
+            _log_error("excel_sheet_refs_rewrite_failed", error=str(e))
+            return
+
+        if cells_fixed or names_fixed or validations_fixed or cf_fixed:
+            _log_info(
+                "excel_sheet_refs_rewritten",
+                cells=cells_fixed,
+                defined_names=names_fixed,
+                validations=validations_fixed,
+                conditional_formats=cf_fixed,
+            )
+
     def _translate_images(self, worksheet: Worksheet, target_language: str) -> None:
         """
         Translate text in images using vision model.