pysfi 0.1.13__py3-none-any.whl → 0.1.15__py3-none-any.whl

sfi/docscan/docscan.py

@@ -91,7 +91,9 @@ def t(key: str, **kwargs) -> str:
     Returns:
         Translated text
     """
-    text = ZH_TRANSLATIONS.get(key, key) if USE_CHINESE else EN_TRANSLATIONS.get(key, key)
+    text = (
+        ZH_TRANSLATIONS.get(key, key) if USE_CHINESE else EN_TRANSLATIONS.get(key, key)
+    )
 
     # Format with kwargs if provided
     if kwargs:
@@ -123,7 +125,9 @@ class Rule:
                 # Use re.ASCII for faster matching when possible
                 self.compiled_pattern = re.compile(self.pattern, flags | re.ASCII)
             except re.error as e:
-                logger.warning(t("invalid_regex_pattern", pattern=self.pattern, error=e))
+                logger.warning(
+                    t("invalid_regex_pattern", pattern=self.pattern, error=e)
+                )
                 self.compiled_pattern = None
         else:
             self.compiled_pattern = None
@@ -274,13 +278,18 @@ class DocumentScanner:
                 "use_pdf_ocr": self.use_pdf_ocr,
                 "use_process_pool": self.use_process_pool,
             },
-            "rules": [{"name": r.name, "pattern": r.pattern, "is_regex": r.is_regex} for r in self.rules],
+            "rules": [
+                {"name": r.name, "pattern": r.pattern, "is_regex": r.is_regex}
+                for r in self.rules
+            ],
             "matches": [],
         }
 
         # Scan files in parallel
         processed = 0
-        executor_class = ProcessPoolExecutor if self.use_process_pool else ThreadPoolExecutor
+        executor_class = (
+            ProcessPoolExecutor if self.use_process_pool else ThreadPoolExecutor
+        )
         executor = executor_class(max_workers=threads)
         self._executor = executor  # Keep reference for forced shutdown
 
@@ -350,10 +359,17 @@ class DocumentScanner:
                     break
 
                 try:
-                    file_result = future.result(timeout=1.0)  # Short timeout to allow quick stop
+                    file_result = future.result(
+                        timeout=1.0
+                    )  # Short timeout to allow quick stop
                     if file_result and file_result["matches"]:
                         results["matches"].append(file_result)
-                        logger.info(t("found_matches_in_file", file_name=Path(file_result.get("file_path", "")).name))
+                        logger.info(
+                            t(
+                                "found_matches_in_file",
+                                file_name=Path(file_result.get("file_path", "")).name,
+                            )
+                        )
                 except TimeoutError:
                     logger.warning(t("task_timeout_scan_may_be_stopping"))
                     if self.stopped:
@@ -366,7 +382,9 @@ class DocumentScanner:
 
                 # Report progress
                 if show_progress and processed % 10 == 0:
-                    logger.info(t("progress_report", processed=processed, total=len(files)))
+                    logger.info(
+                        t("progress_report", processed=processed, total=len(files))
+                    )
 
                 # Call progress callback if set
                 if self._progress_callback:
@@ -391,7 +409,9 @@ class DocumentScanner:
         if self.stopped:
             logger.info(t("scan_stopped_processed_files", processed=processed))
         else:
-            logger.info(t("scan_complete_found_matches", matches_count=len(results["matches"])))
+            logger.info(
+                t("scan_complete_found_matches", matches_count=len(results["matches"]))
+            )
 
         return results
 
@@ -493,7 +513,9 @@ class DocumentScanner:
                     return {}
 
         except Exception as e:
-            logger.warning(t("could_not_extract_text_from_file", file_path=file_path, error=e))
+            logger.warning(
+                t("could_not_extract_text_from_file", file_path=file_path, error=e)
+            )
             return {}
 
         processing_time = time.perf_counter() - file_start_time
@@ -549,14 +571,18 @@ class DocumentScanner:
             try:
                 return self._extract_pdf_fitz(file_path)
             except Exception as e:
-                logger.warning(t("pymupdf_failed_for_file", file_name=file_path.name, error=e))
+                logger.warning(
+                    t("pymupdf_failed_for_file", file_name=file_path.name, error=e)
+                )
 
         # Fallback to pypdf
         if pypdf is not None:
             try:
                 return self._extract_pdf_pypdf(file_path)
             except Exception as e:
-                logger.error(t("pypdf_also_failed_for_file", file_name=file_path.name, error=e))
+                logger.error(
+                    t("pypdf_also_failed_for_file", file_name=file_path.name, error=e)
+                )
                 return "", {}
 
         logger.warning(t("no_pdf_library_installed"))
@@ -632,7 +658,9 @@ class DocumentScanner:
         except Exception as e:
             if doc:
                 doc.close()
-            logger.warning(t("pymupdf_error_trying_fallback", file_path=file_path, error=e))
+            logger.warning(
+                t("pymupdf_error_trying_fallback", file_path=file_path, error=e)
+            )
             # Re-raise to trigger fallback to pypdf
             raise
 
@@ -764,8 +792,12 @@ class DocumentScanner:
                     text_parts.append(text)
 
             metadata = {
-                "title": book.get_metadata("DC", "title")[0][0] if book.get_metadata("DC", "title") else "",  # pyright: ignore[reportAttributeAccessIssue]
-                "author": book.get_metadata("DC", "creator")[0][0] if book.get_metadata("DC", "creator") else "",  # pyright: ignore[reportAttributeAccessIssue]
+                "title": book.get_metadata("DC", "title")[0][0]
+                if book.get_metadata("DC", "title")
+                else "",  # pyright: ignore[reportAttributeAccessIssue]
+                "author": book.get_metadata("DC", "creator")[0][0]
+                if book.get_metadata("DC", "creator")
+                else "",  # pyright: ignore[reportAttributeAccessIssue]
                 "format": "EPUB",
             }
 
@@ -810,7 +842,9 @@ class DocumentScanner:
             root = tree.getroot()
 
             # Extract all text content
-            text_parts = [elem.text for elem in root.iter() if elem.text and elem.text.strip()]
+            text_parts = [
+                elem.text for elem in root.iter() if elem.text and elem.text.strip()
+            ]
             text = "\n".join(text_parts)
 
             metadata = {
@@ -954,7 +988,9 @@ class DocumentScanner:
                         wb.close()
                         return "", {}
 
-                row_text = " | ".join(str(cell) if cell is not None else "" for cell in row)
+                row_text = " | ".join(
+                    str(cell) if cell is not None else "" for cell in row
+                )
                 if row_text.strip():
                     text_parts.append(row_text)
 
@@ -1017,7 +1053,9 @@ class DocumentScanner:
 
             return text, metadata
         except Exception as e:
-            logger.warning(t("could_not_perform_ocr_on_file", file_path=file_path, error=e))
+            logger.warning(
+                t("could_not_perform_ocr_on_file", file_path=file_path, error=e)
+            )
             return "", {}
 
     def _extract_text(self, file_path: Path) -> tuple[str, dict[str, Any]]:
@@ -1047,8 +1085,12 @@ def main():
     USE_CHINESE = temp_args.lang == "zh"
 
     parser = argparse.ArgumentParser(description=t("document_scanner_description"))
-    parser.add_argument("input", type=str, nargs="?", default=str(cwd), help=t("input_directory_help"))
-    parser.add_argument("-r", "--rules", type=str, default="rules.json", help=t("rules_file_help"))
+    parser.add_argument(
+        "input", type=str, nargs="?", default=str(cwd), help=t("input_directory_help")
+    )
+    parser.add_argument(
+        "-r", "--rules", type=str, default="rules.json", help=t("rules_file_help")
+    )
     parser.add_argument("--recursive", action="store_true", help=t("recursive_help"))
     parser.add_argument(
         "-f",
@@ -1056,7 +1098,9 @@ def main():
         help=t("file_types_help"),
         default="pdf,docx,xlsx,pptx,txt,odt,rtf,epub,csv,xml,html,md,jpg,jpeg,png,gif,bmp,tiff",
     )
-    parser.add_argument("--use-pdf-ocr", help=t("use_pdf_ocr_help"), action="store_true")
+    parser.add_argument(
+        "--use-pdf-ocr", help=t("use_pdf_ocr_help"), action="store_true"
+    )
     parser.add_argument(
         "--use-process-pool",
         help=t("use_process_pool_help"),
@@ -1074,7 +1118,9 @@ def main():
     parser.add_argument("-v", "--verbose", help=t("verbose_help"), action="store_true")
 
     # 添加语言参数
-    parser.add_argument("--lang", help=t("language_help"), choices=["en", "zh"], default="zh")
+    parser.add_argument(
+        "--lang", help=t("language_help"), choices=["en", "zh"], default="zh"
+    )
 
     args = parser.parse_args()
 
@@ -1129,11 +1175,20 @@ def main():
     file_types = [ft.strip() for ft in args.file_types.split(",")]
 
     # Create scanner and run scan
-    scanner = DocumentScanner(input_dir, rules, file_types, args.use_pdf_ocr, args.use_process_pool, args.batch_size)
+    scanner = DocumentScanner(
+        input_dir,
+        rules,
+        file_types,
+        args.use_pdf_ocr,
+        args.use_process_pool,
+        args.batch_size,
+    )
     results = scanner.scan(threads=args.threads, show_progress=args.progress)
 
     # Save results to JSON file in input directory
-    output_file = input_dir / f"scan_results_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+    output_file = (
+        input_dir / f"scan_results_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+    )
     with open(output_file, "w", encoding="utf-8") as f:
         json.dump(results, f, indent=2, ensure_ascii=False)
 

sfi/docscan/docscan_gui.py

@@ -45,14 +45,14 @@ except ImportError:
     try:
         from sfi.docscan.docscan import DocumentScanner, Rule
     except ImportError:
-        from src.docscan.docscan import DocumentScanner, Rule
+        from docscan.docscan import DocumentScanner, Rule
 
 # Import translations
 try:
     from sfi.docscan.lang.zhcn import TRANSLATIONS
 except ImportError:
     try:
-        from src.docscan.lang.zhcn import TRANSLATIONS
+        from docscan.lang.zhcn import TRANSLATIONS
     except ImportError:
         TRANSLATIONS = {}
 
@@ -413,15 +413,15 @@ class SettingsDialog(QDialog):
             if item.widget() and isinstance(item.widget(), QGroupBox):
                 group_box = item.widget()
                 if "Language" in group_box.title():  # type: ignore
-                    group_box.setTitle(
+                    group_box.setWindowTitle(
                         t("language_settings", default="Language Settings")
                     )  # type: ignore
                 elif "Processing" in group_box.title():  # type: ignore
-                    group_box.setTitle(
+                    group_box.setWindowTitle(
                         t("processing_options", default="Processing Options")
                     )  # type: ignore
                 elif "Performance" in group_box.title():  # type: ignore
-                    group_box.setTitle(
+                    group_box.setWindowTitle(
                         t("performance_settings", default="Performance Settings")
                     )  # type: ignore
 

sfi/filedate/filedate.py

@@ -11,11 +11,18 @@ import time
 from dataclasses import dataclass
 from functools import cached_property, lru_cache
 from pathlib import Path
-
-DETECT_SEPARATORS: str = "-_#.~"
-SEP: str = "_"
-MAX_RETRY: int = 100
-DATE_PATTERN = re.compile(r"(20|19)\d{2}((0[1-9])|(1[012]))((0[1-9])|([12]\d)|(3[01]))")
+from re import Pattern
+from typing import Final
+
+# Configuration constants
+DETECT_SEPARATORS: Final[str] = "-_#.~"
+SEP: Final[str] = "_"
+MAX_RETRY: Final[int] = 100
+
+# Date pattern for detection
+DATE_PATTERN: Final[Pattern[str]] = re.compile(
+    r"(20|19)\d{2}((0[1-9])|(1[012]))((0[1-9])|([12]\d)|(3[01]))"
+)
 
 logging.basicConfig(level=logging.INFO, format="%(message)s")
 logger = logging.getLogger(__name__)

sfi/img2pdf/img2pdf.py

@@ -33,7 +33,7 @@ class ImageToPdfConfig:
     """Configuration for image to PDF conversion."""
 
     DPI: int = 300
-    EXTENSIONS: set[str] = None
+    EXTENSIONS: set[str] | None = None
 
     def __post_init__(self) -> None:
         # Initialize default extensions if not provided
@@ -68,7 +68,7 @@ class ImageToPdfConfig:
     def save(self) -> None:
         """Save current configuration to file."""
         CONFIG_FILE.parent.mkdir(parents=True, exist_ok=True)
-        config_dict = {"DPI": self.DPI, "EXTENSIONS": list(self.EXTENSIONS)}
+        config_dict = {"DPI": self.DPI, "EXTENSIONS": list(self.EXTENSIONS or set())}
         CONFIG_FILE.write_text(json.dumps(config_dict, indent=4), encoding="utf-8")
 
 
@@ -111,7 +111,7 @@ def is_valid_image(file_path: Path) -> bool:
 
     # Extension validation.
     ext = file_path.suffix.lower()
-    if ext not in conf.EXTENSIONS:
+    if not conf.EXTENSIONS or ext not in conf.EXTENSIONS:
         logger.debug(f"Invalid image extension: {ext}, {file_path}")
         return False
 
@@ -337,8 +337,8 @@ class ImageToPDFRunner:
                 del image
             if "rgb_img" in locals():
                 del rgb_img
-        except:
-            pass  # Ignore cleanup errors
+        except Exception:
+            logger.error(f"Cleanup image: {filepath} failed")
         return None
 
     def _auto_rotate_image(self, image: Image.Image) -> Image.Image:

sfi/llmquantize/llmquantize.py

@@ -31,9 +31,14 @@ from PySide2.QtWidgets import (
 
 CONFIG_FILE = Path.home() / ".pysfi" / "llmquantize.json"
 
-logging.basicConfig(level=logging.INFO)
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
+)
 logger = logging.getLogger(__name__)
 
+__version__ = "1.0.0"
+__build__ = "20260204"
+
 
 @dataclass
 class QuantizerConfig:
@@ -44,9 +49,9 @@ class QuantizerConfig:
     WIN_POS: list[int] = None
     LAST_INPUT_FILE: str = ""
     SELECTED_QUANTS: list[str] = None
-    _loaded_from_file: bool = False
 
     def __post_init__(self) -> None:
+        """初始化默认值并加载配置文件."""
         # 初始化默认值
         if self.WIN_SIZE is None:
             self.WIN_SIZE = [600, 500]
@@ -62,12 +67,7 @@ class QuantizerConfig:
                 # 更新实例属性，只更新存在的属性
                 for key, value in config_data.items():
                     if hasattr(self, key):
-                        if key in ["WIN_SIZE", "WIN_POS", "SELECTED_QUANTS"]:
-                            # 对于列表类型，需要特别处理
-                            setattr(self, key, value)
-                        else:
-                            setattr(self, key, value)
-                self._loaded_from_file = True
+                        setattr(self, key, value)
             except (json.JSONDecodeError, TypeError, AttributeError) as e:
                 logger.warning("Failed to load configuration: %s", e)
                 logger.info("Using default configuration")
@@ -75,7 +75,7 @@ class QuantizerConfig:
             logger.info("Using default configuration")
 
     def save(self) -> None:
-        """保存配置."""
+        """保存配置到文件."""
         CONFIG_FILE.parent.mkdir(parents=True, exist_ok=True)
         # 将数据类转换为字典进行JSON序列化
         config_dict = {}
@@ -97,8 +97,11 @@ atexit.register(conf.save)
 def _process_gguf_stem(filename: str) -> str:
     """处理文件名, 移除可能的F16后缀.
 
+    Args:
+        filename: 输入的文件名(不含扩展名)
+
     Returns:
-        str: 处理后的文件名
+        str: 处理后的文件名, 移除了F16后缀(如果存在)
     """
     if filename.upper().endswith("-F16"):
         filename = filename[:-4]  # 移除-F16后缀
@@ -106,7 +109,13 @@ def _process_gguf_stem(filename: str) -> str:
 
 
 class QuantizationWorker(QThread):
-    """量化执行线程Worker."""
+    """量化执行线程Worker.
+
+    Attributes:
+        progress_msg_updated: 进度消息更新信号
+        progress_count_updated: 进度数值更新信号
+        is_finished: 完成信号
+    """
 
     progress_msg_updated = Signal(str)
     progress_count_updated = Signal(int)
@@ -117,6 +126,12 @@ class QuantizationWorker(QThread):
         input_file: pathlib.Path,
         quant_types: list[str],
     ) -> None:
+        """初始化量化Worker.
+
+        Args:
+            input_file: 输入的F16 GGUF文件路径
+            quant_types: 需要转换的量化类型列表
+        """
         super().__init__()
 
         self.input_file = input_file
@@ -139,7 +154,7 @@ class QuantizationWorker(QThread):
                     f"正在转换到 {quant_type} 格式...",
                 )
 
-                # 构建命令行参数（确保所有参数都是字符串）
+                # 构建命令行参数
                 cmd = [
                     "llama-quantize",
                     str(self.input_file.name),
@@ -147,7 +162,7 @@ class QuantizationWorker(QThread):
                     quant_type,
                 ]
 
-                # 执行转换命令（使用 cwd 参数避免全局目录变更）
+                # 执行转换命令
                 try:
                     process = subprocess.Popen(
                         cmd,
@@ -312,7 +327,7 @@ class GGUFQuantizerGUI(QMainWindow):
         self.setCentralWidget(main_widget)
 
     def select_file(self) -> None:
-        """选择文件."""
+        """选择F16 GGUF文件."""
         # 使用上次选择的目录作为初始目录
         initial_dir = ""
         if conf.LAST_INPUT_FILE and pathlib.Path(conf.LAST_INPUT_FILE).exists():
@@ -361,21 +376,25 @@ class GGUFQuantizerGUI(QMainWindow):
             filename = f"{_process_gguf_stem(self.input_file.stem)}-{quant_type}.gguf"
             expected_file = dir_path / filename
             if expected_file.exists():
-                # 文件已存在，标记但不禁用，允许用户选择重新生成
+                # 文件已存在，标记并禁用，防止重复生成
                 self.quant_checks[quant_type].setText(
                     f"{self.quant_types[quant_type]} (已存在)",
                 )
                 self.quant_checks[quant_type].setStyleSheet("color: orange;")
+                self.quant_checks[quant_type].setChecked(True)
+                self.quant_checks[quant_type].setEnabled(False)
             else:
                 self.quant_checks[quant_type].setText(
                     self.quant_types[quant_type],
                 )
                 self.quant_checks[quant_type].setStyleSheet("")
+                self.quant_checks[quant_type].setEnabled(True)
 
     def _scroll_to_bottom(self) -> None:
         """滚动输出框到底部."""
         scrollbar = self.output_text.verticalScrollBar()
-        scrollbar.setValue(scrollbar.maximum())
+        if scrollbar:
+            scrollbar.setValue(scrollbar.maximum())
 
     def on_quant_type_changed(self, _state: int) -> None:
         """量化类型变更时保存配置."""
@@ -397,7 +416,7 @@ class GGUFQuantizerGUI(QMainWindow):
         return super().resizeEvent(event)
 
     def start_conversion(self) -> None:
-        """开始转换."""
+        """开始转换量化任务."""
         # 检查是否已有任务在运行
         if self.worker and self.worker.isRunning():
             self.output_text.append("已有转换任务正在进行, 请等待完成")
@@ -405,7 +424,9 @@ class GGUFQuantizerGUI(QMainWindow):
             return
 
         selected_quants: list[str] = [
-            q for q, check in self.quant_checks.items() if check.isChecked()
+            q
+            for q, check in self.quant_checks.items()
+            if check.isChecked() and check.isEnabled()
         ]
 
         if not selected_quants:
@@ -418,19 +439,8 @@ class GGUFQuantizerGUI(QMainWindow):
             self._scroll_to_bottom()
             return
 
-        # 检查是否有已存在的文件将被覆盖
-        existing_files = []
-        for quant_type in selected_quants:
-            filename = f"{_process_gguf_stem(self.input_file.stem)}-{quant_type}.gguf"
-            expected_file = self.input_file.parent / filename
-            if expected_file.exists():
-                existing_files.append(filename)
-
-        if existing_files:
-            self.output_text.append("警告: 将覆盖以下已存在文件:")
-            for existing_file in existing_files:
-                self.output_text.append(f"  - {existing_file}")
-            self._scroll_to_bottom()
+        # 注意：由于selected_quants只包含启用的复选框，
+        # 所以不会包含已存在的禁用文件，无需额外检查覆盖
 
         self.convert_btn.setEnabled(False)
         self.progress_bar.setValue(0)
@@ -487,6 +497,7 @@ class GGUFQuantizerGUI(QMainWindow):
 
 
 def main() -> None:
+    """主程序入口."""
     app = QApplication(sys.argv)
 
     # 检查是否安装了llama.cpp
@@ -497,8 +508,8 @@ def main() -> None:
             check=False,
         )
     except FileNotFoundError:
-        logger.exception("错误: 未找到llama.cpp/quantize工具")
-        logger.exception(
+        logger.error("错误: 未找到llama.cpp/quantize工具")
+        logger.error(
             "请确保已编译llama.cpp并将quantize工具放在llama.cpp/目录下",
         )
         sys.exit(1)

sfi/llmserver/__init__.py

@@ -1 +1 @@
- 
+