docvortex 0.2.1__py3-none-any.whl

docvortex/analyzers/native/office/xlsx/package_normalizer.py

@@ -0,0 +1,310 @@
+from io import BytesIO
+import re
+from zipfile import BadZipFile, ZipFile, ZipInfo
+import xml.etree.ElementTree as ET
+
+from ..opc import write_zip_package
+
+SPREADSHEETML_NS = "http://schemas.openxmlformats.org/spreadsheetml/2006/main"
+SHARED_STRINGS_PATH = "xl/sharedStrings.xml"
+STYLES_PATH = "xl/styles.xml"
+WORKSHEET_PREFIX = "xl/worksheets/"
+WORKSHEET_SUFFIX = ".xml"
+UNDERLINE_TAG = f"{{{SPREADSHEETML_NS}}}u"
+FILLS_TAG = f"{{{SPREADSHEETML_NS}}}fills"
+FILL_TAG = f"{{{SPREADSHEETML_NS}}}fill"
+PATTERN_FILL_TAG = f"{{{SPREADSHEETML_NS}}}patternFill"
+AUTOFILTER_TAG = f"{{{SPREADSHEETML_NS}}}autoFilter"
+DIMENSION_TAG = f"{{{SPREADSHEETML_NS}}}dimension"
+SHEET_DATA_TAG = f"{{{SPREADSHEETML_NS}}}sheetData"
+CELL_TAG = f"{{{SPREADSHEETML_NS}}}c"
+UNDERLINE_VAL_ATTRS = ("val", f"{{{SPREADSHEETML_NS}}}val")
+VALID_UNDERLINE_VALUES = {
+    "single",
+    "double",
+    "singleAccounting",
+    "doubleAccounting",
+    "none",
+}
+ROW_ONLY_RANGE_RE = re.compile(r"^\$?([1-9][0-9]*):\$?([1-9][0-9]*)$")
+CELL_REF_RE = re.compile(r"^\$?([A-Za-z]{1,3})\$?[1-9][0-9]*$")
+CELL_RANGE_RE = re.compile(r"^\$?([A-Za-z]{1,3})\$?[1-9][0-9]*:\$?([A-Za-z]{1,3})\$?[1-9][0-9]*$")
+WHOLE_COLUMN_RANGE_RE = re.compile(r"^\$?([A-Za-z]{1,3}):\$?([A-Za-z]{1,3})$")
+MAX_EXCEL_COLUMN = "XFD"
+
+
+def normalize_xlsx_package(file_bytes: bytes) -> bytes:
+    """在进入 openpyxl 前修复常见 XLSX 包级兼容问题。"""
+    try:
+        with ZipFile(BytesIO(file_bytes)) as source:
+            rewritten_members: list[tuple[ZipInfo, bytes]] = []
+            changed = False
+
+            for info in source.infolist():
+                member_data = source.read(info.filename)
+                normalized_data = _normalize_xlsx_member(info.filename, member_data)
+                if normalized_data != member_data:
+                    changed = True
+                member_data = normalized_data
+                rewritten_members.append((info, member_data))
+    except BadZipFile as exc:
+        raise ValueError("Invalid XLSX package: file is not a ZIP archive.") from exc
+
+    if not changed:
+        return file_bytes
+
+    return write_zip_package(rewritten_members)
+
+
+def strip_xlsx_ole_objects_for_openpyxl(file_bytes: bytes) -> bytes:
+    """仅为 openpyxl 读取副本移除会被误判为列定义的 oleObjects。"""
+
+    try:
+        with ZipFile(BytesIO(file_bytes)) as source:
+            rewritten_members: list[tuple[ZipInfo, bytes]] = []
+            changed = False
+            for info in source.infolist():
+                member_data = source.read(info.filename)
+                if _is_worksheet_xml(info.filename):
+                    normalized_data = _remove_worksheet_ole_objects(member_data)
+                    changed = changed or normalized_data != member_data
+                    member_data = normalized_data
+                rewritten_members.append((info, member_data))
+    except BadZipFile as exc:
+        raise ValueError("Invalid XLSX package: file is not a ZIP archive.") from exc
+    return write_zip_package(rewritten_members) if changed else file_bytes
+
+
+def _remove_worksheet_ole_objects(xml_bytes: bytes) -> bytes:
+    """删除 worksheet 的 oleObjects 容器，原始 ZIP 仍由公式解析器持有。"""
+
+    try:
+        root = ET.fromstring(xml_bytes)
+    except ET.ParseError:
+        return xml_bytes
+    changed = False
+    for child in list(root):
+        if child.tag.rsplit("}", 1)[-1] != "oleObjects":
+            continue
+        root.remove(child)
+        changed = True
+    if not changed:
+        return xml_bytes
+    return ET.tostring(root, encoding="utf-8", xml_declaration=True)
+
+
+def _normalize_xlsx_member(member_name: str, member_data: bytes) -> bytes:
+    """根据 XLSX 包内成员路径分发 XML 兼容性规范化逻辑。"""
+    if member_name == SHARED_STRINGS_PATH:
+        return _normalize_shared_strings_xml(member_data)
+    if member_name == STYLES_PATH:
+        return _normalize_styles_xml(member_data)
+    if _is_worksheet_xml(member_name):
+        return _normalize_worksheet_xml(member_data)
+    return member_data
+
+
+def _normalize_shared_strings_xml(xml_bytes: bytes) -> bytes:
+    """规范化共享字符串 XML 中 openpyxl 无法接受的富文本属性。"""
+    try:
+        root = ET.fromstring(xml_bytes)
+    except ET.ParseError:
+        return xml_bytes
+
+    changed = False
+    for underline in root.findall(f".//{UNDERLINE_TAG}"):
+        if _drop_blank_underline_value(underline):
+            changed = True
+
+    if not changed:
+        return xml_bytes
+
+    return ET.tostring(root, encoding="utf-8", xml_declaration=True)
+
+
+def _normalize_styles_xml(xml_bytes: bytes) -> bytes:
+    """规范化 styles.xml 中 openpyxl 无法接受的空 fill 节点。"""
+    try:
+        root = ET.fromstring(xml_bytes)
+    except ET.ParseError:
+        return xml_bytes
+
+    changed = False
+    fills = root.find(FILLS_TAG)
+    if fills is not None:
+        for fill in fills.findall(FILL_TAG):
+            if _ensure_fill_has_pattern(fill):
+                changed = True
+
+    if not changed:
+        return xml_bytes
+
+    return ET.tostring(root, encoding="utf-8", xml_declaration=True)
+
+
+def _normalize_worksheet_xml(xml_bytes: bytes) -> bytes:
+    """规范化 worksheet XML 中会阻断 openpyxl 加载的行范围筛选器。"""
+    try:
+        root = ET.fromstring(xml_bytes)
+    except ET.ParseError:
+        return xml_bytes
+
+    changed = False
+    for auto_filter in root.findall(f".//{AUTOFILTER_TAG}"):
+        if _normalize_auto_filter_ref(root, auto_filter):
+            changed = True
+
+    if not changed:
+        return xml_bytes
+
+    return ET.tostring(root, encoding="utf-8", xml_declaration=True)
+
+
+def _drop_blank_underline_value(underline: ET.Element) -> bool:
+    """删除空白 underline val，保留 <u/> 的默认单下划线语义。"""
+    for attr_name in UNDERLINE_VAL_ATTRS:
+        attr_value = underline.attrib.get(attr_name)
+        if attr_value is None:
+            continue
+        if attr_value.strip() == "":
+            del underline.attrib[attr_name]
+            return True
+        if attr_value not in VALID_UNDERLINE_VALUES:
+            return False
+    return False
+
+
+def _ensure_fill_has_pattern(fill: ET.Element) -> bool:
+    """为空 fill 补一个空 patternFill，保留原 fillId 顺序和样式引用。"""
+    if len(fill):
+        return False
+
+    fill.append(ET.Element(PATTERN_FILL_TAG))
+    return True
+
+
+def _normalize_auto_filter_ref(root: ET.Element, auto_filter: ET.Element) -> bool:
+    """将 autoFilter 的纯行范围 ref 扩展为 openpyxl 可接受的单元格范围。"""
+    ref = auto_filter.attrib.get("ref")
+    if not ref:
+        return False
+
+    rows = _parse_row_only_range(ref)
+    if rows is None:
+        return False
+
+    min_col, max_col = _resolve_worksheet_column_bounds(root)
+    auto_filter.set("ref", f"{min_col}{rows[0]}:{max_col}{rows[1]}")
+    return True
+
+
+def _parse_row_only_range(ref: str) -> tuple[int, int] | None:
+    """解析 Excel 行范围写法，仅接受正整数且起止顺序有效的范围。"""
+    match = ROW_ONLY_RANGE_RE.match(ref)
+    if match is None:
+        return None
+
+    min_row, max_row = int(match.group(1)), int(match.group(2))
+    if min_row > max_row:
+        return None
+    return min_row, max_row
+
+
+def _resolve_worksheet_column_bounds(root: ET.Element) -> tuple[str, str]:
+    """优先从 dimension 获取列边界，失败时退回扫描 sheetData。"""
+    dimension = root.find(DIMENSION_TAG)
+    if dimension is not None:
+        columns = _column_bounds_from_ref(dimension.attrib.get("ref", ""))
+        if columns is not None:
+            return columns
+
+    columns = _column_bounds_from_sheet_data(root)
+    if columns is not None:
+        return columns
+
+    return "A", MAX_EXCEL_COLUMN
+
+
+def _column_bounds_from_ref(ref: str) -> tuple[str, str] | None:
+    """从 dimension/ref 这类范围字符串中解析起止列。"""
+    ref = ref.strip()
+    if not ref:
+        return None
+
+    range_match = CELL_RANGE_RE.match(ref)
+    if range_match is not None:
+        return (
+            range_match.group(1).upper(),
+            range_match.group(2).upper(),
+        )
+
+    cell_match = CELL_REF_RE.match(ref)
+    if cell_match is not None:
+        column = cell_match.group(1).upper()
+        return column, column
+
+    whole_column_match = WHOLE_COLUMN_RANGE_RE.match(ref)
+    if whole_column_match is not None:
+        return (
+            whole_column_match.group(1).upper(),
+            whole_column_match.group(2).upper(),
+        )
+
+    return None
+
+
+def _column_bounds_from_sheet_data(root: ET.Element) -> tuple[str, str] | None:
+    """扫描 sheetData 里的单元格引用，推导工作表真实列范围。"""
+    sheet_data = root.find(SHEET_DATA_TAG)
+    if sheet_data is None:
+        return None
+
+    column_indexes = []
+    for cell in sheet_data.findall(f".//{CELL_TAG}"):
+        column = _column_from_cell_ref(cell.attrib.get("r", ""))
+        if column is not None:
+            column_indexes.append(_column_to_index(column))
+
+    if not column_indexes:
+        return None
+
+    return (
+        _index_to_column(min(column_indexes)),
+        _index_to_column(max(column_indexes)),
+    )
+
+
+def _column_from_cell_ref(cell_ref: str) -> str | None:
+    """从 A1 形式的单元格引用中提取列名。"""
+    match = CELL_REF_RE.match(cell_ref)
+    if match is None:
+        return None
+    return match.group(1).upper()
+
+
+def _column_to_index(column: str) -> int:
+    """将 Excel 列名转换为 1-based 数字索引。"""
+    index = 0
+    for char in column.upper():
+        if not "A" <= char <= "Z":
+            return 0
+        index = index * 26 + ord(char) - ord("A") + 1
+    return index
+
+
+def _index_to_column(index: int) -> str:
+    """将 1-based 数字索引转换为 Excel 列名。"""
+    column = []
+    while index:
+        index, remainder = divmod(index - 1, 26)
+        column.append(chr(ord("A") + remainder))
+    return "".join(reversed(column))
+
+
+def _is_worksheet_xml(member_name: str) -> bool:
+    """判断包内成员是否为普通 worksheet XML。"""
+    return (
+        member_name.startswith(WORKSHEET_PREFIX)
+        and member_name.endswith(WORKSHEET_SUFFIX)
+        and not member_name.startswith(f"{WORKSHEET_PREFIX}_rels/")
+    )