v0.5.6

md2gost/office/uno_helpers.py

@@ -0,0 +1,209 @@
+# -*- coding: utf-8 -*-
+"""Pure table-split helpers for UNO worker (Py2.7 + Py3, standalone, no pywin32)."""
+
+from __future__ import print_function, unicode_literals
+
+import os
+import re
+import sys
+
+try:
+    from urllib.parse import quote
+except ImportError:
+    from urllib import quote as _urllib_quote
+
+    def quote(value, safe="/:@"):
+        if isinstance(value, bytes):
+            value = value.decode("utf-8")
+        return _urllib_quote(value.encode("utf-8"), safe)
+
+
+DOCX_FILTER_CANDIDATES = {
+    "libre": ("Office Open XML Text", "MS Word 2007 XML", "Word 2007 XML"),
+    "openoffice": (
+        "MS Word 2007 XML",
+        "Office Open XML Text",
+        "Word 2007 XML",
+    ),
+}
+
+DOCX_SAVE_FILTER_SPECS = {
+    "libre": (
+        "Office Open XML Text",
+        "MS Word 2007 XML",
+    ),
+    "openoffice": (
+        "MS Word 2007 XML",
+        "Office Open XML Text",
+    ),
+}
+
+MAX_PASSES = 20
+
+# UNO enum numeric values (``from com.sun.star.*`` is unavailable in some LO/OO Pythons).
+BREAK_NONE = 0
+BREAK_PAGE_BEFORE = 3
+CONTROL_PARAGRAPH_BREAK = 0
+
+_UNO_MISSING = object()
+
+try:
+    _TEXT_TYPE = unicode
+except NameError:
+    _TEXT_TYPE = str
+
+
+def as_unicode(value):
+    """Coerce value to a Unicode string (safe on Python 2.7 + 3)."""
+    if value is None:
+        return _TEXT_TYPE()
+    if isinstance(value, _TEXT_TYPE):
+        return value
+    if isinstance(value, bytes):
+        for encoding in ("utf-8", sys.getfilesystemencoding(), "mbcs"):
+            if not encoding:
+                continue
+            try:
+                return value.decode(encoding)
+            except (UnicodeDecodeError, LookupError):
+                continue
+        return value.decode("utf-8", "replace")
+    try:
+        return _TEXT_TYPE(value)
+    except Exception:
+        return _TEXT_TYPE(repr(value))
+
+
+def uno_error_text(exc):
+    """Readable UNO/Python exception text (Py2 often has empty str(exc))."""
+    for attr in ("Message", "Context", "Name"):
+        try:
+            val = getattr(exc, attr, None)
+            if val:
+                text = as_unicode(val)
+                if text:
+                    return text
+        except Exception:
+            continue
+    text = as_unicode(exc)
+    if text:
+        return text
+    return repr(exc)
+
+
+def uno_call(obj, base, default=_UNO_MISSING):
+    """Invoke UNO getter ``getBase()`` or read attribute ``Base`` (OO Py2 vs LO Py3)."""
+    getter = "get" + base
+    for access in (getter, base):
+        try:
+            member = getattr(obj, access)
+            return member() if callable(member) else member
+        except Exception:
+            continue
+    if default is not _UNO_MISSING:
+        return default
+    raise AttributeError(base)
+
+_RE_FLAGS = re.IGNORECASE
+if sys.version_info[0] < 3:
+    _RE_FLAGS |= re.UNICODE
+
+_CAPTION_RE = re.compile(
+    r"^(?:Продолжение\s+)?"
+    r"(?P<kind>Таблица|Таблицы|Листинг|Листинга)\s+"
+    r"(?P<number>[\d.]+)",
+    _RE_FLAGS,
+)
+_CONTINUATION_RE = re.compile(
+    r"^Продолжение\s+(?:Таблицы|Листинга)\b", _RE_FLAGS
+)
+
+
+class CaptionInfo(object):
+    __slots__ = ("kind", "number", "is_continuation")
+
+    def __init__(self, kind, number, is_continuation=False):
+        self.kind = kind
+        self.number = number
+        self.is_continuation = is_continuation
+
+
+def parse_caption_text(text):
+    raw = (text or "").replace("\r", "").replace("\x07", "").strip()
+    if not raw:
+        return None
+    first = raw.split("\n", 1)[0].strip()
+    m = _CAPTION_RE.match(first)
+    if not m:
+        return None
+    kind_raw = m.group("kind").lower()
+    if kind_raw.startswith("табл"):
+        kind = "table"
+    elif kind_raw.startswith("лист"):
+        kind = "listing"
+    else:
+        return None
+    return CaptionInfo(
+        kind=kind,
+        number=m.group("number"),
+        is_continuation=bool(_CONTINUATION_RE.match(first)),
+    )
+
+
+def find_page_break_row(page_numbers):
+    if len(page_numbers) < 2:
+        return None
+    for i in range(1, len(page_numbers)):
+        if page_numbers[i] > page_numbers[i - 1]:
+            row_1based = i + 1
+            if row_1based <= 1:
+                return None
+            return row_1based
+    return None
+
+
+def is_orphan_header_break(break_at, had_header):
+    return bool(had_header and break_at == 2)
+
+
+def is_caption_orphaned(caption_page, first_row_page):
+    if caption_page is None or first_row_page is None:
+        return False
+    return caption_page < first_row_page
+
+
+def continuation_label(kind, number):
+    if kind == "listing":
+        return "Продолжение Листинга {0}".format(number)
+    return "Продолжение Таблицы {0}".format(number)
+
+
+def caption_style_name(kind):
+    return "Caption Listing" if kind == "listing" else "Название таблицы"
+
+
+def file_url_from_path(path):
+    abs_path = os.path.abspath(path)
+    posix = abs_path.replace("\\", "/")
+    if len(posix) >= 2 and posix[1] == ":":
+        return "file:///" + quote(posix, safe="/:@")
+    return "file://" + quote(posix, safe="/:@")
+
+
+def path_to_file_url(uno, path):
+    abs_path = os.path.abspath(path)
+    try:
+        url = uno.systemPathToFileUrl(abs_path)
+        if url:
+            return url
+    except Exception:
+        pass
+    return file_url_from_path(abs_path)
+
+
+def docx_filter_candidates(engine):
+    return DOCX_FILTER_CANDIDATES.get(engine, DOCX_FILTER_CANDIDATES["libre"])
+
+
+def docx_save_filter_specs(engine):
+    return DOCX_SAVE_FILTER_SPECS.get(engine, DOCX_SAVE_FILTER_SPECS["libre"])