hot-fix 0.4.3

- fix Не появляется пустая странициа перед названием таблицы - fix Не появляется пустая страница перед альбомной ориентацией - fix При отказе от изменений создаётся новый файл - fix Замена "..." на «...» больше не затрагивает uml и mermaid блоки
2026-09-05 10:24:13 +03:00
parent 8508d04bbc
commit febf62c257
18 changed files with 267 additions and 49 deletions
@@ -9,7 +9,6 @@ __pycache__/
 # Distribution / packaging
 .Python
 build/
-build-fast/
 develop-eggs/
 dist/
 downloads/
@@ -181,4 +180,5 @@ cython_debug/
 md2gost.schemes.json
 md2gost.include-cache.json
 include-cache/
-.md2gost-cache/
+.md2gost-cache/
+/build-fast
@@ -15,12 +15,12 @@ IMPORTANT: Do NOT post this list to the issue-tracker. Use it as a basis for
            tracking down the missing module yourself. Thanks!

 missing module named 'org.python' - imported by copy (optional), xml.sax (delayed, conditional)
-missing module named grp - imported by shutil (delayed, optional), tarfile (optional), pathlib (delayed, optional), subprocess (delayed, conditional, optional), backports.tarfile (optional), distutils.archive_util (optional), setuptools._distutils.archive_util (optional)
-missing module named pwd - imported by posixpath (delayed, conditional), shutil (delayed, optional), tarfile (optional), pathlib (delayed, optional), subprocess (delayed, conditional, optional), getpass (delayed), distutils.util (delayed, conditional, optional), netrc (delayed, conditional), backports.tarfile (optional), distutils.archive_util (optional), http.server (delayed, optional), webbrowser (delayed), psutil (optional), setuptools._distutils.util (delayed, conditional, optional), setuptools._distutils.archive_util (optional)
+missing module named grp - imported by subprocess (delayed, conditional, optional), shutil (delayed, optional), tarfile (optional), pathlib (delayed, optional), backports.tarfile (optional), distutils.archive_util (optional), setuptools._distutils.archive_util (optional)
+missing module named pwd - imported by posixpath (delayed, conditional), subprocess (delayed, conditional, optional), shutil (delayed, optional), tarfile (optional), pathlib (delayed, optional), getpass (delayed), distutils.util (delayed, conditional, optional), netrc (delayed, conditional), backports.tarfile (optional), distutils.archive_util (optional), http.server (delayed, optional), webbrowser (delayed), psutil (optional), setuptools._distutils.util (delayed, conditional, optional), setuptools._distutils.archive_util (optional)
 missing module named urllib.urlopen - imported by urllib (delayed, optional), lxml.html (delayed, optional)
 missing module named urllib.urlencode - imported by urllib (delayed, optional), lxml.html (delayed, optional)
 missing module named pep517 - imported by importlib.metadata (delayed)
-missing module named posix - imported by shutil (conditional), importlib._bootstrap_external (conditional), os (conditional, optional)
+missing module named posix - imported by os (conditional, optional), shutil (conditional), importlib._bootstrap_external (conditional)
 missing module named resource - imported by posix (top-level)
 missing module named _frozen_importlib_external - imported by importlib._bootstrap (delayed), importlib (optional), importlib.abc (optional), zipimport (top-level)
 excluded module named _frozen_importlib - imported by importlib (optional), importlib.abc (optional), zipimport (top-level)
@@ -27176,6 +27176,7 @@ imported by:
 &#8226;   <a href="#md2gost.renderable.caption">md2gost.renderable.caption</a>
 &#8226;   <a href="#md2gost.renderable.listing">md2gost.renderable.listing</a>
 &#8226;   <a href="#md2gost.renderable.table">md2gost.renderable.table</a>
+ &#8226;   <a href="#md2gost.renderer">md2gost.renderer</a>
 &#8226;   <a href="#md2gost_exe.py">md2gost_exe.py</a>

  </div>
@@ -27780,6 +27781,7 @@ imports:
 &#8226;   <a href="#md2gost.layout_tracker">md2gost.layout_tracker</a>
 &#8226;   <a href="#md2gost.numberer">md2gost.numberer</a>
 &#8226;   <a href="#md2gost.page_geometry">md2gost.page_geometry</a>
+ &#8226;   <a href="#md2gost.profiles">md2gost.profiles</a>
 &#8226;   <a href="#md2gost.renderable">md2gost.renderable</a>
 &#8226;   <a href="#md2gost.renderable.diagram">md2gost.renderable.diagram</a>
 &#8226;   <a href="#md2gost.renderable.equation">md2gost.renderable.equation</a>
@@ -21,7 +21,7 @@ from .diagram_schemes import (
 )
 from .dnd import enable_file_drop, first_markdown
 from .help_content import SCHEMES_HELP, USAGE_HELP, load_prompt_catalog
-from .pipeline import ConvertRequest, convert, default_output_path
+from .pipeline import ConvertRequest, convert, default_output_path, timestamped_output_path
 from .profiles import (
    DEFAULT_HEADING_NUMBERING,
    DEFAULT_TABLE_CONTINUATION,
@@ -1200,9 +1200,19 @@ class Md2GostApp:
        if not req.filename:
            messagebox.showinfo("md2gost", "Сначала перетащите или выберите markdown-файл.")
            return
-        if not req.check_only and req.output and os.path.isfile(req.output):
-            if not messagebox.askyesno("md2gost", f"Файл уже есть:\n{req.output}\n\nПерезаписать?"):
-                return
+        if not req.check_only:
+            target = (req.output or "").strip() or (
+                default_output_path(req.filename) if req.filename else ""
+            )
+            if target and os.path.isfile(target):
+                if not messagebox.askyesno(
+                    "md2gost", f"Файл уже есть:\n{target}\n\nПерезаписать?"
+                ):
+                    target = timestamped_output_path(target)
+                    self._append_log(f"Не перезаписываем, пишем в {target}")
+                req.output = target
+                self.output_var.set(target)
+                self._auto_output = False
        self._busy = True
        self.convert_btn.configure(state=tk.DISABLED, bg=NAVY_DIM)
        self.progress.start(12)
@@ -1225,6 +1235,7 @@ class Md2GostApp:
            self._append_log("Дебаг сброшен после этой сборки.")
        if result.ok:
            if result.output_path:
+                self.output_var.set(result.output_path)
                self._append_log("Готово.")
            elif result.check_report:
                self._append_log("Проверка завершена.")
@@ -16,6 +16,7 @@ USAGE_HELP = """md2gost — Markdown → DOCX (ТЗ МИРЭА / ГОСТ 7.32)
   PlantUML / Kroki — Настройки → Диаграммы.
   Свои UML-схемы — меню «Шаблоны UML».
 3. Нажмите «Конвертировать». Документ сохранится рядом с исходником (или по пути «Выходной DOCX»).
+   Если файл уже есть и вы откажетесь перезаписывать — сохранится как имя_гггг-мм-дд-ЧЧ-ММ.docx.
   Дебаг (меню сверху) — следующая сборка с отладочными данными в документе.

 Типы: practice (по умолчанию) / coursework / vkr — ГОСТ МИРЭА; PIS_custom — отчёт по практикам ПИС; APID_coursework — курсовая АПИД.
@@ -8,6 +8,7 @@ import platform
 import subprocess
 import traceback
 from dataclasses import dataclass
+from datetime import datetime
 from getpass import getuser
 from typing import Callable

@@ -73,6 +74,25 @@ def default_output_path(filename: str) -> str:
    return os.path.join(os.path.dirname(os.path.abspath(filename)), base + ".docx")


+def timestamped_output_path(path: str) -> str:
+    """report.docx → report_2026-09-05-10-05.docx (local time; seconds if taken)."""
+    directory, name = os.path.split(os.path.abspath(path))
+    stem, ext = os.path.splitext(name)
+    if ext.lower() != ".docx":
+        ext = ".docx"
+    stamp = datetime.now().strftime("%Y-%m-%d-%H-%M")
+    candidate = os.path.join(directory, f"{stem}_{stamp}{ext}")
+    if not os.path.exists(candidate):
+        return candidate
+    stamp = datetime.now().strftime("%Y-%m-%d-%H-%M-%S")
+    candidate = os.path.join(directory, f"{stem}_{stamp}{ext}")
+    n = 2
+    while os.path.exists(candidate):
+        candidate = os.path.join(directory, f"{stem}_{stamp}_{n}{ext}")
+        n += 1
+    return candidate
+
+
 def default_template_path() -> str:
    return os.path.join(package_dir(), "Template.docx")

@@ -235,7 +255,13 @@ def convert(req: ConvertRequest, log: LogFn | None = None) -> ConvertResult:

        document.core_properties.author = getuser()
        document.core_properties.comments = "Создано при помощи md2gost (ТЗ МИРЭА)"
-        document.save(output)
+        try:
+            document.save(output)
+        except PermissionError:
+            alt = timestamped_output_path(output)
+            emit(f"Не удалось записать {output} (файл занят). Сохраняю как {alt}")
+            document.save(alt)
+            output = alt
    except Exception as exc:
        emit(traceback.format_exc())
        return ConvertResult(
@@ -99,6 +99,38 @@ def get_profile(doc_type: str) -> DocProfile:

 # --- text autofixes ---

+# ```uml / ```mermaid / ```uml-c4 / ```bpmn … — кавычки и тире там синтаксис
+_FENCE_RE = re.compile(
+    r"^(?P<fence>`{3,}|~{3,})(?P<info>[^\n]*)\n"
+    r"(?P<body>[\s\S]*?)"
+    r"^(?P=fence)[ \t]*(?:\n|$)",
+    re.M,
+)
+
+
+def _fence_lang(info: str) -> str:
+    return (info.strip().split() or [""])[0].lower()
+
+
+def _is_diagram_fence_info(info: str) -> bool:
+    from .diagram_schemes import is_diagram_lang
+    return is_diagram_lang(_fence_lang(info))
+
+
+def _map_outside_diagram_fences(text: str, fn) -> str:
+    """Apply fn only outside UML / Mermaid / scheme fences."""
+    out: list[str] = []
+    pos = 0
+    for m in _FENCE_RE.finditer(text):
+        if not _is_diagram_fence_info(m.group("info")):
+            continue
+        out.append(fn(text[pos:m.start()]))
+        out.append(m.group(0))
+        pos = m.end()
+    out.append(fn(text[pos:]))
+    return "".join(out)
+
+
 def fix_russian_quotes(text: str) -> str:
    """Replace "..." with «...» for Russian text segments (heuristic)."""
    def repl(m):
@@ -142,12 +174,14 @@ def dash_separator() -> str:
 def preprocess_markdown(text: str, emdash_to_hyphen: bool | None = None) -> str:
    if emdash_to_hyphen is not None:
        set_emdash_to_hyphen(emdash_to_hyphen)
-    text = fix_russian_quotes(text)
-    if _EMDASH_TO_HYPHEN:
-        # Keep ASCII hyphens; flatten any em dashes from the source
-        text = replace_emdash_with_hyphen(text)
-    else:
-        text = fix_dashes(text)
+
+    def _apply(chunk: str) -> str:
+        chunk = fix_russian_quotes(chunk)
+        if _EMDASH_TO_HYPHEN:
+            return replace_emdash_with_hyphen(chunk)
+        return fix_dashes(chunk)
+
+    text = _map_outside_diagram_fences(text, _apply)
    text = separate_biblio_lines(text)
    return text

@@ -67,14 +67,23 @@ class Caption(Renderable):
            layout_state.max_width
        ).calculate_height()

-        if self._before and ((height_data.lines + 2 - 1) * height_data.line_spacing + 1) * height_data.line_height \
-                > layout_state.remaining_page_height:
-            self._docx_paragraph.paragraph_format.page_break_before = True
+        # Table/listing captions use keep_with_next. pageBreakBefore + keepWithNext
+        # on a paragraph that Word already wraps to a new page produces a blank
+        # page that the layout tracker / debug overlay never sees.
+        leftover = 0
+        need_new_page = (
+            self._before
+            and layout_state.current_page_height > 0
+            and ((height_data.lines + 2 - 1) * height_data.line_spacing + 1) * height_data.line_height
+            > layout_state.remaining_page_height
+        )
+        if need_new_page:
+            leftover = layout_state.remaining_page_height
            height_data = ParagraphSizer(
                self._docx_paragraph,
                None,
                layout_state.max_width
            ).calculate_height()

-        yield RenderedInfo(self._docx_paragraph, height_data.full + (layout_state.remaining_page_height
-                                                                     if self._docx_paragraph.paragraph_format.page_break_before else 0))
+        self._docx_paragraph.paragraph_format.page_break_before = False
+        yield RenderedInfo(self._docx_paragraph, height_data.full + leftover)
@@ -123,21 +123,13 @@ class Listing(Renderable, RequiresNumbering):
    def _emit_page_break_and_optional_caption(
        self, layout_state: LayoutState
    ) -> Generator[RenderedInfo, None, None]:
-        mode = self._continuation_mode
-        if mode == "legacy":
-            continuation_paragraph = self._make_continuation_paragraph()
-            continuation_paragraph.page_break_before = True
-            info = next(continuation_paragraph.render(None, copy(layout_state)))
-            layout_state.add_height(info.height)
-            yield info
-            return
-
        page_break_info = next(PageBreak(self._parent).render(None, layout_state))
        layout_state.add_height(page_break_info.height)
        yield page_break_info

        continuation_paragraph = self._make_continuation_paragraph()
        continuation_paragraph._docx_paragraph.paragraph_format.keep_with_next = True
+        continuation_paragraph._docx_paragraph.paragraph_format.page_break_before = False
        info = next(continuation_paragraph.render(None, copy(layout_state)))
        layout_state.add_height(info.height)
        yield info
@@ -154,23 +154,15 @@ class Table(Renderable, RequiresNumbering):
        self, layout_state: LayoutState
    ) -> Generator[RenderedInfo, None, None]:
        """For legacy/caption only: break page and insert «Продолжение Таблицы N»."""
-        mode = self._continuation_mode
-
-        if mode == "legacy":
-            continuation_paragraph = self._make_continuation_paragraph()
-            continuation_paragraph.page_break_before = True
-            info = next(continuation_paragraph.render(None, copy(layout_state)))
-            layout_state.add_height(info.height)
-            yield info
-            return
-
-        # caption
        page_break_info = next(PageBreak(self._parent).render(None, layout_state))
        layout_state.add_height(page_break_info.height)
        yield page_break_info

+        # Never page_break_before on «Название таблицы»: keep_with_next +
+        # pageBreakBefore makes a blank page that debug does not count.
        continuation_paragraph = self._make_continuation_paragraph()
        continuation_paragraph._docx_paragraph.paragraph_format.keep_with_next = True
+        continuation_paragraph._docx_paragraph.paragraph_format.page_break_before = False
        info = next(continuation_paragraph.render(None, copy(layout_state)))
        layout_state.add_height(info.height)
        yield info
@@ -7,7 +7,6 @@ from docx.enum.text import WD_TAB_LEADER, WD_TAB_ALIGNMENT, WD_PARAGRAPH_ALIGNME
 from docx.shared import Parented

 from . import Paragraph
-from .page_break import PageBreak
 from .renderable import Renderable
 from ..layout_tracker import LayoutState
 from ..rendered_info import RenderedInfo
@@ -142,4 +141,5 @@ class ToC(Renderable):

        for rendered_info in self._paragraph.render(previous_rendered, copy(layout_state)):
            yield RenderedInfo(rendered_info.docx_element, 0)
-        yield from PageBreak(self._parent).render(None, copy(layout_state))
+        # Page break after TOC is a section break in Renderer (body + PAGE).
+        # An extra w:br here + NEW_PAGE section = blank portrait page.
@@ -95,12 +95,11 @@ class Renderer:
            self._landscape_depth += 1
            return
        self._flush_to_new_screen()
-        if self._after_toc:
-            self._ensure_body_section_with_page_numbers()
-        elif not self._body_section_started:
-            # No TOC — still need page numbers on subsequent sections.
+        if not self._body_section_started:
+            # Do not add_section(portrait) here. A portrait NEW_PAGE plus the
+            # landscape NEW_PAGE below is two breaks: text → empty portrait → album.
            self._body_section_started = True
-            apply_centered_page_footer(self._document.sections[0])
+            apply_centered_page_footer(self._document.sections[-1])
        section = self._document.add_section(WD_SECTION.NEW_PAGE)
        apply_section_geometry(section, landscape=True)
        apply_centered_page_footer(section)
@@ -366,6 +365,13 @@ class Renderer:
            extra = self._after_current.pop(0)
            self.render(extra)

+        if isinstance(renderable, ToC):
+            # One NEXT_PAGE section after the TOC field — not a w:br plus a section.
+            self._ensure_body_section_with_page_numbers()
+            state = self._layout_tracker.current_state
+            if state.current_page_height > 0:
+                self._layout_tracker.new_page()
+
        if deferred_listing is not None:
            self.render(deferred_listing)

@@ -156,6 +156,86 @@ def test_landscape_places_figure_inside_landscape_section(tmp_path, monkeypatch)
    assert image_section == 1


+W_NS = "{http://schemas.openxmlformats.org/wordprocessingml/2006/main}"
+
+
+def _is_break_paragraph(el) -> bool:
+    """Empty paragraph that is a page break and/or hosts a section break."""
+    if el.tag != f"{W_NS}p":
+        return False
+    texts = [t.text or "" for t in el.findall(f".//{W_NS}t")]
+    if any(t.strip() for t in texts):
+        return False
+    pPr = el.find(f"{W_NS}pPr")
+    has_sect = pPr is not None and pPr.find(f"{W_NS}sectPr") is not None
+    has_br = any(br.get(f"{W_NS}type") == "page" for br in el.findall(f".//{W_NS}br"))
+    return has_sect or has_br
+
+
+def _breaks_between_last_text_and_table(doc) -> int:
+    """Empty page/section-break paras between the last body sentence and the table."""
+    children = list(doc.element.body)
+    table_i = next(i for i, el in enumerate(children) if el.tag == f"{W_NS}tbl")
+    j = table_i - 1
+    breaks = 0
+    while j >= 0:
+        el = children[j]
+        if _is_break_paragraph(el):
+            breaks += 1
+            j -= 1
+            continue
+        if el.tag == f"{W_NS}p":
+            text = "".join(t.text or "" for t in el.findall(f".//{W_NS}t")).strip()
+            if text.startswith("Таблица") or text.startswith("Продолжение"):
+                j -= 1
+                continue
+        break
+    return breaks
+
+
+def test_portrait_to_landscape_is_single_section_break():
+    """Text then +landscape: one next-page section, no extra blank portrait page."""
+    doc = Document(os.path.join(package_dir(), "Template.docx"))
+    body = doc._body._element
+    for child in list(body):
+        if child.tag.endswith("}sectPr"):
+            continue
+        body.remove(child)
+
+    from md2gost.renderable.heading import Heading
+    from md2gost.renderable.paragraph import Paragraph
+    from md2gost.renderable.toc import ToC
+
+    h = Heading(doc._body, 1, False)
+    h.add_run("СОДЕРЖАНИЕ")
+    toc = ToC(doc._body, toc_mode="native")
+    para = Paragraph(doc._body)
+    para.add_run("Текст перед широкой таблицей.")
+    table = Table(doc._body, 2, 3, CaptionInfo("wide", "Карта", landscape=True))
+    for r in range(2):
+        for c in range(3):
+            table.add_paragraph_to_cell(r, c).add_run(f"{r}{c}")
+
+    Renderer(doc, skip_numbering=True).process([h, toc, para, table])
+    assert _breaks_between_last_text_and_table(doc) == 1
+
+    # Same without TOC — still a single break, not page-break + section-break.
+    doc2 = Document(os.path.join(package_dir(), "Template.docx"))
+    body2 = doc2._body._element
+    for child in list(body2):
+        if child.tag.endswith("}sectPr"):
+            continue
+        body2.remove(child)
+    para2 = Paragraph(doc2._body)
+    para2.add_run("Текст перед широкой таблицей.")
+    table2 = Table(doc2._body, 2, 3, CaptionInfo("wide2", "Карта", landscape=True))
+    for r in range(2):
+        for c in range(3):
+            table2.add_paragraph_to_cell(r, c).add_run(f"{r}{c}")
+    Renderer(doc2, skip_numbering=True).process([para2, table2])
+    assert _breaks_between_last_text_and_table(doc2) == 1
+
+
 def test_landscape_listing_deferred_after_section(tmp_path, monkeypatch):
    """+landscape +listing → listing must not sit in the landscape section."""
    from unittest.mock import patch
@@ -190,6 +190,27 @@ def test_preprocess_quotes():
    assert "«тест»" in preprocess_markdown('Он сказал "тест" вслух.')


+def test_preprocess_skips_uml_and_mermaid_fences():
+    md = (
+        'В тексте "кавычки".\n\n'
+        "```uml\n"
+        'rectangle "Акционеры" as own\n'
+        "```\n\n"
+        "```mermaid\n"
+        'title "Выручка"\n'
+        "```\n\n"
+        "```uml-c4context\n"
+        'Person(user, "Заявитель")\n'
+        "```\n"
+    )
+    out = preprocess_markdown(md)
+    assert "«кавычки»" in out
+    assert 'rectangle "Акционеры" as own' in out
+    assert 'title "Выручка"' in out
+    assert 'Person(user, "Заявитель")' in out
+    assert "«Акционеры»" not in out
+
+
 def test_formula_refs():
    assert find_formula_refs("см. @Формула:eq1 и далее") == {"eq1"}

@@ -466,6 +487,28 @@ def test_appendix_index_auto_inserted():
    assert out2 == [sec, manual, a, b]


+def test_table_caption_does_not_set_page_break_before():
+    import os
+    import docx
+    from docx.shared import Mm
+    from md2gost import package_dir
+    from md2gost.styles import apply_mirea_styles
+    from md2gost.layout_tracker import LayoutState
+    from md2gost.renderable.caption import Caption, CaptionInfo
+
+    doc = docx.Document(os.path.join(package_dir(), "Template.docx"))
+    apply_mirea_styles(doc)
+    cap = Caption(doc._body, "Таблица", CaptionInfo("t1", "Реквизиты"), "1", True)
+    state = LayoutState(Mm(250), Mm(160))
+    state.add_height(Mm(248))
+    leftover = state.remaining_page_height
+    infos = list(cap.render(None, state))
+    assert len(infos) == 1
+    assert cap._docx_paragraph.style.name == "Название таблицы"
+    assert cap._docx_paragraph.paragraph_format.page_break_before in (None, False)
+    assert infos[0].height > leftover
+
+
 def test_caption_table_keep_with_next_and_toc_styles():
    import docx
    from docx.enum.text import WD_ALIGN_PARAGRAPH, WD_LINE_SPACING
@@ -4,7 +4,13 @@ import os
 import sys

 from md2gost.dnd import first_markdown, normalize_drop_paths, parse_tkdnd_files
-from md2gost.pipeline import ConvertRequest, convert, default_output_path, should_launch_gui
+from md2gost.pipeline import (
+    ConvertRequest,
+    convert,
+    default_output_path,
+    should_launch_gui,
+    timestamped_output_path,
+)
 from md2gost.__main__ import build_parser, request_from_args


@@ -20,6 +26,22 @@ def test_default_output_path():
    assert path.endswith("report.docx")


+def test_timestamped_output_path(tmp_path):
+    import re
+    src = tmp_path / "ПР1.docx"
+    src.write_bytes(b"x")
+    out = timestamped_output_path(str(src))
+    name = os.path.basename(out)
+    assert re.fullmatch(r"ПР1_\d{4}-\d{2}-\d{2}-\d{2}-\d{2}\.docx", name)
+    assert not os.path.exists(out)
+    # collision → seconds (or seconds_2)
+    open(out, "wb").close()
+    out2 = timestamped_output_path(str(src))
+    assert out2 != out
+    assert os.path.basename(out2).startswith("ПР1_")
+    assert out2.endswith(".docx")
+
+
 def test_parse_tkdnd_files():
    raw = r"{C:\My Files\a.md} C:\tmp\b.md"
    assert parse_tkdnd_files(raw) == [r"C:\My Files\a.md", r"C:\tmp\b.md"]