feat: 新增论文排版脚本并输出标准格式文档

example/format_thesis_docx.py

@@ -0,0 +1,197 @@
+from docx import Document
+from docx.enum.text import WD_ALIGN_PARAGRAPH
+from docx.oxml import OxmlElement
+from docx.oxml.ns import qn
+from docx.shared import Cm, Pt, RGBColor
+
+
+SRC = r"/Users/apple/code/bs/mying/example/萌贝母婴商城毕业论文初稿-2026版.docx"
+DST = r"/Users/apple/code/bs/mying/example/萌贝母婴商城毕业论文初稿-2026版-排版.docx"
+
+
+def set_style_font(
+    style,
+    east_asia_font: str,
+    size_pt: float,
+    bold: bool | None = None,
+    west_font: str = "Times New Roman",
+):
+    font = style.font
+    font.name = west_font
+    font.size = Pt(size_pt)
+    if bold is not None:
+        font.bold = bold
+    font.color.rgb = RGBColor(0, 0, 0)
+    rfonts = style.element.get_or_add_rPr().get_or_add_rFonts()
+    rfonts.set(qn("w:ascii"), west_font)
+    rfonts.set(qn("w:hAnsi"), west_font)
+    rfonts.set(qn("w:eastAsia"), east_asia_font)
+
+
+def set_runs_font(
+    paragraph,
+    east_asia_font: str,
+    size_pt: float,
+    bold: bool | None = None,
+    west_font: str = "Times New Roman",
+):
+    for run in paragraph.runs:
+        run.font.name = west_font
+        run.font.size = Pt(size_pt)
+        if bold is not None:
+            run.font.bold = bold
+        run.font.color.rgb = RGBColor(0, 0, 0)
+        rpr = run._element.get_or_add_rPr()
+        rfonts = rpr.get_or_add_rFonts()
+        rfonts.set(qn("w:ascii"), west_font)
+        rfonts.set(qn("w:hAnsi"), west_font)
+        rfonts.set(qn("w:eastAsia"), east_asia_font)
+
+
+def set_runs_common(paragraph, italic: bool | None = None, color_black: bool = True):
+    for run in paragraph.runs:
+        if italic is not None:
+            run.font.italic = italic
+        if color_black:
+            run.font.color.rgb = RGBColor(0, 0, 0)
+
+
+def is_numbered_paragraph(paragraph) -> bool:
+    ppr = paragraph._p.pPr
+    if ppr is None:
+        return False
+    return ppr.numPr is not None
+
+
+def set_table_all_borders_black(table):
+    for row in table.rows:
+        for cell in row.cells:
+            tc = cell._tc
+            tc_pr = tc.get_or_add_tcPr()
+            tc_borders = tc_pr.find(qn("w:tcBorders"))
+            if tc_borders is None:
+                tc_borders = OxmlElement("w:tcBorders")
+                tc_pr.append(tc_borders)
+            for edge in ("top", "left", "bottom", "right", "insideH", "insideV"):
+                edge_tag = qn(f"w:{edge}")
+                elem = tc_borders.find(edge_tag)
+                if elem is None:
+                    elem = OxmlElement(f"w:{edge}")
+                    tc_borders.append(elem)
+                elem.set(qn("w:val"), "single")
+                elem.set(qn("w:sz"), "4")
+                elem.set(qn("w:color"), "000000")
+                elem.set(qn("w:space"), "0")
+def iter_table_paragraphs(table):
+    for row in table.rows:
+        for cell in row.cells:
+            for p in cell.paragraphs:
+                yield p
+            for t in cell.tables:
+                yield from iter_table_paragraphs(t)
+
+def set_first_line_two_chars(paragraph, twips: int = 420, chars: int = 200):
+    ppr = paragraph._p.get_or_add_pPr()
+    ind = ppr.find(qn("w:ind"))
+    if ind is None:
+        ind = OxmlElement("w:ind")
+        ppr.append(ind)
+    ind.set(qn("w:firstLine"), str(twips))
+    ind.set(qn("w:firstLineChars"), str(chars))
+
+def format_paragraph(p):
+    style_name = p.style.name if p.style is not None else ""
+    if style_name == "Heading 1":
+        apply_para_format(p, 1.5, 0, WD_ALIGN_PARAGRAPH.CENTER)
+        set_runs_font(p, "黑体", 22, True)
+    elif style_name == "Heading 2":
+        apply_para_format(p, 1.5, 32)
+        set_runs_font(p, "黑体", 16, True)
+    elif style_name == "Heading 3":
+        apply_para_format(p, 1.5, 28)
+        set_runs_font(p, "宋体", 14, True)
+    elif style_name == "Heading 4":
+        apply_para_format(p, 1.5, 24)
+        set_runs_font(p, "宋体", 12, True)
+        set_runs_common(p, italic=False, color_black=True)
+    elif is_numbered_paragraph(p) or style_name.startswith("List Number"):
+        p.paragraph_format.line_spacing = 1.5
+        set_runs_font(p, "宋体", 12)
+        set_runs_common(p, color_black=True)
+    else:
+        apply_para_format(p, 1.5, 24)
+        set_runs_font(p, "宋体", 10.5)
+        set_runs_common(p, color_black=True)
+
+
+def apply_para_format(paragraph, line_spacing: float, first_line_pt: float | None = None, align=None):
+    fmt = paragraph.paragraph_format
+    fmt.line_spacing = line_spacing
+    fmt.space_before = Pt(0)
+    fmt.space_after = Pt(0)
+    if first_line_pt is not None:
+        fmt.first_line_indent = Pt(first_line_pt)
+        set_first_line_two_chars(paragraph)
+
+    if align is not None:
+        paragraph.alignment = align
+def set_page_layout(doc):
+    for section in doc.sections:
+        section.page_width = Cm(21.0)
+        section.page_height = Cm(29.7)
+        section.top_margin = Cm(2.5)
+        section.bottom_margin = Cm(2.5)
+        section.left_margin = Cm(2.5)
+        section.right_margin = Cm(2.5)
+        section.header_distance = Cm(1.5)
+        section.footer_distance = Cm(1.75)
+
+def main():
+    doc = Document(SRC)
+
+    normal = doc.styles["Normal"]
+    h1 = doc.styles["Heading 1"]
+    h2 = doc.styles["Heading 2"]
+    h3 = doc.styles["Heading 3"]
+    h4 = doc.styles["Heading 4"]
+
+    # 正文：宋体小四，首行缩进2字符（约24pt），1.5倍行距
+    set_style_font(normal, "宋体", 10.5)
+    normal.paragraph_format.line_spacing = 1.5
+    normal.paragraph_format.first_line_indent = Pt(21)
+
+    # 标题1：黑体二号，加粗，居中，1.5倍行距
+    set_style_font(h1, "黑体", 22, True)
+    h1.paragraph_format.line_spacing = 1.5
+    h1.paragraph_format.first_line_indent = Pt(0)
+
+    # 标题2：黑体三号，加粗，首行缩进2字符，1.5倍行距
+    set_style_font(h2, "黑体", 16, True)
+    h2.paragraph_format.line_spacing = 1.5
+    h2.paragraph_format.first_line_indent = Pt(32)
+
+    # 标题3：宋体四号，加粗，首行缩进2字符，1.5倍行距
+    set_style_font(h3, "黑体", 14, True)
+    h3.paragraph_format.line_spacing = 1.5
+    h3.paragraph_format.first_line_indent = Pt(28)
+
+    # 标题4：加粗，取消斜体，黑色，1.5倍行距
+    set_style_font(h4, "黑体", 14, True)
+    h4.font.italic = False
+    h4.paragraph_format.line_spacing = 1.5
+    h4.paragraph_format.first_line_indent = Pt(24)
+    set_page_layout(doc)
+    for p in doc.paragraphs:
+        format_paragraph(p)
+
+    for t in doc.tables:
+        set_table_all_borders_black(t)
+        for p in iter_table_paragraphs(t):
+            format_paragraph(p)
+
+    doc.save(DST)
+    print(DST)
+
+
+if __name__ == "__main__":
+    main()