feat(transit): 添加参考文献解析功能并修复段落编号继承问题

新增了参考文献处理模块,支持按照 GB 7714 《文后参考文献著录规则》顺序编码制
解析和格式化参考文献。同时修复了段落替换过程中自动编号丢失的问题。

- 新增 transit/references.py 模块,提供参考文献解析和格式化功能
- 在 body.py 的 replace_placeholder 函数中实现段落编号属性的正确继承
- 修改 transit/__init__.py 导入新的参考文献处理函数
- 更新 transit/config.py 添加参考文献样式配置项
- 修改 transit/renderer.py 集成参考文献处理流程
This commit is contained in:
zzy
2026-05-08 22:14:51 +08:00
parent c29a3e6af0
commit 74d28ea2d8
5 changed files with 149 additions and 4 deletions

View File

@@ -4,6 +4,7 @@ from .parser import parse_markdown
from .body import body_to_paragraphs, replace_placeholder from .body import body_to_paragraphs, replace_placeholder
from .renderer import generate_thesis from .renderer import generate_thesis
from .config import load_config, ThesisConfig from .config import load_config, ThesisConfig
from .references import references_to_paragraphs, format_gb7714
__all__ = [ __all__ = [
"parse_markdown", "parse_markdown",
@@ -12,4 +13,6 @@ __all__ = [
"generate_thesis", "generate_thesis",
"load_config", "load_config",
"ThesisConfig", "ThesisConfig",
"references_to_paragraphs",
"format_gb7714",
] ]

View File

@@ -6,7 +6,9 @@ Markdown 正文 → Word 段落转换。
""" """
import re import re
from copy import deepcopy
from docx import Document from docx import Document
from docx.oxml.ns import qn
_PAT_HEADING = re.compile(r"^(#{1,6})\s+(.+)$", re.MULTILINE) _PAT_HEADING = re.compile(r"^(#{1,6})\s+(.+)$", re.MULTILINE)
@@ -84,6 +86,11 @@ def replace_placeholder(
placeholder_style = para.style.name if para.style else None placeholder_style = para.style.name if para.style else None
parent = para._element.getparent() parent = para._element.getparent()
idx = list(parent).index(para._element) idx = list(parent).index(para._element)
# 保存原段落的编号属性numPr用于继承自动编号
orig_pPr = para._element.find(qn("w:pPr"))
numPr = orig_pPr.find(qn("w:numPr")) if orig_pPr is not None else None
parent.remove(para._element) parent.remove(para._element)
for pd_data in reversed(paragraphs): for pd_data in reversed(paragraphs):
@@ -93,15 +100,24 @@ def replace_placeholder(
# 尝试应用样式,逐步降级 # 尝试应用样式,逐步降级
applied = _apply_style(new_p, doc, style_name) applied = _apply_style(new_p, doc, style_name)
if not applied and style_name.startswith("Heading"): if not applied and style_name.startswith("Heading"):
# 标题样式找不到
new_p.style = doc.styles["Normal"] new_p.style = doc.styles["Normal"]
elif not applied: elif not applied:
# 正文样式找不到 → 尝试占位符自身的样式
if placeholder_style: if placeholder_style:
_apply_style(new_p, doc, placeholder_style) _apply_style(new_p, doc, placeholder_style)
if new_p.style.name == "Normal" and placeholder_style: if new_p.style.name == "Normal" and placeholder_style:
new_p.style = doc.styles[placeholder_style] new_p.style = doc.styles[placeholder_style]
# 继承原段落的编号属性(自动编号)
if numPr is not None:
new_pPr = new_p._element.find(qn("w:pPr"))
if new_pPr is None:
new_pPr = new_p._element.makeelement(qn("w:pPr"), {})
new_p._element.insert(0, new_pPr)
existing = new_pPr.find(qn("w:numPr"))
if existing is not None:
new_pPr.remove(existing)
new_pPr.append(deepcopy(numPr))
parent.insert(idx, new_p._element) parent.insert(idx, new_p._element)
break break

View File

@@ -24,6 +24,7 @@ class ThesisConfig:
) )
body_style: str = "Body Text Indent" body_style: str = "Body Text Indent"
level_offset: int = -1 level_offset: int = -1
reference_style: str = "列出段落1"
def to_dict(self) -> dict: def to_dict(self) -> dict:
"""转成模板渲染用的扁平字典,排除 options 命名空间。""" """转成模板渲染用的扁平字典,排除 options 命名空间。"""
@@ -64,4 +65,5 @@ def load_config(path: str | Path) -> ThesisConfig:
), ),
body_style=opts.get("body_style", "Body Text Indent"), body_style=opts.get("body_style", "Body Text Indent"),
level_offset=opts.get("level_offset", -1), level_offset=opts.get("level_offset", -1),
reference_style=opts.get("reference_style", "列出段落1"),
) )

114
transit/references.py Normal file
View File

@@ -0,0 +1,114 @@
"""
GB 7714 《文后参考文献著录规则》顺序编码制 — 参考文献解析与格式化。
将 Markdown 中 ``[N] ... [TYPE] ...`` 格式的参考文献逐条解析,
按文献类型(书 M、期刊 J、会议 C、学位论文 D、标准 S、电子资源 EB/OL 等)
重新编排为符合 GB 7714 规范的格式,并输出为独立段落。
"""
import re
from typing import Optional
# 单行匹配: [N] 开头
_RE_LINE = re.compile(r"^\[(\d+)\]\s*(.*)$")
# 文献类型标记: [M] [J] [C] [D] [S] [EB/OL] [P] [N]
_RE_TYPE = re.compile(r"\[(\w+(?:/\w+)?)\]")
# GB 7714 中各文献类型的标准格式模板(仅用于说明,实际格式化直接拼接)
TYPE_LABELS: dict[str, str] = {
"M": "专著",
"J": "期刊文章",
"C": "会议论文",
"D": "学位论文",
"S": "标准",
"EB/OL": "电子资源",
"P": "专利",
"N": "报纸文章",
}
def parse_reference_line(line: str) -> Optional[dict]:
"""解析单行参考文献,提取序号、作者+标题、文献类型、来源信息。
期望格式::
[N] Authors. Title[TYPE]. Source info.
"""
line = line.strip()
m = _RE_LINE.match(line)
if not m:
return None
number = int(m.group(1))
rest = m.group(2).strip()
# 定位文献类型标记 [TYPE]
tm = _RE_TYPE.search(rest)
if not tm:
return None
before_type = rest[: tm.start()].strip().rstrip(".")
doc_type = tm.group(1)
after_type = rest[tm.end() :].strip().lstrip(".").strip()
return {
"number": number,
"before_type": before_type, # "Authors. Title"
"doc_type": doc_type, # "M", "J", "EB/OL", …
"after_type": after_type, # 来源信息
}
def _normalize_period(text: str) -> str:
"""确保文本以英文句点结尾GB 7714 要求)。"""
text = text.rstrip()
if text and not text.endswith("."):
text += "."
return text
def format_gb7714(ref: dict) -> str:
"""按 GB 7714 重新编排一条参考文献(不含序号前缀,由 Word 样式自动编号)。
格式::
Authors. Title[TYPE]. Source.
"""
bt = ref["before_type"]
dt = ref["doc_type"]
at = ref["after_type"]
formatted = f"{bt}[{dt}]. {at}"
return _normalize_period(formatted)
def references_to_paragraphs(
ref_text: str,
ref_style: str = "列出段落1",
) -> list[dict]:
"""将参考文献原始文本转换为格式化段落列表。
返回的每个元素::
{"text": str, "level": 0, "style": ref_style}
每条参考文献为一个独立段落。
"""
if not ref_text or ref_text == "<None>":
return [{"text": "<None>", "level": 0, "style": ref_style}]
lines = [l.strip() for l in ref_text.strip().split("\n") if l.strip()]
paragraphs: list[dict] = []
for line in lines:
ref = parse_reference_line(line)
if ref:
formatted = format_gb7714(ref)
else:
# 无法解析时,至少去掉 [N] 前缀
fallback = re.sub(r"^\[\d+\]\s*", "", line)
formatted = _normalize_period(fallback)
paragraphs.append({"text": formatted, "level": 0, "style": ref_style})
return paragraphs

View File

@@ -12,6 +12,7 @@ from docx import Document
from .config import load_config, ThesisConfig from .config import load_config, ThesisConfig
from .parser import parse_markdown from .parser import parse_markdown
from .body import body_to_paragraphs, replace_placeholder from .body import body_to_paragraphs, replace_placeholder
from .references import references_to_paragraphs
_TEXT_FIELDS = [ _TEXT_FIELDS = [
@@ -97,8 +98,13 @@ def generate_thesis(
if body_md else [] if body_md else []
) )
# 6. 占位符 # 6. 解析参考文献为段落列表
ref_text = ctx.get("reference", "")
ref_paragraphs = references_to_paragraphs(ref_text, ref_style=config.reference_style)
# 7. 占位符(替代模板变量,后处理时替换)
ctx["body_placeholder"] = "__CONTEXT_PLACEHOLDER__" ctx["body_placeholder"] = "__CONTEXT_PLACEHOLDER__"
ctx["reference"] = "__REFERENCE_PLACEHOLDER__"
# 7. 渲染模板 # 7. 渲染模板
doc = DocxTemplate(str(template_path)) doc = DocxTemplate(str(template_path))
@@ -108,12 +114,16 @@ def generate_thesis(
temp_path = Path(output_path).with_suffix(".tmp") temp_path = Path(output_path).with_suffix(".tmp")
doc.save(str(temp_path)) doc.save(str(temp_path))
# 9. 正文注入 # 9. 正文注入+参考文献注入
final_doc = Document(str(temp_path)) final_doc = Document(str(temp_path))
replace_placeholder( replace_placeholder(
final_doc, "__CONTEXT_PLACEHOLDER__", body_paragraphs, final_doc, "__CONTEXT_PLACEHOLDER__", body_paragraphs,
default_body_style=config.body_style, default_body_style=config.body_style,
) )
replace_placeholder(
final_doc, "__REFERENCE_PLACEHOLDER__", ref_paragraphs,
default_body_style=config.reference_style,
)
final_doc.save(str(output_path)) final_doc.save(str(output_path))
temp_path.unlink(missing_ok=True) temp_path.unlink(missing_ok=True)