Files
MDHelper/mdhelper/pdf_export.py
T

304 lines
9.7 KiB
Python

from __future__ import annotations
import html
import re
from dataclasses import dataclass
from io import BytesIO
from pathlib import Path
from PySide6.QtCore import QMarginsF
from PySide6.QtGui import QPageLayout, QPageSize
from PySide6.QtWidgets import (
QDialog,
QDialogButtonBox,
QDoubleSpinBox,
QFormLayout,
QLabel,
QVBoxLayout,
QWidget,
)
from .markdown_utils import Heading
PDF_BODY_MARKER = "MDHELPER_PDF_BODY_START_7A61"
PDF_HEADING_MARKER_PREFIX = "MDHELPER_PDF_HEADING_"
@dataclass(frozen=True)
class PdfMargins:
top: float = 15.0
bottom: float = 15.0
left: float = 15.0
right: float = 15.0
def build_pdf_page_layout(margins: PdfMargins | None = None) -> QPageLayout:
selected = margins or PdfMargins()
return QPageLayout(
QPageSize(QPageSize.PageSizeId.A4),
QPageLayout.Orientation.Portrait,
QMarginsF(selected.left, selected.top, selected.right, selected.bottom),
QPageLayout.Unit.Millimeter,
)
class PdfExportSettingsDialog(QDialog):
def __init__(self, margins: PdfMargins, parent: QWidget | None = None) -> None:
super().__init__(parent)
self.setWindowTitle("PDF 导出设置")
self.setModal(True)
note = QLabel("纸张:A4;导出背景:白色;正文前自动生成带页码目录。")
note.setWordWrap(True)
self.top_spin = self._margin_spin(margins.top)
self.bottom_spin = self._margin_spin(margins.bottom)
self.left_spin = self._margin_spin(margins.left)
self.right_spin = self._margin_spin(margins.right)
form = QFormLayout()
form.addRow("上边距", self.top_spin)
form.addRow("下边距", self.bottom_spin)
form.addRow("左边距", self.left_spin)
form.addRow("右边距", self.right_spin)
buttons = QDialogButtonBox(QDialogButtonBox.StandardButton.Ok | QDialogButtonBox.StandardButton.Cancel)
buttons.accepted.connect(self.accept)
buttons.rejected.connect(self.reject)
layout = QVBoxLayout(self)
layout.addWidget(note)
layout.addLayout(form)
layout.addWidget(buttons)
def margins(self) -> PdfMargins:
return PdfMargins(
top=float(self.top_spin.value()),
bottom=float(self.bottom_spin.value()),
left=float(self.left_spin.value()),
right=float(self.right_spin.value()),
)
def _margin_spin(self, value: float) -> QDoubleSpinBox:
spin = QDoubleSpinBox()
spin.setRange(0.0, 50.0)
spin.setDecimals(1)
spin.setSingleStep(1.0)
spin.setSuffix(" mm")
spin.setValue(value)
return spin
def build_pdf_export_html(
html_document: str,
headings: list[Heading],
page_numbers: dict[int, int] | None = None,
*,
include_page_markers: bool = True,
) -> str:
printable_headings = [heading for heading in headings if heading.level <= 3]
marked_document = _insert_heading_markers(html_document, printable_headings) if include_page_markers else html_document
toc_html = _toc_html(printable_headings, page_numbers or {})
print_css = _print_css(len(printable_headings))
marked_document = marked_document.replace("</style>", f"{print_css}\n </style>", 1)
body_class = ' class="pdf-export"'
marked_document = marked_document.replace("<body>", f"<body{body_class}>", 1)
body_marker = f'<div class="pdf-body-start-marker">{PDF_BODY_MARKER}</div>\n' if include_page_markers else ""
export_front_matter = f"\n{toc_html}\n{body_marker}"
return marked_document.replace(f"<body{body_class}>", f"<body{body_class}>{export_front_matter}", 1)
def extract_heading_page_numbers(pdf_data: bytes, heading_count: int) -> dict[int, int]:
from pypdf import PdfReader
reader = PdfReader(BytesIO(pdf_data))
page_texts = [(page.extract_text() or "") for page in reader.pages]
normalized_page_texts = [_normalized_marker_text(text) for text in page_texts]
normalized_body_marker = _normalized_marker_text(PDF_BODY_MARKER)
body_page_index = next(
(index for index, text in enumerate(normalized_page_texts) if normalized_body_marker in text),
None,
)
if body_page_index is None:
raise ValueError("无法在预打印 PDF 中定位正文起始页。")
page_numbers: dict[int, int] = {}
for heading_index in range(heading_count):
marker = _normalized_marker_text(f"{PDF_HEADING_MARKER_PREFIX}{heading_index:04d}")
page_index = next(
(index for index in range(body_page_index, len(normalized_page_texts)) if marker in normalized_page_texts[index]),
None,
)
if page_index is not None:
page_numbers[heading_index] = page_index + 1
if heading_count and len(page_numbers) != heading_count:
missing = sorted(set(range(heading_count)) - set(page_numbers))
raise ValueError(f"无法识别全部目录页码,缺少标题索引:{missing}")
return page_numbers
def _normalized_marker_text(value: str) -> str:
return re.sub(r"[^A-Z0-9]", "", value.upper())
def _insert_heading_markers(html_document: str, headings: list[Heading]) -> str:
marked = html_document
search_from = 0
for index, heading in enumerate(headings):
escaped_slug = html.escape(heading.slug, quote=True)
positions = [
position
for needle in (f'id="{escaped_slug}"', f"id='{escaped_slug}'")
if (position := marked.find(needle, search_from)) >= 0
]
if not positions:
continue
attribute_position = min(positions)
tag_end = marked.find(">", attribute_position)
if tag_end < 0:
continue
marker = (
'<span class="pdf-heading-marker">'
f"{PDF_HEADING_MARKER_PREFIX}{index:04d}"
"</span>"
)
marked = f"{marked[:tag_end + 1]}{marker}{marked[tag_end + 1:]}"
search_from = tag_end + len(marker) + 1
return marked
def _toc_html(headings: list[Heading], page_numbers: dict[int, int]) -> str:
density_class = " pdf-toc--dense" if len(headings) > 28 else ""
if len(headings) > 70:
density_class += " pdf-toc--very-dense"
rows = []
for index, heading in enumerate(headings):
title = html.escape(heading.title)
slug = html.escape(heading.slug, quote=True)
page_number = str(page_numbers.get(index, "--"))
rows.append(
f'<li class="pdf-toc__item pdf-toc__item--level-{heading.level}">'
f'<a href="#{slug}"><span class="pdf-toc__text">{title}</span>'
'<span class="pdf-toc__dots"></span>'
f'<span class="pdf-toc__page">{page_number}</span></a></li>'
)
items = "\n".join(rows) or '<li class="pdf-toc__empty">本文档没有可用的章节标题。</li>'
return (
f'<section class="pdf-toc{density_class}">'
'<h1>目录</h1>'
f'<ol class="pdf-toc__list">{items}</ol>'
"</section>"
)
def _print_css(heading_count: int) -> str:
column_count = 2 if heading_count > 28 else 1
return f"""
@media print {{
html,
body,
.reader-shell,
.reader {{
background: #ffffff !important;
background-image: none !important;
}}
html,
body {{
margin: 0 !important;
padding: 0 !important;
}}
.reader-shell {{
display: block !important;
width: 100% !important;
min-height: 0 !important;
padding: 0 !important;
}}
.reader {{
box-sizing: border-box !important;
width: 100% !important;
max-width: none !important;
min-height: 0 !important;
margin: 0 !important;
padding: 0 !important;
border: 0 !important;
box-shadow: none !important;
}}
.image-lightbox {{
display: none !important;
}}
.pdf-toc {{
display: block !important;
box-sizing: border-box;
width: 100%;
min-height: 100%;
color: #111827;
background: #ffffff !important;
break-after: page;
page-break-after: always;
}}
.pdf-toc h1 {{
margin: 0 0 10mm;
padding: 0 0 4mm;
border-bottom: 1px solid #cbd5e1;
color: #111827;
font-size: 24pt;
line-height: 1.2;
}}
.pdf-toc__list {{
column-count: {column_count};
column-gap: 10mm;
margin: 0;
padding: 0;
list-style: none;
}}
.pdf-toc__item {{
margin: 0 0 3.2mm;
break-inside: avoid;
page-break-inside: avoid;
font-size: 10.5pt;
line-height: 1.35;
}}
.pdf-toc--dense .pdf-toc__item {{
margin-bottom: 2.2mm;
font-size: 9.5pt;
}}
.pdf-toc--very-dense .pdf-toc__item {{
margin-bottom: 1.4mm;
font-size: 8.5pt;
}}
.pdf-toc__item a {{
display: flex;
align-items: baseline;
color: inherit !important;
text-decoration: none !important;
}}
.pdf-toc__item--level-2 {{ padding-left: 5mm; }}
.pdf-toc__item--level-3 {{ padding-left: 10mm; }}
.pdf-toc__text {{ max-width: 78%; }}
.pdf-toc__dots {{
min-width: 8mm;
margin: 0 2mm;
flex: 1;
border-bottom: 1px dotted #94a3b8;
}}
.pdf-toc__page {{
min-width: 8mm;
font-variant-numeric: tabular-nums;
text-align: right;
}}
.pdf-body-start-marker,
.pdf-heading-marker {{
color: #ffffff !important;
font-size: 1px !important;
line-height: 1px !important;
}}
.pdf-body-start-marker {{
height: 1px;
overflow: hidden;
}}
}}
"""