pengxiao
219232de74
以讲义内容为骨架迁移到标准目录格式： - officefile/ 主内容（12章 + 附录 + CC4SI补充） - dofile/ 代码示例（11个Python脚本） - data/ 图片资源 - output/ 生成输出（忽略） - Archive/ 归档旧目录（忽略） - .claude/skills/ 保留markdown-to-docx工具链 - .pandoc/ 保留CSL和本地化配置 Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
112 lines
3.0 KiB
Bash
112 lines
3.0 KiB
Bash
#!/usr/bin/env bash
|
||
set -euo pipefail
|
||
|
||
usage() {
|
||
cat >&2 <<'EOF'
|
||
Usage:
|
||
render_markdown_with_dotx.sh <source-md> <output-docx> <template-dotx-or-docx> [book-title] [resource-root] [shortcut-template]
|
||
EOF
|
||
exit 1
|
||
}
|
||
|
||
[[ $# -lt 3 || $# -gt 6 ]] && usage
|
||
|
||
SOURCE_MD="$1"
|
||
OUTPUT_DOCX="$2"
|
||
TEMPLATE_DOC="$3"
|
||
BOOK_TITLE="${4:-}"
|
||
SHORTCUT_TEMPLATE="${6:-}"
|
||
SOURCE_NAME="$(basename "$SOURCE_MD")"
|
||
|
||
if [[ ! -f "$SOURCE_MD" ]]; then
|
||
echo "Error: source markdown not found: $SOURCE_MD" >&2
|
||
exit 1
|
||
fi
|
||
|
||
if [[ ! -f "$TEMPLATE_DOC" ]]; then
|
||
echo "Error: template file not found: $TEMPLATE_DOC" >&2
|
||
exit 1
|
||
fi
|
||
|
||
if ! command -v pandoc >/dev/null 2>&1; then
|
||
echo "Error: pandoc is not installed or not in PATH." >&2
|
||
exit 1
|
||
fi
|
||
|
||
template_has_keymap_customizations() {
|
||
python3 - "$1" <<'PY'
|
||
import sys
|
||
import zipfile
|
||
|
||
try:
|
||
with zipfile.ZipFile(sys.argv[1]) as zf:
|
||
raise SystemExit(0 if "word/customizations.xml" in zf.namelist() else 1)
|
||
except Exception:
|
||
raise SystemExit(1)
|
||
PY
|
||
}
|
||
|
||
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
|
||
SOURCE_DIR="$(cd "$(dirname "$SOURCE_MD")" && pwd -P)"
|
||
SOURCE_PARENT="$(cd "$SOURCE_DIR/.." && pwd -P)"
|
||
RESOURCE_ROOT="${5:-$SOURCE_PARENT}"
|
||
|
||
if [[ -z "$SHORTCUT_TEMPLATE" ]] && template_has_keymap_customizations "$TEMPLATE_DOC"; then
|
||
SHORTCUT_TEMPLATE="$TEMPLATE_DOC"
|
||
fi
|
||
|
||
mkdir -p "$(dirname "$OUTPUT_DOCX")"
|
||
|
||
TMP_DIR="$(mktemp -d)"
|
||
trap 'rm -rf "$TMP_DIR"' EXIT
|
||
|
||
NORMALIZED_MD="$TMP_DIR/normalized.md"
|
||
TMP_RESOURCES_DIR="$TMP_DIR/resources"
|
||
|
||
perl -0pe '
|
||
s{^图：([^\n!]+)!\[\[([^]|]+)\|[0-9]+\]\]}{\n\n图：$1}mg;
|
||
s{!\[\[([^]|]+)\|[0-9]+\]\]}{}g;
|
||
s{!\[\[([^]|]+)\]\]}{}g;
|
||
s{^\*([图表]：[^\n*]+)\*$}{$1}mg;
|
||
s{(?m)^(!\[[^\n]*\]\([^\n]+\))$}{\n$1\n}g;
|
||
s{(?m)^([图表]：[^\n]+)$}{\n$1\n}g;
|
||
s{(?m)^(\*\*[^\n*]+\*\*)$}{\n$1\n}g;
|
||
s{^---$}{}mg;
|
||
s{^(#{3,})\s+\d+\.\d+(?:\.\d+)?\s+(小结|可执行清单)}{$1 $2}mg;
|
||
s{\n{3,}}{\n\n}g;
|
||
' "$SOURCE_MD" > "$NORMALIZED_MD"
|
||
|
||
python3 "$SCRIPT_DIR/render_mermaid_blocks_for_docx.py" \
|
||
"$NORMALIZED_MD" \
|
||
"$SOURCE_NAME" \
|
||
"$RESOURCE_ROOT" \
|
||
"$TMP_RESOURCES_DIR"
|
||
|
||
# Auto-fix missing table/figure captions before conversion
|
||
python3 "$SCRIPT_DIR/validate_captions.py" fix "$NORMALIZED_MD"
|
||
|
||
CHAPTER_TITLE="$(sed -n 's/^# //p' "$NORMALIZED_MD" | head -n 1)"
|
||
CHAPTER_PREFIX="$(printf '%s\n' "$CHAPTER_TITLE" | perl -ne 'print "$1\n" if /(第[0-9]+章)/')"
|
||
|
||
HEADER_TEXT="${CHAPTER_TITLE:-Markdown Export}"
|
||
if [[ -n "$BOOK_TITLE" ]]; then
|
||
HEADER_TEXT="《${BOOK_TITLE}》"
|
||
fi
|
||
if [[ -n "$BOOK_TITLE" && -n "$CHAPTER_PREFIX" ]]; then
|
||
HEADER_TEXT="${HEADER_TEXT}${CHAPTER_PREFIX}"
|
||
fi
|
||
|
||
RESOURCE_PATH="$TMP_DIR:$TMP_RESOURCES_DIR:$SOURCE_DIR:$SOURCE_PARENT:$RESOURCE_ROOT:$RESOURCE_ROOT/resources"
|
||
|
||
pandoc "$NORMALIZED_MD" \
|
||
-f markdown \
|
||
-t docx \
|
||
--reference-doc="$TEMPLATE_DOC" \
|
||
--lua-filter="$SCRIPT_DIR/template_style_filter.lua" \
|
||
--resource-path="$RESOURCE_PATH" \
|
||
-o "$OUTPUT_DOCX"
|
||
|
||
"$SCRIPT_DIR/postprocess_template_docx.py" "$OUTPUT_DOCX" "$TEMPLATE_DOC" "$HEADER_TEXT" "$SHORTCUT_TEMPLATE"
|
||
|
||
printf 'OK\t%s\n' "$OUTPUT_DOCX"
|