#!/usr/bin/env python3
import html
import re
import shutil
import struct
from pathlib import Path
from zipfile import ZipFile, ZIP_DEFLATED
WORKDIR = Path("/Applications/MAMP/htdocs/toolchat")
SOURCE_MD = WORKDIR / "HUONG_DAN_SU_DUNG_FLYCHAT.md"
REFERENCE_DOCX = Path("/Users/quangqa/.codex/plugins/cache/openai-curated-remote/openai-templates/0.1.1/skills/artifact-template-strategy-memorandum/assets/reference.docx")
OUTPUT_DOCX = WORKDIR / "HUONG_DAN_SU_DUNG_FLYCHAT.docx"
NS = (
'xmlns:wpc="http://schemas.microsoft.com/office/word/2010/wordprocessingCanvas" '
'xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006" '
'xmlns:o="urn:schemas-microsoft-com:office:office" '
'xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" '
'xmlns:m="http://schemas.openxmlformats.org/officeDocument/2006/math" '
'xmlns:v="urn:schemas-microsoft-com:vml" '
'xmlns:wp14="http://schemas.microsoft.com/office/word/2010/wordprocessingDrawing" '
'xmlns:wp="http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing" '
'xmlns:w10="urn:schemas-microsoft-com:office:word" '
'xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main" '
'xmlns:w14="http://schemas.microsoft.com/office/word/2010/wordml" '
'xmlns:wpg="http://schemas.microsoft.com/office/word/2010/wordprocessingGroup" '
'xmlns:wpi="http://schemas.microsoft.com/office/word/2010/wordprocessingInk" '
'xmlns:wne="http://schemas.microsoft.com/office/word/2006/wordml" '
'xmlns:wps="http://schemas.microsoft.com/office/word/2010/wordprocessingShape" '
'mc:Ignorable="w14 wp14"'
)
SECT_PR = (
''
''
''
''
''
)
def esc(text):
return html.escape(text, quote=False)
def image_size(path):
with path.open("rb") as f:
sig = f.read(65536)
if sig[:8] != b"\x89PNG\r\n\x1a\n":
if sig[:2] == b"\xff\xd8":
idx = 2
while idx + 9 < len(sig):
if sig[idx] != 0xFF:
idx += 1
continue
marker = sig[idx + 1]
idx += 2
if marker in (0xD8, 0xD9):
continue
size = int.from_bytes(sig[idx:idx + 2], "big")
if marker in range(0xC0, 0xC4):
return struct.unpack(">HH", sig[idx + 3:idx + 7])[1], struct.unpack(">HH", sig[idx + 3:idx + 7])[0]
idx += size
raise ValueError(f"Unsupported screenshot format: {path}")
return struct.unpack(">II", sig[16:24])
def text_runs(markdown_text):
text = markdown_text.replace("**", "")
parts = re.split(r"(`[^`]+`)", text)
runs = []
for part in parts:
if not part:
continue
mono = part.startswith("`") and part.endswith("`")
value = part[1:-1] if mono else part
font = '' if mono else ""
runs.append(f'{font}{esc(value)}')
return "".join(runs)
def paragraph(text="", style=None, bold=False, italic=False, color=None, before=0, after=120, line=276, align=None):
ppr = []
if style:
ppr.append(f'')
if align:
ppr.append(f'')
ppr.append(f'')
rpr = []
if bold:
rpr.append("")
if italic:
rpr.append("")
if color:
rpr.append(f'')
run = f'{"".join(rpr)}{esc(text)}' if (bold or italic or color) else text_runs(text)
return f'{"".join(ppr)}{run}'
def bullet(text, level=0):
left = 720 + level * 360
hanging = 360
return (
f''
f''
f''
f'• {text_runs(text)}'
)
def numbered(text, n, level=0):
left = 720 + level * 360
hanging = 360
return (
f''
f''
f''
f'{n}. {text_runs(text)}'
)
def quote(text):
return (
''
''
''
f'{text_runs(text)}'
)
def table(rows):
widths = [2300, 2600, 4460]
grid = "".join(f'' for w in widths)
trs = []
for idx, row in enumerate(rows):
cells = []
for cidx, cell in enumerate(row):
fill = "D9EAF7" if idx == 0 else "FFFFFF"
bold = "" if idx == 0 else ""
cells.append(
f''
f''
''
''
f'{bold}{esc(cell)}'
)
trs.append(f'{"".join(cells)}')
return (
''
''
''
''
''
f'{grid}{"".join(trs)}'
)
def image_paragraph(rid, path, target_name, caption):
px_w, px_h = image_size(path)
max_cx = 5_943_600
cx = min(max_cx, int(px_w * 914400 / 144))
cy = int(cx * px_h / px_w)
docpr_id = int(rid[3:]) + 100
drawing = f'''
'''
return (
''
f'{drawing}'
f'{paragraph(caption, italic=True, color="667085", after=180, align="center")}'
)
def parse_md(lines):
body = []
image_refs = []
table_rows = []
list_stack = []
in_table = False
for raw in lines:
line = raw.rstrip("\n")
if not line.strip():
if in_table and table_rows:
body.append(table(table_rows))
table_rows = []
in_table = False
continue
if line.startswith("|") and line.endswith("|"):
cells = [c.strip() for c in line.strip("|").split("|")]
if all(set(c) <= {"-", ":"} for c in cells):
continue
table_rows.append(cells)
in_table = True
continue
if in_table and table_rows:
body.append(table(table_rows))
table_rows = []
in_table = False
m = re.match(r"!\[(.*?)\]\((.*?)\)", line)
if m:
caption, rel = m.groups()
rid = f"rId{20 + len(image_refs)}"
img_path = (WORKDIR / rel).resolve()
target_name = f"flychat-screenshot-{len(image_refs) + 1:02d}.jpg"
image_refs.append((rid, img_path, target_name))
body.append(image_paragraph(rid, img_path, target_name, caption))
continue
h = re.match(r"^(#{1,3})\s+(.+)$", line)
if h:
level = len(h.group(1))
style = "Title" if level == 1 else f"Heading{min(level, 3)}"
body.append(paragraph(h.group(2), style=style, bold=(level == 1), before=180 if level > 1 else 0, after=160))
continue
if line.startswith("> "):
body.append(quote(line[2:]))
continue
m = re.match(r"^(\s*)-\s+(.+)$", line)
if m:
body.append(bullet(m.group(2), len(m.group(1)) // 2))
continue
m = re.match(r"^(\s*)(\d+)\.\s+(.+)$", line)
if m:
body.append(numbered(m.group(3), int(m.group(2)), len(m.group(1)) // 3))
continue
if line.strip() == "---":
body.append(paragraph("", after=80))
continue
body.append(paragraph(line, after=140))
if table_rows:
body.append(table(table_rows))
return body, image_refs
def rels_xml(existing, image_refs):
end = ""
additions = "".join(
f''
for rid, path, target_name in image_refs
)
return existing.replace(end, additions + end)
def content_types_xml(existing):
if 'Extension="jpg"' in existing:
return existing
marker = ''
addition = marker + ''
return existing.replace(marker, addition)
def header_xml(existing):
return existing.replace("Strategy Memo", "FlyChat User Guide")
def main():
lines = SOURCE_MD.read_text(encoding="utf-8").splitlines()
body, image_refs = parse_md(lines)
document_xml = f"{''.join(body)}{SECT_PR}"
tmp = OUTPUT_DOCX.with_suffix(".tmp.docx")
shutil.copy2(REFERENCE_DOCX, tmp)
with ZipFile(tmp, "r") as zin, ZipFile(OUTPUT_DOCX, "w", ZIP_DEFLATED) as zout:
written = set()
for item in zin.infolist():
data = zin.read(item.filename)
if item.filename == "word/document.xml":
data = document_xml.encode("utf-8")
elif item.filename == "word/_rels/document.xml.rels":
data = rels_xml(data.decode("utf-8"), image_refs).encode("utf-8")
elif item.filename == "[Content_Types].xml":
data = content_types_xml(data.decode("utf-8")).encode("utf-8")
elif item.filename == "word/header1.xml":
data = header_xml(data.decode("utf-8")).encode("utf-8")
if item.filename in written:
continue
zout.writestr(item, data)
written.add(item.filename)
for _, img_path, target_name in image_refs:
target = f"word/media/{target_name}"
if target not in written:
zout.write(img_path, target)
written.add(target)
tmp.unlink(missing_ok=True)
print(OUTPUT_DOCX)
if __name__ == "__main__":
main()