#!/usr/bin/env python3
"""Author deterministic unembedded Japanese CID H/V fixtures (ReportLab)."""
import hashlib
import json
from pathlib import Path

from pypdf import PdfReader
from reportlab.pdfbase import pdfmetrics
from reportlab.pdfbase.cidfonts import UnicodeCIDFont
from reportlab.pdfgen.canvas import Canvas

ROOT = Path(__file__).resolve().parent / "fonts"
ROOT.mkdir(exist_ok=True)
OUTPUT = ROOT / "unembedded-japanese.pdf"
TEXTS = ["日本語の文書を表示します。", "漢字とひらがな、カタカナ。", "「縦書き」の句読点を確認。"]
for face in ("HeiseiMin-W3", "HeiseiKakuGo-W5"):
    for vertical in (False, True):
        font = UnicodeCIDFont(face, isVertical=vertical)
        font.name = font.fontName = face + ("-V" if vertical else "-H")
        pdfmetrics.registerFont(font)

canvas = Canvas(str(OUTPUT), pagesize=(600, 760), invariant=1, pageCompression=0)
canvas.setTitle("DocView unembedded Japanese CID H/V fixture")
canvas.setAuthor("DocView synthetic test corpus")
for page, vertical in enumerate((False, True), 1):
    canvas.setFont("Helvetica-Bold", 16)
    canvas.drawString(36, 722, "Unembedded Japanese CID fonts")
    canvas.setFont("Helvetica", 11)
    canvas.drawString(36, 700, "UniJIS-UCS2-" + ("V: native vertical writing mode" if vertical else "H: horizontal writing mode"))
    for row, face in enumerate(("HeiseiMin-W3", "HeiseiKakuGo-W5")):
        canvas.setFont("Helvetica", 10)
        if vertical:
            canvas.drawString(330 - row * 280, 660, face)
            for col, text in enumerate(TEXTS):
                canvas.setFont(face + "-V", 20)
                canvas.drawString(520 - row * 280 - col * 62, 620, text)
        else:
            canvas.drawString(36, 648 - row * 260, face)
            canvas.setFont(face + "-H", 20)
            for line, text in enumerate(TEXTS):
                canvas.drawString(36, 610 - row * 260 - line * 42, text)
    canvas.setFont("Helvetica", 9)
    canvas.drawString(36, 28, f"Synthetic fixture / page {page} / no font program embedded")
    canvas.showPage()
canvas.save()

reader = PdfReader(OUTPUT)
encodings = set()
for page in reader.pages:
    for ref in page["/Resources"]["/Font"].values():
        font = ref.get_object()
        if font.get("/Subtype") != "/Type0":
            continue
        encodings.add(str(font["/Encoding"]))
        for descendant in font["/DescendantFonts"]:
            descriptor = descendant.get_object()["/FontDescriptor"].get_object()
            assert not any(key in descriptor for key in ("/FontFile", "/FontFile2", "/FontFile3"))
assert encodings == {"/UniJIS-UCS2-H", "/UniJIS-UCS2-V"}, encodings
manifest = {
    "schemaVersion": 1,
    "generator": "generate_fonts.py (ReportLab UnicodeCIDFont, invariant PDF)",
    "documents": [{
        "file": OUTPUT.name,
        "sha256": hashlib.sha256(OUTPUT.read_bytes()).hexdigest(),
        "pages": 2,
        "sizePt": [600, 760],
        "fontProgramsEmbedded": False,
        "encodings": sorted(encodings),
        "faces": ["HeiseiMin-W3", "HeiseiKakuGo-W5"],
        "expectedText": TEXTS,
        "writingModes": ["horizontal", "vertical"],
    }],
}
(ROOT / "manifest.json").write_text(json.dumps(manifest, ensure_ascii=False, indent=2) + "\n")
print(json.dumps(manifest, ensure_ascii=False, indent=2))