PlantAI-dev/manuscript/plantai_aiia_zh/scripts/test_dual_html.py

import base64
import tempfile
import unittest
from html.parser import HTMLParser
from pathlib import Path

from build_dual_html import (
    EXPERIMENT_OUTPUT,
    MANUSCRIPT_OUTPUT,
    ROOT,
    build_experiment_plan,
    build_manuscript_reader,
    document_shell,
    figure_card,
    image_data_uri,
    inline_markup,
    markdown_blocks,
    markdown_table,
)


class CountingParser(HTMLParser):
    def __init__(self):
        super().__init__()
        self.ids = []
        self.images = 0
        self.tables = 0

    def handle_starttag(self, tag, attrs):
        values = dict(attrs)
        if "id" in values:
            self.ids.append(values["id"])
        if tag == "img":
            self.images += 1
        if tag == "table":
            self.tables += 1


class MarkdownPrimitiveTests(unittest.TestCase):
    def test_inline_markup_escapes_and_formats(self):
        source = "**PlantAI** uses `Manifest` & [AIIA](https://example.org)."
        rendered = inline_markup(source)
        self.assertIn("<strong>PlantAI</strong>", rendered)
        self.assertIn("<code>Manifest</code>", rendered)
        self.assertIn("&amp;", rendered)
        self.assertIn('href="https://example.org"', rendered)

    def test_markdown_table_is_semantic(self):
        source = "| A | B |\n|---|---|\n| x | y |\n"
        rendered = markdown_table(source)
        self.assertIn("<thead>", rendered)
        self.assertIn("<tbody>", rendered)
        self.assertIn("<th>A</th>", rendered)
        self.assertIn("<td>y</td>", rendered)

    def test_markdown_blocks_supports_heading_list_quote_and_paragraph(self):
        source = "## 标题\n\n> 提示\n\n- 项目一\n- 项目二\n\n正文。\n"
        rendered = markdown_blocks(source)
        self.assertIn("<h2", rendered)
        self.assertIn("<blockquote>", rendered)
        self.assertIn("<ul>", rendered)
        self.assertIn("<p>正文。</p>", rendered)

    def test_png_is_embedded_as_data_uri(self):
        with tempfile.TemporaryDirectory() as tmp:
            path = Path(tmp) / "plot.png"
            payload = b"\x89PNG\r\n\x1a\nexample"
            path.write_bytes(payload)
            uri = image_data_uri(path)
        self.assertEqual(
            uri,
            "data:image/png;base64," + base64.b64encode(payload).decode("ascii"),
        )


class ManuscriptReaderTests(unittest.TestCase):
    def test_document_shell_is_standalone(self):
        page = document_shell(
            "T", "E", "S", ["P"], [("x", "X")], ['<section id="x">X</section>']
        )
        self.assertIn("<!doctype html>", page.lower())
        self.assertIn("@media print", page)
        self.assertNotRegex(page, r'<(?:link|script)[^>]+(?:href|src)="https?://')

    def test_simulated_figure_card_has_visible_warning(self):
        png = ROOT / "figures/exports/figure2_rnaseq_scenarios_SIMULATED.png"
        card = figure_card(2, "RNA-seq", "caption", png, True)
        self.assertIn("SIMULATED DATA — DRAFT ONLY", card)
        self.assertIn("data:image/png;base64,", card)

    def test_manuscript_reader_contains_five_figures_and_three_tables(self):
        page = build_manuscript_reader()
        parser = CountingParser()
        parser.feed(page)
        self.assertEqual(parser.images, 5)
        self.assertGreaterEqual(parser.tables, 3)
        self.assertIn("PlantAI-dev", page)
        self.assertIn("PlantAI-RAG", page)
        self.assertGreaterEqual(page.count("SIMULATED DATA — DRAFT ONLY"), 8)


class ExperimentPlanTests(unittest.TestCase):
    def test_experiment_plan_has_17_numbered_sections_and_all_assets(self):
        page = build_experiment_plan()
        parser = CountingParser()
        parser.feed(page)
        for number in range(17):
            self.assertIn(f'id="plan-{number}"', page)
        self.assertEqual(parser.images, 5)
        self.assertGreaterEqual(parser.tables, 3)
        self.assertGreaterEqual(page.count("SIMULATED DATA — DRAFT ONLY"), 8)

    def test_experiment_plan_contains_five_experiments_and_reviewer_views(self):
        page = build_experiment_plan()
        for phrase in (
            "系统级可执行闭环",
            "多场景RNA-seq",
            "RNA-seq到候选家族",
            "轻量Extension",
            "Graph RAG",
            "审稿人视角",
            "生信分析师视角",
            "智能体工程师视角",
        ):
            self.assertIn(phrase, page)

    def test_experiment_plan_does_not_claim_excluded_capabilities(self):
        page = build_experiment_plan()
        self.assertIn("Diagnosis Agent不纳入论文贡献", page)
        self.assertIn("agentic RAG未进入当前主路径", page)
        self.assertIn("重测序—GWAS仅作为后续方向", page)


class OutputIntegrityTests(unittest.TestCase):
    def test_both_outputs_are_self_contained_and_parseable(self):
        for path in (MANUSCRIPT_OUTPUT, EXPERIMENT_OUTPUT):
            page = path.read_text(encoding="utf-8")
            parser = CountingParser()
            parser.feed(page)
            self.assertEqual(parser.images, 5)
            self.assertEqual(len(parser.ids), len(set(parser.ids)))
            self.assertNotIn("file://", page)
            self.assertNotRegex(page, r'<img[^>]+src=["\'](?:/|https?://)')
            self.assertNotRegex(page, r"\{\{[A-Z_][A-Z0-9_]*\}\}")
            self.assertIn("@media print", page)

    def test_required_terms_and_simulation_labels_are_present(self):
        for path in (MANUSCRIPT_OUTPUT, EXPERIMENT_OUTPUT):
            page = path.read_text(encoding="utf-8")
            for term in (
                "Artificial Intelligence in Agriculture",
                "PlantAI-dev",
                "PlantAI-RAG",
                "Graph RAG",
                "Extension",
                "SIMULATED DATA — DRAFT ONLY",
            ):
                self.assertIn(term, page)


if __name__ == "__main__":
    unittest.main()