from __future__ import annotations
from datetime import date
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
FIXTURES_DIR = ROOT / "benches" / "fixtures"
SAMPLES = [
("commonmark-5k.md", 5 * 1024),
("commonmark-20k.md", 20 * 1024),
("commonmark-50k.md", 50 * 1024),
]
def build_article(target_bytes: int, label: str) -> str:
parts: list[str] = []
current_bytes = 0
header = (
"<!-- Generated on "
f"{date.today().isoformat()} to ~{label}. Synthetic wiki-style article. "
"Do not edit by hand. -->\n\n"
)
parts.append(header)
current_bytes += len(header.encode("utf-8"))
def add(text: str) -> None:
nonlocal current_bytes
parts.append(text)
current_bytes += len(text.encode("utf-8"))
def add_para(text: str) -> None:
add(text.rstrip("\n") + "\n\n")
def compact_base_blocks() -> list[str]:
return [
(
"# Northbridge District\n\n"
"Northbridge District is a fictional metropolitan area used in technical "
"documentation to describe planning, governance, and infrastructure patterns.\n\n"
"The article is written in a Wikipedia-like style, with many paragraphs, "
"lists, and code blocks. It also demonstrates CommonMark features such as "
"reference links, autolinks, inline HTML, and blockquotes.\n\n"
),
(
"Overview\n========\n\n"
"Northbridge sits along a river delta and connects industrial corridors to "
"residential zones. The local charter emphasizes long-term planning and "
"public access.\n\n"
"Key characteristics:\n\n"
"- Mixed-use zoning across historic corridors\n"
"- **High-frequency** transit loops with *steady* headways\n"
"- Public archives with long-term audit trails\n"
" - Urban catalog revisions\n"
" - Civic ledger backups\n\n"
),
(
"## History\n\n"
"Settlement began near the riverbank warehouses, followed by rail expansion "
"and market squares. Records refer to the \\*operator\\* policy and "
"Northbridge District as a formal term.\n\n"
"> \"We planned for continuity as much as for growth.\" — Committee Minutes, 1987\n\n"
"### Milestones\n\n"
"1. 1889 — First bridge completed.\n"
"2. 1957 — Flood response plan adopted.\n"
"3. 1984 — Density plan revised.\n\n"
),
(
"## Infrastructure\n\n"
"Operators publish schedules at <https://example.org/northbridge/transit> "
"and accept inquiries at <mailto:info@example.org>.\n\n"
"In maintenance guides, a hard line break is used to separate shift notes. \n"
"This line continues on the next row with explicit formatting.\n\n"
),
(
"## Notes on Formatting\n\n"
"Some records use character entities like © and & to preserve "
"licensing text. Backslash escapes such as \\[brackets\\] and \\_underscores\\_ "
"appear in transcripts.\n\n"
"Inline tags like <span class=\"label\">draft</span> show up in civic memos.\n\n"
"<div class=\"infobox\">\n"
"<p><strong>Northbridge Gallery</strong></p>\n"
"<p>Founded: 1979</p>\n"
"</div>\n\n"
),
(
"## Education and Research\n\n"
"A representative data transformation is included below:\n\n"
"```rust\n"
"fn normalize_score(score: i32) -> i32 {\n"
" if score < 0 { 0 } else { score }\n"
"}\n"
"```\n\n"
"Legacy memos still use indented code blocks for configuration samples:\n\n"
" [archive]\n"
" retention_years = 25\n"
" checksum = true\n\n"
),
(
"## See also\n\n"
"- [Regional planning overview](https://example.org/region)\n"
"- [Northbridge data portal][portal]\n"
"- ![District map](https://example.org/assets/map.png \"Map\")\n\n"
"---\n\n"
),
]
def full_base_blocks() -> list[str]:
blocks = compact_base_blocks()
blocks.extend(
[
(
"## Geography\n\n"
"Northbridge occupies a low-lying delta bordered by tidal flats and a "
"steep northern ridge. The main river splits into distributaries that "
"form natural boundaries between neighborhoods.\n\n"
"### Neighborhoods\n\n"
"- Harbor Row\n"
"- Eastbank\n"
"- Civic Plateau\n"
"- Archive Ward\n"
" - Old Registry\n"
" - Survey Annex\n\n"
),
(
"## Climate\n\n"
"The climate is temperate with long wet seasons and mild summers. "
"Annual rainfall supports an extensive greenway system along the river.\n\n"
),
(
"## Demographics\n\n"
"The population is diverse, with several long-established communities "
"and a steady influx of students and researchers.\n\n"
),
(
"## Economy\n\n"
"The local economy centers on logistics, planning services, and archival "
"technology. Small manufacturing remains present along the rail corridor.\n\n"
"Several firms specialize in planning software. One dataset is described "
"in the [Municipal Archive][archive].\n\n"
),
(
"## Culture\n\n"
"Northbridge has a strong public arts program, with murals documenting "
"planning eras and neighborhood transitions.\n\n"
),
(
"## Governance\n\n"
"The governing council publishes a yearly plan outlining priorities for "
"public transit, housing stability, and record preservation.\n\n"
"The most cited policy is the [Civic Charter][charter].\n\n"
"### Council structure\n\n"
"- Chair\n"
"- Planning director\n"
"- Archive commissioner\n\n"
),
(
"## Media\n\n"
"Local media outlets focus on infrastructure, planning decisions, and "
"archival releases. A common phrase in editorials is **\"plan for today\"**.\n\n"
),
]
)
return blocks
base_blocks = compact_base_blocks() if target_bytes <= 8 * 1024 else full_base_blocks()
for block in base_blocks:
add(block)
add("## Supplementary Notes\n\n")
def block_transit(idx: int) -> str:
return (
f"### Transit Loop {idx}\n\n"
"The loop schedule is coordinated with river traffic and freight windows. "
"Operators track **turnaround time** and *dwell time* at major stations.\n\n"
"- Peak interval: 6 minutes\n"
"- Off-peak interval: 12 minutes\n"
"- Night service: 20 minutes\n"
" - Limited stops at Harbor Row\n"
" - Signal priority near Eastbank\n\n"
)
def block_archives(idx: int) -> str:
return (
f"### Archive Protocol {idx}\n\n"
"Archival metadata is recorded with a consistent schema. A typical entry "
"includes `series_id`, a retention flag, and a public access marker.\n\n"
"> Records are validated nightly, and only then indexed for search.\n"
"> Use [Municipal Archive][archive] for official requests.\n\n"
)
def block_lists(idx: int) -> str:
return (
f"### Field Survey {idx}\n\n"
"Survey teams use a shared checklist for neighborhood audits:\n\n"
"1. Measure sidewalk clearance.\n"
"2. Record lighting levels.\n"
"3. Note accessibility markers.\n"
"\n"
" - ADA ramp condition\n"
" - Crosswalk timing\n\n"
)
def block_fenced(idx: int) -> str:
return (
f"### Routing Script {idx}\n\n"
"A small script is used to normalize incoming schedules:\n\n"
"```python\n"
"def normalize(minutes):\n"
" return max(0, int(minutes))\n"
"```\n\n"
)
def block_indented(idx: int) -> str:
return (
f"### Legacy Config {idx}\n\n"
"Older maintenance forms embed raw configuration blocks:\n\n"
" [loop]\n"
" interval = 12\n"
" priority = true\n\n"
)
def block_inline_html(idx: int) -> str:
return (
f"### Exhibit Note {idx}\n\n"
"Curators often mark temporary exhibits with inline tags like "
"<span class=\"tag\">rotating</span> to aid indexing.\n\n"
"<aside>\n"
"<p>Exhibit schedule updates are posted monthly.</p>\n"
"</aside>\n\n"
)
def block_emphasis(idx: int) -> str:
return (
f"### Planning Memo {idx}\n\n"
"Reports include escaped markers such as \\*priority\\* and "
"\\_status\\_ in transcripts. Several lines use forced breaks. \n"
"The second line keeps the break visible in the output.\n\n"
)
def block_links(idx: int) -> str:
return (
f"### Public Outreach {idx}\n\n"
"Notices refer to the [Civic Charter][charter] and an online portal at "
"<https://example.org/northbridge/outreach>. A separate memo references "
"[Planning Office][planning].\n\n"
)
def block_quote(idx: int) -> str:
return (
f"### Council Excerpt {idx}\n\n"
"> We will revisit the flood plan after the next cycle.\n"
"> - Follow-up audits\n"
"> - Community feedback\n\n"
)
def block_separator(_: int) -> str:
return "---\n\n"
blocks = [
block_transit,
block_archives,
block_lists,
block_fenced,
block_indented,
block_inline_html,
block_emphasis,
block_links,
block_quote,
block_separator,
]
ref_defs = (
"[archive]: https://example.org/archive \"Municipal Archive\"\n"
"[planning]: https://example.org/planning 'Planning Office'\n"
"[charter]: https://example.org/charter (Civic Charter)\n"
"[portal]: https://example.org/data \"Data Portal\"\n"
)
ref_defs_len = len(ref_defs.encode("utf-8"))
idx = 1
while current_bytes < target_bytes - ref_defs_len:
block = blocks[(idx - 1) % len(blocks)]
add(block(idx))
idx += 1
add(ref_defs)
return "".join(parts)
def main() -> None:
FIXTURES_DIR.mkdir(parents=True, exist_ok=True)
for filename, target in SAMPLES:
label = filename.split(".")[0].replace("commonmark-", "")
sample = build_article(target, label)
path = FIXTURES_DIR / filename
path.write_text(sample, encoding="utf-8")
print(f"Wrote {path} ({path.stat().st_size} bytes)")
if __name__ == "__main__":
main()