nyangsum 0.1.0

Korean cat-sound lorem ipsum generator: 냥, 냐옹, 냐앙, 냐아앙.

Nyangsum

English | 한국어

Korean cat-sound lorem ipsum generator. Produces text composed of “냥/냐옹/나앙/냐아앙 …” variants.

Key properties:

• Tokens: 냥, 냐옹, 나앙(요청 반영), 늘임형 “냐” + “아”*k + (“앙”|“옹”)
• Default distribution: 냥 0.45, 냐옹 0.30, 나앙 0.15, 늘임형 0.10
• Punctuation: sentences can include commas, inner !/?/..., end with [., !, ?, ...], quotes occasionally
• Deterministic mode via seed()

Install (local dev)

python -m pip install -e .

Quickstart

Python:

from nyangsum import nyang

print(nyang.words(5)) # e.g., "냥 냐옹 냥 냥 냐옹" print(nyang.sentence(words=8)) # one phrase with exactly 8 words print("\n\n".join(nyang.paragraphs(2))) # two paragraphs

CLI:

Exactly N phrases of M words

python3 -m nyangsum --phrases 2 --words 6

N sentences (word count sampled in [min, max])

python3 -m nyangsum --sentences 3 --min-words 5 --max-words 10

N paragraphs, optionally HTML-wrapped

python3 -m nyangsum --paragraphs 2 --html

Deterministic output with seed

python3 -m nyangsum --phrases 1 --words 10 --seed 42

Examples

$ python3 -m nyangsum --phrases 1 --words 6 냥 냐옹 냥, 냐옹 나앙 냥.

$ python3 -m nyangsum --paragraphs 1 --html

냐옹 냥 냥 냥? 냥 냐옹 냐옹 냥 냥 냥. 냥 냥 냥...

API

The public API mirrors common lorem-ipsum ergonomics but outputs Korean cat sounds.

• nyang.seed(value: int | None) -> None Seed the internal RNG for reproducible output.

• nyang.word(max_len=6, capitalize=False, weights: dict | None = None, elong_min_a=1, elong_max_a=4, tail_weights=None) -> str Generate one token:

  • Base tokens favored: "냥", "냐옹", "나앙"
  • Elongated variants: "냐" + "아"*k + ("앙" | "옹") with k in [elong_min_a, elong_max_a]
  • weights override example: {"냥":0.5, "냐옹":0.3, "나앙":0.15, "elongated":0.05}

• nyang.words(n=1, as_list=False, sep=" ") -> str | list[str] Generate n words (joined by sep unless as_list=True).

• nyang.phrase(words_count: int, punctuation=True) -> str Exactly words_count tokens, internal punctuation possible, ends with [., !, ?, ...], may wrap in quotes.

• nyang.phrases(n: int, words_count: int, as_list=False, punctuation=True) -> str | list[str] Generate n phrases, each with words_count words. Joined by newline unless as_list=True.

• nyang.sentence(words: int | None = None, word_range=(4, 12), punctuation=True) -> str Convenience alias for a single phrase. If words is None, samples uniformly from word_range.

• nyang.sentences(n=1, words: int | None = None, word_range=(4, 12), as_list=False, punctuation=True) -> str | list[str] Multiple sentences. Joined by space unless as_list=True.

• nyang.paragraph(sentence_range=(3, 7)) -> str Paragraph with a sampled number of sentences.

• nyang.paragraphs(n=1, sentence_range=(3, 7), as_list=False) -> str | list[str] Multiple paragraphs (double-newline separated unless as_list=True).

• nyang.text(paragraphs_count=3, sentence_range=(3, 7), html=False) -> str Produce multiple paragraphs; html=True wraps each as

  ...

  .

CLI

Console module: python3 -m nyangsum

• Selection

  • --phrases N --words M Generate N phrases, each with exactly M words
  • --sentences N Generate N sentences (pair with --words or use --min-words/--max-words)
  • --paragraphs N Generate N paragraphs (pair with --html optionally)
  • --words M Generate exactly M words (standalone)

• Ranges

  • --min-words X --max-words Y
  • --min-sentences A --max-sentences B

• Formatting

  • --html Wrap paragraphs in

    ...

    (only with --paragraphs)
  • --no-punct Disable punctuation for phrases/sentences

• Reproducibility / tuning

  • --seed INT
  • --weights JSON Override base/elongated weights e.g., --weights '{"냥":0.5,"냐옹":0.3,"나앙":0.15,"elongated":0.05}'

Notes:

• Default behavior: prints a single paragraph of 500 words (one long phrase).
• Base tokens “냥/냐옹/나앙” are most frequent by default; elongated variants appear occasionally.
• Length caps (max_len) keep visual width reasonable.

Rationale

• Follows the reference “meowsum” structure and UX in Python.
• Korean-specific tokens with adjustable weights to favor 냥/냐옹/나앙.

License

MIT