How many poems are in the database?

Roughly 78,000 unique poems and ci-lyrics. The page itself ships 300+ hand-annotated classics covering Tang shi, Song ci and Yuan qu — 唐诗 (from 蘅塘退士《唐诗三百首》 1763), 宋词 (from 朱孝臧《宋词三百首》 1924) and 元曲 (from 关汉卿/马致远/白朴/张养浩/张可久 etc.) — with translation, notes, theme tags, and grade level — plus a 1,000-poem "canon" chunk (唐诗三百首 + 宋词三百首 + 全部诗经 + 全部楚辞, ≈190 KB gzipped, loaded on first open). Hitting the "search full corpus" button or picking 唐 / 宋 from the dynasty filter loads the rest on demand — the 全唐诗 chunk is ≈56,000 poems / 6 MB gzipped, the 全宋词 chunk is ≈21,000 ci-lyrics / 3 MB gzipped. The 210 curated entries are the only ones with translations and notes; the larger corpus gives breadth and exact-text search so you can find any line, but doesn't carry classroom-style glosses for every entry.

Where does the data come from?

The 80,000-poem corpus is from the open-source chinese-poetry/chinese-poetry GitHub repository (MIT-licensed, the largest community-curated Chinese classical poetry dataset). 全唐诗 raw shards are in traditional Chinese; we run them through OpenCC TSCharacters to convert to simplified at build time so search works for users typing on a mainland keyboard. The 210 hand-annotated classics live in the source code itself and were verified against at least two reference editions (中华书局 / 上海古籍 / 唐诗三百首通行本). Modern translations are interpretive paraphrases written in plain modern Mandarin — the way a 语文 teacher would explain the poem — not literal academic glosses. We explicitly did NOT use an LLM to generate any poem text, because the failure mode is hallucinated "lost" Li Bai lines that read plausible but are simply fake.

Why these poets specifically?

We followed the 部编版 (PRC Ministry of Education) recommended reading list for grades 1-12 first, then layered in the most-frequent classics that show up in college entrance exam 高考 question banks since 2010. That means dataset weight tracks real classroom value, not aesthetic preference — so Li Bai and Du Fu dominate the Tang section, Su Shi and Li Qingzhao dominate the Song section, and we include the politically/historically loaded pieces (杜甫《茅屋为秋风所破歌》, 文天祥《过零丁洋》, 陆游《示儿》) that show up in essay prompts. Less-frequently-tested but historically major poets (李贺, 韩愈, 元稹) get 1-3 representative pieces.

Can I get the same "poem of the day" tomorrow?

No — by design. The poem-of-the-day function uses your local calendar date as a deterministic seed (YYYY-MM-DD → simple integer hash → index into the curated dataset), so every visitor in the same timezone sees the same poem on the same day, but the poem changes at local midnight. This makes the daily pick shareable ("我们今天都看到了《静夜思》") without any server-side state, while preventing same-poem-twice-in-a-row monotony. The daily pick rotates through the 210-poem curated set, so you'll cycle through it over roughly 7 months — long enough that the same poem rarely repeats, short enough that every famous classic eventually surfaces.

Are the translations literal or interpretive?

Interpretive — the goal is what a 中学语文 teacher would say to a student trying to understand the poem for the first time, not a word-by-word academic gloss. So 床前明月光,疑是地上霜 becomes "床前洒下一片明亮的月光,远远望去好像地上结了一层白霜" rather than the literal "bed-front bright-moon-light, suspect is ground-on frost". The optional 注释 (annotation) field adds the key character glosses (e.g. 疑 = 好像, 举头 = 抬头) when individual words would block comprehension. For competition-level literal translations consult 中华书局 or 上海古籍 academic editions.

Chinese Classical Poetry Search — 300+ Hand-Annotated Tang/Song/Yuan Canon + 80,000 Searchable Poems

Chinese classical poetry — 300+ hand-annotated Tang/Song/Yuan canon (唐诗宋词元曲) plus 80,000+ searchable Tang/Song/Shijing/Chuci poems on demand.

Runs locally
Category Text
Best for Removing repetitive cleanup work from everyday writing and operations.

309 of 309 poems · 309 hand-annotated classics · 0 more in the full corpus when expanded

What this tool does

Look up Chinese classical poetry (古诗词) directly in your browser — no API call, no signup, no ads in the way. Two-layer dataset: 300+ hand-annotated classics covering Tang shi, Song ci and Yuan qu (唐诗宋词元曲) ship in the page itself (full modern-Mandarin translation, key character glosses, theme tags, grade level), and 80,000+ more poems from the public-domain corpus load on demand. Coverage spans the entire Tang corpus 全唐诗 (≈56,000 poems), the Song ci-lyric corpus 全宋词 (≈21,000), the canonical 唐诗三百首 / 宋词三百首 selections, the Shijing 诗经 (305 poems) and the Chuci 楚辞 (≈50 poems). Search three ways: by author (李白 surfaces hundreds of Li Bai pieces), by title fragment (静夜思), or by any line you remember (床前明月光). Filter further by theme, dynasty (唐/宋/元/明/清/先秦), or grade level. Picking a dynasty filter or hitting "search full corpus" triggers the matching chunk to download — the full Tang corpus is a one-time 6 MB gzipped load that stays cached for the rest of your session. A deterministic "poem of the day" surfaces a different classic each calendar day — same poem for every visitor on the same date, no cache trickery. 100% client-side: every query runs locally, your search history never leaves the browser. We explicitly do NOT use an LLM to generate poems, because the failure mode is plausible-sounding fake "Li Bai" lines that would mislead exactly the students this tool exists to help.

Tool details

Input: Form fields; The page exposes text boxes, numeric controls, file pickers, or structured inputs depending on the tool.
Output: Live result + Download; The result area focuses on usable output, with copy, download, or preview actions when supported.
Privacy: Browser-side processing; The main tool logic does not call an external API, so inputs normally stay in the current tab.
Save / share: No account required; Open the page and use it; whether results survive refresh depends on the tool.
Performance budget: Initial JS <= 60 KB; No WASM budget is declared, keeping the tool quick to open on mobile.
Best fit: Text · Student; Category and role tags drive related tools, internal links, and quick fit checks.

How to use

1. Input

Paste or drop your content into the tool panel.
2. Process

Click the button. All processing is local in your browser.
3. Copy / Download

Copy the result or download to disk in one click.

How Chinese Poetry Search fits into your work

Use it to clean, compare, reshape, or extract plain text before it goes into a document, CMS, spreadsheet, or prompt.

Text jobs

Removing repetitive cleanup work from everyday writing and operations.
Making text easier to compare, paste, publish, or feed into another tool.
Working with content locally when the text is private or unfinished.

Text checks

Scan for unintended whitespace, duplicate lines, and lost punctuation.
For long text, test the first few lines before applying the whole change.
Copy the final output only after checking the preview.

Good next steps

These links move the current task into a more complete workflow.

FAQ

Tool combos

Folks in your role tend to reach for these alongside this tool.

Browse all tools for this role

Chinese Classical Poetry Search — 300+ Hand-Annotated Tang/Song/Yuan Canon + 80,000 Searchable Poems

What this tool does

Tool details

How to use

1. Input

2. Process

3. Copy / Download

How Chinese Poetry Search fits into your work

Text jobs

Text checks

Good next steps

FAQ

Chinese Idiom Search

Chinese Pinyin Converter

Traditional ⇄ Simplified Chinese Converter

Lunar Calendar Converter (Chinese Nongli)

Chinese Name Generator

Chinese Stroke Counter

Chinese Zodiac Fortune (Daily / Monthly / Yearly)

BaZi (Four Pillars) Calculator

Chinese Acupoint Locator

Add Line Numbers

AI Eval Planner

AI Model Comparison