ਖੋਜੀ ਉਪਜੈ ਬਾਦੀ ਬਿਨਸੈ ਹਉ ਬਲਿ ਬਲਿ ਗੁਰ ਕਰਤਾਰਾ ॥
Sri Guru Granth Sahib Ji, Ang 1255
Start typing on the left: first letters of a tuk, a word or phrase, or build an akhar with its lagaan-matra under Letters & Signs.
Distribution across Raags and Authors
ਤਤਕਰਾ ਰਾਗਾਂ ਕਾ Index of Raags & Banis
ਬਾਣੀਕਾਰ Contributors
Counts are of tuks (verse-lines) excluding headings. Attribution follows the headings (ਮਹਲਾ / ਮਃ, bhagat titles) as they occur in the text.
ਲਿਪੀ-ਅੰਕੜੇ Orthographic Inventory of Sri Guru Granth Sahib Ji
Every word of the text is segmented into written akhar (base letter + pairin + lagaan-matra). The tables below count those akhar. Click any row or cell to open the matching tuks.
Letter × Sign matrix ਅੱਖਰ × ਲਗ
Each cell counts akhar in which the row letter carries the column sign (alone or alongside other signs). Independent vowels are counted under their bearer: ਓ = ੳ + ੋ.
Lagaan-matra & diacritics
Letters bearing two or more signs
Subscripts (pairin) & half letters
Conjunct akhar with pairin
The Gurbani Search App is a study instrument for the text of Sri Guru Granth Sahib Ji.
Akhar model
Each written akhar is analysed as base letter + pairin + signs:
- Base: one of the 35 letters of the painti and ੜ. Independent vowels are analysed as bearer + sign, so ਆ = ਅ + ਾ, ਈ = ੲ + ੀ, ਓ = ੳ + ੋ.
- Pairin: a letter written small beneath the base: pairin haha ੍ਹ (ਪੜ੍ਹਹਿ, ਵਰ੍ਹਿਆ), ੍ਰ ੍ਵ, the half yayya ੍ਯ੍ਯ (ਧ੍ਯ੍ਯਾਨੁ, ਰਖ੍ਯ੍ਯਾ), and the rarer ੍ਨ ੍ਚ ੍ਟ (ਇਸ੍ਨਾਨੁ, ਨਿਸ੍ਚਲ, ਦ੍ਰਿਸ੍ਟਿ).
- Signs: kanna, sihari, bihari, aunkar, dulainkar, lavan, dulavan, hora, kanaura, bindi, tippi, visarga, and the halant under a letter: under ਨ ਮ ੜ ਲ in ਗਾਵਨ੍ਹਿ, ਤਿਨ੍ਹ, ਤੁਮ੍ਹ, and under ਸ before ਤ in ਮਸ੍ਤਕ, ਹਸ੍ਤ.
The well-known double marking of hora with aunkar (ਤੋੁਹੀ, ਲੋੁਭਾਨੁ, ਜੋੁਲਾਹਾ, ਓੁਹੀ) is encoded in the text as ੋ followed by ੁ on the same letter. The engine reads the set of signs on a letter, not their order, so any combination can be searched: choose “Only these signs” for that pair alone, or “With other signs too” to also allow further signs.
Word model
A word is the run of akhar between two spaces, read exactly as written, with every lagaan-matra, pairin and halant. So ਨਾਮੁ, ਨਾਮ and ਨਾਮਿ are three different word-forms, each with its own count.
- Whole word (ਪੂਰਾ ਸ਼ਬਦ) finds tuks containing that word-form; several words find tuks containing all of them, in any order.
- Part of a word (ਸ਼ਬਦ ਦਾ ਹਿੱਸਾ) finds every word that begins with, ends with, contains or is exactly the typed part: ਸਤਿ finds ਸਤਿਗੁਰ, ਸਤਿਗੁਰੂ, ਸਤਿਨਾਮੁ.
- The halant and the pairin haha may be typed either way (ਤਿਨ੍ਹ or ਤਿਨ੍); a word with pairin haha is kept apart from the same word without it (ਪੜ੍ਹਹਿ is not ਪੜਹਿ).
- Every result lists the word-forms found, with how often each occurs; click one to see only its tuks.
- Under Letters & Signs, an akhar can be placed within the word: first letter, in-between, last letter, or a word of one akhar only. A one-akhar word (ਸੋ, ਤੂ) counts as a first letter.
First-letter search
The first akhar of each word is folded to its base letter, and vowels to their bearer (ਆ ਐ ਔ → ਅ; ਇ ਈ ਏ → ੲ; ਉ ਊ ਓ → ੳ), as in conventional Gurbani software. Words and letters may also be typed in English as they sound: satgur, waheguru, shabad. The usual variant spellings are accepted (aa/a, ee/i, oo/u, sh/s, bh/b), and short vowels may be left out. For first letters, each English letter stands for the Gurmukhi letters it can spell, e.g. t for ਟ ਠ ਤ ਥ.
Attribution and its limits
Raag sections follow the Tatkara at the front of the Granth. Authorship is inferred from headings (ਮਹਲਾ ੧–੯, ਮਃ, ਕਬੀਰ ਜੀਉ and so on). It is reliable for the great majority of the text but is a computational reading, not an authority; the Bhatt Swayye are grouped under one heading, and composite shabads follow their opening heading. Always confirm against the saroop.
Built as a self-contained offline app: open index.html in any modern browser. The data file is regenerated from the docx with tools/build_data.py.