Data & method

The food data

Nutrient values for single ingredients are taken from published composition tables — principally the Indian Food Composition Tables 2017, published by the National Institute of Nutrition (ICMR) in Hyderabad, which is the current authoritative analysis of Indian foods measured on Indian samples.

Citation: Longvah T, Ananthan R, Bhaskarachary K, Venkaiah K. Indian Food Composition Tables 2017. National Institute of Nutrition, Indian Council of Medical Research, Hyderabad. Nutrient values are measured scientific quantities, reproduced here as data with attribution.

IFCT 2017 does not carry every fasting food. It has no entry for buckwheat, barnyard millet, fox nut, sago or arrowroot, and the vrat flours are milled products rather than the whole seeds it analysed. Where IFCT has no row, this site uses USDA FoodData Central or a named peer-reviewed composition study, and says which. Where nothing usable exists, the page says the value is unverified rather than dressing an estimate as a citation.

What "verified" means here, food by food

This is the part of the site most worth checking, so it is stated per food rather than in general. Every entry below carries one of four labels, and the same label appears on the food's own page:

• Verified — the value was read off a named row of a named table, or a named paper. The food code or identifier is given.
• Recipe analysis — a cooked dish that appears in no composition table. Built from an ingredient list, each ingredient taken from a sourced row above, divided by the finished weight. The ingredient list is printed on the dish's page.
• Not individually verified — carried over from the tables when the site was built, and not re-checked line by line in the July 2026 audit of the fasting-specific foods. Indicative, not cited. This bucket is now empty: every food that was in it has been read off a named table row or moved to one of the labels above or below.
• Unverified — no table and no paper could be found for this exact food. The number is an estimate and is labelled as one wherever it appears.

The July 2026 audit, and what it found

Two rounds of checking, against the tables rather than against this site's own arithmetic. The first covered the ten vrat flours and starches. The second covered every remaining food that had not been individually checked — the tubers, the vegetables, the fruit and three dairy rows.

Both rounds found the same defect, and it is worth describing because it is invisible to every internal check this site runs. Several foods were carrying carbohydrate by difference — which is everything left after protein, fat, ash and water, and therefore includes the fibre — in the column this site labels available carbohydrate, which excludes it. The energy figure had then been reconciled against that inflated carbohydrate, so the pair was internally consistent and sailed through the Atwater check below. Two wrong numbers that agree with each other pass every test a site can run on itself. Only an outside source catches it.

The worst case was boiled potato, a core vrat food, which shipped at 97 kcal and 22.6 g of carbohydrate against IFCT's 69.8 kcal and 14.9 g — an energy overstatement of about 39% on something that appears in half the dishes here. Kuttu was out by 8.4 g of carbohydrate for the same reason. Every dish built on either of them moved when they were corrected.

Because of that, the Indian tables are now preferred wherever they have the food: IFCT 2017 publishes available carbohydrate directly, so nothing has to be converted and nothing can be conflated. Where USDA is the only source, it publishes carbohydrate by difference and the fibre is subtracted here to give available carbohydrate — that subtraction is stated in the note on every affected food.

An em dash is not a zero

Composition tables print a dash when a nutrient was never determined. Makhana's crude fibre is the case in point: ICMR–NIN never measured it. Until July 2026 this site shipped that as 0, which was explained in the footnote on makhana's own page and was still wrong everywhere else — a zero sorts last in a fibre ranking, compares as "contains none", and reads as a measurement. It now renders as an em dash, and any food whose value was never measured is excluded from rankings and comparisons for that nutrient rather than ordered as though it carried none. The same rule applies to any dish built entirely from ingredients with that gap.

Shared with ThaliFacts

Where a food also appears on ThaliFacts, the values here are copied from that dataset rather than re-derived, so the two sites cannot disagree about what a katori of curd carries. Banana, curd, ghee, almonds, jaggery, sabudana khichdi, sabudana vada and around thirty others come across unchanged.

Composite vrat dishes

Kuttu ki puri, samak pulao, singhare ka halwa and the rest are recipes, not ingredients, so they do not appear in any composition table. They are built by recipe analysis: a normal home recipe written out ingredient by ingredient, each ingredient's composition taken from a sourced row above, and the result divided by the finished weight. Where a dish is built this way the ingredient list and the finished weight are printed on its page, so the arithmetic can be checked and argued with. The fat assumption is always stated, because on a vrat the ghee is usually the largest single variable — a heavier hand moves a dish by 30 to 50%.

Energy for a recipe-analysed dish is the Atwater sum of its ingredients rather than an independently measured figure, which is why those rows reconcile exactly.

The energy check

Energy has to agree with the macronutrients that produce it. Every build runs each food through Atwater factors — 4 kcal per gram of protein, 4 per gram of available carbohydrate, 9 per gram of fat — and refuses to publish if the stated energy is more than 15% away from that sum, or 20% for high-fibre foods. All 83 entries currently reconcile. A hand-entered dataset will contain typos; this is what stops one becoming a wrong page.

How the observance pages are written

Three rules govern every description on this site:

• Commonly observed, never required. Customs differ by family, region, community and sampradaya. Every observance page carries an explicit note saying so, written specifically for that observance rather than pasted in.
• No prescription. Nothing here tells anyone their tradition is wrong, and nothing suggests modifying a religious practice for a fitness reason. Where training or blood-sugar guidance could collide with an observance, it is offered as an option and the reader's tradition and doctor decide.
• Specific safety, not a generic footer. Every observance page names the particular risk of that observance — heat for Nirjala Ekadashi, cold-water immersion and 36 hours for Chhath, sleep loss for Shivratri, medication timing for a monthly Ekadashi.

Why there are no fasting or prayer times

Sehri, iftar, moonrise, sunset and the start and end of any religious fast are locally and religiously determined. Publishing a wrong one would cause real harm — someone eating when they should have stopped, or going without when they need not have. This site computes none of them, and the build actively refuses to ship a page containing a clock time outside the personal intermittent-fasting calculator, which is not a religious schedule. Use your local mosque's timetable or your community's announced panchang.

What this site cannot tell you

• What your tradition requires. It describes what is common. Those are different things, and only the people you would normally ask can answer the second.
• Whether you personally should fast. That is a medical question and a religious one, in that order for safety and in your own order for meaning.
• What is in your kitchen. A heavier hand with the ghee moves a vrat dish by 30 to 50%. The assumption is stated so you can adjust it yourself.
• Regional variation in recipes. Sabudana khichdi in Maharashtra is not sabudana khichdi in Uttar Pradesh. One common version is used.

Corrections

Corrections to food values are prioritised, and corrections to how a practice is described are prioritised above those. If something here misrepresents your tradition, tell me and it will be fixed on the next build. General information only — not medical or religious guidance. Fasting is not safe for everyone, particularly during pregnancy or with diabetes or other conditions. Speak to a doctor first.

Every source, with its publication and last-checked date →

ਸਰੋਤ·Published vrat traditions + IFCT 2017·composition tables and observance accounts·ਸਮੀਖਿਆ ਜੁਲਾਈ 2026