{ "category": "Our 10 featured functions", "display_name": "FILTER", "entry_kind": "built_in_function", "eval_api": { "method": "POST", "url": "https://handbook.dnacalc.com/api/eval" }, "function_id": "FUNC.FILTER", "implementation_state": { "answer_bank": null, "detail": "Recorded: 42 workbook cells.", "dnacalc_commit": "9e1884369a678158dd82d3d74188daf986995e87", "in_catalogue": true, "state": "matches", "state_label": "Matches every recorded Excel answer", "summary": "Matches all 42 recorded answers", "workbook_corpus": { "cells": 42, "matched": 42 } }, "implementations": [ { "commit": "9e1884369a678158dd82d3d74188daf986995e87", "flavour": "excel-compatibility-candidate", "language": "Rust", "name": "CalcFunc", "own_code": [ "https://github.com/DnaCalc/DnaCalc/blob/9e1884369a678158dd82d3d74188daf986995e87/crates/calc-func/src/catalogue/arrays.rs#L633-L636", "https://github.com/DnaCalc/DnaCalc/blob/9e1884369a678158dd82d3d74188daf986995e87/crates/calc-func/src/ported/dynamic_array_reshape_family.rs#L222-L229", "https://github.com/DnaCalc/DnaCalc/blob/9e1884369a678158dd82d3d74188daf986995e87/crates/calc-func/src/ported/dynamic_array_reshape_family.rs#L629-L670", "https://github.com/DnaCalc/DnaCalc/blob/9e1884369a678158dd82d3d74188daf986995e87/crates/calc-func/src/ported/dynamic_array_reshape_family.rs#L672-L722" ], "repository": "https://github.com/DnaCalc/DnaCalc", "shared_code": [ "catalogue/arrays::source_error", "catalogue/arrays::prepared", "catalogue/arrays::prepared_one", "catalogue/arrays::result", "catalogue/arrays::run", "ported/dynamic_array_reshape_family::scalar_cell", "ported/dynamic_array_reshape_family::materialize_array_arg", "ported/dynamic_array_reshape_family::shape", "ported/dynamic_array_reshape_family::get", "ported/dynamic_array_reshape_family::build_array", "ported/dynamic_array_reshape_family::map_dynamic_array_reshape_error_to_ws" ], "source": "https://github.com/DnaCalc/DnaCalc/blob/9e1884369a678158dd82d3d74188daf986995e87/crates/calc-func/src/catalogue/arrays.rs#L633-L636", "state": "matches", "suite": null, "summary": "Matches all 42 recorded answers" } ], "json_url": "https://handbook.dnacalc.com/api/functions/FUNC.FILTER.json", "name": "FILTER", "notes": [], "schema": "efh.function-reference/v1", "section_order": [ "what-it-does", "try-it", "quirks", "how-to-implement", "version-differences", "alternatives" ], "sections": { "alternatives": "\n### Relationships\n\n\n- **`SORT` and `UNIQUE`** are the other two members of the classic dynamic-array trio;\n `FILTER` then `SORT` then `UNIQUE` is the canonical query pipeline, and all three live in the\n same reference-implementation module.\n- **`CHOOSEROWS` / `CHOOSECOLS`** select by index where `FILTER` selects by predicate. When the\n predicate is really \"positions 1, 4 and 9\", the index functions say it directly.\n- **`TAKE` / `DROP`** select contiguous edge bands, which is the other common special case.\n- **Advanced Filter and the database functions (`DSUM` and family)** are the pre-dynamic-array\n answers to the same question, with criteria expressed as a range of criteria rows rather than\n as a boolean array.\n- **`XLOOKUP`** answers \"the one matching row\" where `FILTER` answers \"all matching rows\".\n Reaching for `FILTER` and then taking the first row is a common but weaker idiom, because it\n computes everything to use one.\n- **`SUMIFS` / `COUNTIFS`** aggregate under criteria without materializing the matching rows;\n when you only want a total, they do less work and produce no spill.\n\n\n", "how-to-implement": "\n### Notes for implementers\n\n- **Axis inference from `include`'s shape is the crux.** The rule has to distinguish a\n column-shaped criterion from a row-shaped one, and to reject anything that is neither — with\n `#VALUE!`, per the documentation. A 1×1 criterion is ambiguous for a 1×1 array and needs a\n declared policy.\n- **The empty result must be intercepted before the array is constructed**, exactly as in\n `DROP`: a zero-row array is not representable, so the `if_empty`-or-`#CALC!` decision belongs\n at the point where the survivor count is known.\n- **Truth-value interpretation of `include` is per-family policy**, not engine law. The shared\n model is explicit that to-logical behaviour is distributed across functions rather than\n centralized, so an implementation must declare its rule for numbers, text, blanks and errors\n in the criterion rather than inheriting one.\n- **Stability.** Preserving input order is part of the semantics, not an implementation\n accident; any parallel or partitioned implementation must be order-preserving.\n\n\n", "quirks": "\n### Result and edge cases\n\n\nReturn kind: `Array` — the surviving rows (or columns), spilled. The result's width equals\n`array`'s width when filtering rows, and its height equals `array`'s height when filtering\ncolumns.\n\n- **Every row passes** — the result equals the input.\n- **Exactly one row passes** — a one-row array, not a scalar. Downstream code that expects a\n scalar should not assume; use `@` or a reducer.\n- **No row passes** — the empty case, handled by `if_empty` or by `#CALC!` (below).\n- **Errors inside `include`** are a different matter: an error in the criterion array has no\n truth value. What `FILTER` does with one is not established here, and it is a case real\n workbooks hit constantly, because criterion arrays are usually computed.\n- **A `FILTER` over a whole-column reference** carries the whole column's blanks into the\n criterion, which is why the idiom is usually paired with a non-blank test.\n\n\n\n### Errors\n\n\nMicrosoft's page documents two conditions:\n\n- `#CALC!` when no entries in `include` are true and `if_empty` is omitted. This is the\n empty-array case: the value model admits no empty array at the published boundary\n ([value universe](../model/01-value-universe.md)), so \"nothing matched\" has to surface as an\n error unless you supplied a substitute.\n- `#VALUE!` when `include` is not of a dimension compatible with `array`.\n\nThe `#CALC!` case is the one that changes how formulas must be written: any `FILTER` whose\ncriterion depends on data should either carry an `if_empty` or be wrapped. Note that `FILTER`\nis unusual in the reshape family for offering `if_empty` at all — `DROP` and its siblings give\nyou `#CALC!` with no substitute.\n\nAn error value supplied as an argument propagates under the universal rule in\n[coercion and lifting](../model/02-coercion-and-lifting.md).\n\n\n", "sources": "\n- Microsoft, FILTER function —\n \n (argument definitions, the `if_empty` substitute, the `#CALC!` empty-result condition, and the\n `#VALUE!` dimension-incompatibility condition).\n- Handbook `content/model/01-value-universe.md` (`#CALC!`; no empty array at the published\n boundary).\n- Handbook `content/model/02-coercion-and-lifting.md` (to-logical as per-family policy;\n elementwise lifting of comparisons; error propagation).\n- Handbook `content/model/03-call-pipeline.md` (`ValuesOnlyPreAdapter`; host-side spill\n adaptation).\n- Handbook `data/functions/FUNC.FILTER.json` and `data/presence/FUNC.FILTER.json` (arity,\n category placement, classification axes, shared reshape-family module).\n\n", "version-differences": "No comparison between specific Excel builds is recorded here yet. This does not establish that all builds behave alike.", "what-it-does": "\n`FILTER(array, include, [if_empty])` keeps the slices of `array` for which the corresponding\nelement of `include` is true, and drops the rest. Order is preserved: the surviving slices come\nback in their original sequence.\n\nThe word doing the work is *slices*. `FILTER` does not select individual cells; it selects whole\nrows or whole columns, and **which axis it filters is decided by the shape of `include`, not by\nan argument**:\n\n- `include` shaped like a single column with one entry per row of `array` — rows are filtered;\n- `include` shaped like a single row with one entry per column of `array` — columns are\n filtered.\n\nThat is the single most important thing to understand about this function, and it is the source\nof most `#VALUE!` results people meet with it: the criterion array must be parallel to the axis\nyou mean to filter, and getting it transposed does not filter the other way, it fails.\n\n`include` is an array of truth values, typically produced by a comparison over a range —\n`(A2:A100=\"North\")` — which is itself an elementwise lifted operation\n([coercion and lifting](../model/02-coercion-and-lifting.md)). Combining criteria is arithmetic\non those arrays: `*` for AND, `+` for OR. `FILTER` supplies no criteria language of its own,\nwhich is why it composes so freely and why it has no equivalent of the `COUNTIF` family's\ncriteria strings.\n\n\n\n### Arguments\n\n\n| Argument | Meaning |\n|---|---|\n| `array` | The array or range to filter. Required. |\n| `include` | A boolean array parallel to the filtered axis of `array`. Required. |\n| `if_empty` | The value to return when nothing passes the filter. Optional. |\n\nNotes on each:\n\n- **`array`** is prepared under `ValuesOnlyPreAdapter`, so a reference is resolved to its values\n before the function runs; the result is values and carries no reference structure.\n- **`include`** must have exactly one entry per row (or per column) of `array`. Non-logical\n values are interpreted by ordinary to-logical rules — zero is false, non-zero true — but which\n contexts accept text logicals is per-family policy in the shared model rather than a universal\n rule, and the Handbook has not established `FILTER`'s policy.\n- **`if_empty`** is the argument people omit and then wish they had not. Without it, an empty\n result is an error (below). With it, `FILTER(…, …, \"\")` or `FILTER(…, …, \"none\")` degrades\n gracefully. It accepts any scalar; whether it accepts an array is not established here.\n\n\n" }, "signature": { "display": "FILTER(array, include, [if_empty])", "parameters": [ { "description": null, "name": "array", "optional": false, "repeats": false }, { "description": null, "name": "include", "optional": false, "repeats": false }, { "description": null, "name": "if_empty", "optional": true, "repeats": false } ], "trailing_repeats": false }, "sources": { "metadata": { "admission": { "label": "implemented", "reason": null }, "arity": { "max": 3, "min": 2 }, "category": "Our 10 featured functions", "classification": { "arg_preparation_profile": "ValuesOnlyPreAdapter", "coercion_lift_profile": "Custom", "determinism_class": "Deterministic", "error_collapse_profile": "None", "fec_dependency_profile": "None", "host_interaction_class": "None", "kernel_signature_class": "Custom", "rich_value_usage": "RichBlind", "surface_fec_dependency_profile": "RefOnly", "thread_safety_class": "SafePure", "volatility_class": "NonVolatile" }, "descriptions": { "long": null, "short": "The rows or columns of an array for which a condition holds", "short_source": "dna-calc-own" }, "docs": { "english_description": "Use this function to filter a range of data based on criteria you define.", "microsoft_url": "https://support.microsoft.com/en-us/office/filter-function-f4f7cb66-82eb-4767-8f7c-4877ad80c759" }, "entry_kind": "built_in_function", "localized_family_names": null, "localized_names": [ { "locale": "ar-SA", "name": "FILTER" }, { "locale": "bg-BG", "name": "FILTER" }, { "locale": "cs-CZ", "name": "FILTER" }, { "locale": "da-DK", "name": "FILTRER" }, { "locale": "de-DE", "name": "FILTER" }, { "locale": "el-GR", "name": "FILTER" }, { "locale": "en-AU", "name": "FILTER" }, { "locale": "en-GB", "name": "FILTER" }, { "locale": "en-US", "name": "FILTER" }, { "locale": "es-ES", "name": "FILTRAR" }, { "locale": "et-EE", "name": "FILTER" }, { "locale": "fi-FI", "name": "SUODATA" }, { "locale": "fr-FR", "name": "FILTRE*" }, { "locale": "he-IL", "name": "FILTER" }, { "locale": "hr-HR", "name": "FILTER" }, { "locale": "hu-HU", "name": "SZŰRŐ" }, { "locale": "id-ID", "name": "FILTER" }, { "locale": "it-IT", "name": "FILTER" }, { "locale": "ja-JP", "name": "FILTER" }, { "locale": "ko-KR", "name": "FILTER" }, { "locale": "lt-LT", "name": "FILTER" }, { "locale": "lv-LV", "name": "FILTER" }, { "locale": "nb-NO", "name": "FILTER" }, { "locale": "nl-NL", "name": "FILTER" }, { "locale": "pl-PL", "name": "FILTRUJ" }, { "locale": "pt-BR", "name": "FILTER" }, { "locale": "pt-PT", "name": "FILTRAR" }, { "locale": "ro-RO", "name": "FILTER" }, { "locale": "ru-RU", "name": "ФИЛЬТР" }, { "locale": "sk-SK", "name": "FILTER" }, { "locale": "sl-SI", "name": "FILTER" }, { "locale": "sr-Latn-RS", "name": "FILTER" }, { "locale": "sv-SE", "name": "FILTER" }, { "locale": "th-TH", "name": "FILTER" }, { "locale": "tr-TR", "name": "FİLTRE" }, { "locale": "uk-UA", "name": "FILTER" }, { "locale": "vi-VN", "name": "FILTER" }, { "locale": "zh-CN", "name": "FILTER" }, { "locale": "zh-HK", "name": "FILTER" }, { "locale": "zh-TW", "name": "FILTER" } ], "signature": { "display": "FILTER(array, include, [if_empty])", "parameters": [ { "description": null, "name": "array", "optional": false, "repeats": false }, { "description": null, "name": "include", "optional": false, "repeats": false }, { "description": null, "name": "if_empty", "optional": true, "repeats": false } ], "trailing_repeats": false }, "surface_name": "FILTER", "version_marker": null, "xlcall": null }, "microsoft": "https://support.microsoft.com/en-us/office/filter-function-f4f7cb66-82eb-4767-8f7c-4877ad80c759" }, "try_it": { "formula": "=FILTER({1;2;3},{TRUE;FALSE;TRUE})" }, "url": "https://handbook.dnacalc.com/functions/FILTER.html", "variants": [], "versions": [] }