| 1 | package calc |
| 2 | |
| 3 | // fn_cond_apply.go: the six conditional aggregates themselves. The |
| 4 | // criteria machinery they share is in fn_cond.go; split for the 300-line |
| 5 | // cap, and along the seam that already existed — one file decides |
| 6 | // whether a cell matches, the other decides what to do about it. |
| 7 | |
| 8 | // condPair is one criteria range and the criterion applied to it. |
| 9 | type condPair struct { |
| 10 | over rect |
| 11 | crit criterion |
| 12 | } |
| 13 | |
| 14 | // selected walks the first pair's rectangle and calls fn for every |
| 15 | // offset where EVERY pair matches. |
| 16 | // |
| 17 | // Offsets, not addresses. Each further criteria range is reshaped to the |
| 18 | // first one's dimensions (offsetRect), so ranges of different written |
| 19 | // sizes still line up cell for cell — Excel's rule, and the one that |
| 20 | // makes SUMIFS over ranges somebody wrote inconsistently give the answer |
| 21 | // they expect rather than #VALUE!. |
| 22 | func selected(pairs []condPair, env Env, fn func(dr, dc int)) { |
| 23 | base := pairs[0].over |
| 24 | for dr := 0; dr <= base.r1-base.r0; dr++ { |
| 25 | for dc := 0; dc <= base.c1-base.c0; dc++ { |
| 26 | all := true |
| 27 | for _, p := range pairs { |
| 28 | at := offsetRect(p.over, base) |
| 29 | if !p.crit.matches(cellValueAt(env, at.r0+dr, at.c0+dc)) { |
| 30 | all = false |
| 31 | break |
| 32 | } |
| 33 | } |
| 34 | if all { |
| 35 | fn(dr, dc) |
| 36 | } |
| 37 | } |
| 38 | } |
| 39 | } |
| 40 | |
| 41 | // readPairs decodes (range, criteria) argument couples starting at |
| 42 | // argNodes[from], stepping two at a time. |
| 43 | // |
| 44 | // ok is false for a bad count, a non-range where a range must be, or a |
| 45 | // range over the cell cap — the same three refusals fn_lookup.go makes, |
| 46 | // answered the same way, because a caller cannot tell them apart and |
| 47 | // should not have to. |
| 48 | func readPairs(argNodes []Node, from int, env Env) ([]condPair, Value, bool) { |
| 49 | rest := argNodes[from:] |
| 50 | if len(rest) == 0 || len(rest)%2 != 0 { |
| 51 | return nil, errValue(ErrValue), false |
| 52 | } |
| 53 | var pairs []condPair |
| 54 | for i := 0; i < len(rest); i += 2 { |
| 55 | box, ok := rectOf(rest[i]) |
| 56 | if !ok { |
| 57 | return nil, errValue(ErrValue), false |
| 58 | } |
| 59 | c := Eval(rest[i+1], env) |
| 60 | if c.Kind == KindErr { |
| 61 | return nil, c, false |
| 62 | } |
| 63 | pairs = append(pairs, condPair{over: box, crit: parseCriterion(c)}) |
| 64 | } |
| 65 | // Every criteria range must have the same SHAPE as the first — |
| 66 | // width and height, not area. |
| 67 | // |
| 68 | // Area was the first version of this check and it was wrong in the |
| 69 | // way that matters: A1:A3 is 3x1 and A1:C1 is 1x3, so the two have |
| 70 | // the same cell count, the pair was accepted, and offsetRect |
| 71 | // reshaped one onto the other. The result was a plausible number |
| 72 | // read out of cells nobody named — COUNTIFS(A1:A3,">5",A1:C1,"<>") |
| 73 | // answered 3. Caught by the branch review; LibreOffice answers |
| 74 | // Err:502 to that expression and excelize #VALUE!, so both oracles |
| 75 | // say refuse. |
| 76 | for _, p := range pairs[1:] { |
| 77 | if p.over.r1-p.over.r0 != pairs[0].over.r1-pairs[0].over.r0 || |
| 78 | p.over.c1-p.over.c0 != pairs[0].over.c1-pairs[0].over.c0 { |
| 79 | return nil, errValue(ErrValue), false |
| 80 | } |
| 81 | } |
| 82 | return pairs, Value{}, true |
| 83 | } |
| 84 | |
| 85 | // total adds up the values at the selected offsets of `over`, counts |
| 86 | // them, and reports the first error it met, so SUM and AVERAGE share one |
| 87 | // walk. |
| 88 | // |
| 89 | // The error return is the fix for a real inconsistency: SUM over a range |
| 90 | // holding a #DIV/0! answers #DIV/0!, and SUMIF over the same range |
| 91 | // answered 40 — it skipped the error along with the text and the blanks. |
| 92 | // That contradicted this function's own comment, which said a |
| 93 | // conditional sum must not disagree with an unconditional one about what |
| 94 | // a number is. An error is not a value to skip; it is the answer. |
| 95 | // |
| 96 | // Neither oracle could settle it. excelize does not propagate at all — |
| 97 | // it answers 40 to SUM(A1:A3) over the same range, so it disagrees with |
| 98 | // our corpus-validated SUM before the conditional forms are even |
| 99 | // reached — and the corpus generator refuses to record an error as a |
| 100 | // value. So this follows the invariant already written here and Excel's |
| 101 | // documented behaviour, and says so rather than claiming an oracle it |
| 102 | // does not have. |
| 103 | func total(pairs []condPair, over rect, env Env) (sum float64, n int, bad Value) { |
| 104 | at := offsetRect(over, pairs[0].over) |
| 105 | base := pairs[0].over |
| 106 | |
| 107 | // The error scan is SEPARATE from the selection walk, and it has to |
| 108 | // be. An error matches no criterion — criterion.matches returns |
| 109 | // false for one, deliberately, so a bad formula does not inflate a |
| 110 | // COUNTIF — so a cell holding one is never selected and the walk |
| 111 | // below never sees it. The first version of this fix put the check |
| 112 | // inside the walk and changed nothing at all. |
| 113 | // |
| 114 | // Excel propagates an error anywhere in the sum range, matched or |
| 115 | // not, which is the same rule SUM follows over the whole range. |
| 116 | for dr := 0; dr <= base.r1-base.r0 && bad.Kind != KindErr; dr++ { |
| 117 | for dc := 0; dc <= base.c1-base.c0; dc++ { |
| 118 | if v := cellValueAt(env, at.r0+dr, at.c0+dc); v.Kind == KindErr { |
| 119 | bad = v |
| 120 | break |
| 121 | } |
| 122 | } |
| 123 | } |
| 124 | if bad.Kind == KindErr { |
| 125 | return 0, 0, bad |
| 126 | } |
| 127 | |
| 128 | selected(pairs, env, func(dr, dc int) { |
| 129 | v := cellValueAt(env, at.r0+dr, at.c0+dc) |
| 130 | if v.Kind != KindNum { |
| 131 | // Text, blanks and logicals read from a reference are |
| 132 | // skipped, exactly as SUM skips them (fn_math.go's |
| 133 | // taggedValue comment). |
| 134 | return |
| 135 | } |
| 136 | sum += v.Num |
| 137 | n++ |
| 138 | }) |
| 139 | return sum, n, bad |
| 140 | } |
| 141 | |
| 142 | // ---- SUMIF / COUNTIF / AVERAGEIF ------------------------------------ |
| 143 | // |
| 144 | // The singulars take (range, criteria, [apply_range]) with the applied |
| 145 | // range LAST and optional. The plurals take (apply_range, range1, |
| 146 | // criteria1, …) with it FIRST and required. That inconsistency is |
| 147 | // Excel's, not ours, and it is the single most common way these get |
| 148 | // written wrongly — so the two shapes are decoded by separate functions |
| 149 | // rather than by one with a flag. |
| 150 | |
| 151 | func fnSumIf(argNodes []Node, env Env) Value { |
| 152 | pairs, over, bad, ok := singular(argNodes, env) |
| 153 | if !ok { |
| 154 | return bad |
| 155 | } |
| 156 | sum, _, bad := total(pairs, over, env) |
| 157 | if bad.Kind == KindErr { |
| 158 | return bad |
| 159 | } |
| 160 | return Value{Kind: KindNum, Num: sum} |
| 161 | } |
| 162 | |
| 163 | func fnCountIf(argNodes []Node, env Env) Value { |
| 164 | if len(argNodes) != 2 { |
| 165 | return errValue(ErrValue) |
| 166 | } |
| 167 | pairs, bad, ok := readPairs(argNodes, 0, env) |
| 168 | if !ok { |
| 169 | return bad |
| 170 | } |
| 171 | n := 0 |
| 172 | selected(pairs, env, func(int, int) { n++ }) |
| 173 | return Value{Kind: KindNum, Num: float64(n)} |
| 174 | } |
| 175 | |
| 176 | func fnAverageIf(argNodes []Node, env Env) Value { |
| 177 | pairs, over, bad, ok := singular(argNodes, env) |
| 178 | if !ok { |
| 179 | return bad |
| 180 | } |
| 181 | sum, n, bad := total(pairs, over, env) |
| 182 | if bad.Kind == KindErr { |
| 183 | return bad |
| 184 | } |
| 185 | if n == 0 { |
| 186 | // Excel's answer for an average of nothing, and deliberately |
| 187 | // not zero: zero is a number somebody would act on. |
| 188 | return errValue(ErrDiv0) |
| 189 | } |
| 190 | return Value{Kind: KindNum, Num: sum / float64(n)} |
| 191 | } |
| 192 | |
| 193 | // singular decodes the (range, criteria, [apply_range]) shape and |
| 194 | // returns the rectangle to aggregate over — the criteria range itself |
| 195 | // when no third argument was written, which is what makes |
| 196 | // SUMIF(A1:A9,">5") sum the cells it tested. |
| 197 | func singular(argNodes []Node, env Env) ([]condPair, rect, Value, bool) { |
| 198 | if len(argNodes) != 2 && len(argNodes) != 3 { |
| 199 | return nil, rect{}, errValue(ErrValue), false |
| 200 | } |
| 201 | pairs, bad, ok := readPairs(argNodes[:2], 0, env) |
| 202 | if !ok { |
| 203 | return nil, rect{}, bad, false |
| 204 | } |
| 205 | over := pairs[0].over |
| 206 | if len(argNodes) == 3 { |
| 207 | box, ok := rectOf(argNodes[2]) |
| 208 | if !ok { |
| 209 | return nil, rect{}, errValue(ErrValue), false |
| 210 | } |
| 211 | over = box |
| 212 | } |
| 213 | return pairs, over, Value{}, true |
| 214 | } |
| 215 | |
| 216 | // ---- SUMIFS / COUNTIFS / AVERAGEIFS --------------------------------- |
| 217 | |
| 218 | func fnSumIfs(argNodes []Node, env Env) Value { |
| 219 | over, pairs, bad, ok := plural(argNodes, env) |
| 220 | if !ok { |
| 221 | return bad |
| 222 | } |
| 223 | sum, _, bad := total(pairs, over, env) |
| 224 | if bad.Kind == KindErr { |
| 225 | return bad |
| 226 | } |
| 227 | return Value{Kind: KindNum, Num: sum} |
| 228 | } |
| 229 | |
| 230 | func fnAverageIfs(argNodes []Node, env Env) Value { |
| 231 | over, pairs, bad, ok := plural(argNodes, env) |
| 232 | if !ok { |
| 233 | return bad |
| 234 | } |
| 235 | sum, n, bad := total(pairs, over, env) |
| 236 | if bad.Kind == KindErr { |
| 237 | return bad |
| 238 | } |
| 239 | if n == 0 { |
| 240 | return errValue(ErrDiv0) |
| 241 | } |
| 242 | return Value{Kind: KindNum, Num: sum / float64(n)} |
| 243 | } |
| 244 | |
| 245 | // fnCountIfs has no applied range at all — it counts matches — so it |
| 246 | // decodes pairs from argument zero like COUNTIF rather than through |
| 247 | // plural. |
| 248 | func fnCountIfs(argNodes []Node, env Env) Value { |
| 249 | pairs, bad, ok := readPairs(argNodes, 0, env) |
| 250 | if !ok { |
| 251 | return bad |
| 252 | } |
| 253 | n := 0 |
| 254 | selected(pairs, env, func(int, int) { n++ }) |
| 255 | return Value{Kind: KindNum, Num: float64(n)} |
| 256 | } |
| 257 | |
| 258 | // plural decodes the (apply_range, range1, criteria1, …) shape. |
| 259 | func plural(argNodes []Node, env Env) (rect, []condPair, Value, bool) { |
| 260 | if len(argNodes) < 3 { |
| 261 | return rect{}, nil, errValue(ErrValue), false |
| 262 | } |
| 263 | over, ok := rectOf(argNodes[0]) |
| 264 | if !ok { |
| 265 | return rect{}, nil, errValue(ErrValue), false |
| 266 | } |
| 267 | pairs, bad, ok := readPairs(argNodes, 1, env) |
| 268 | if !ok { |
| 269 | return rect{}, nil, bad, false |
| 270 | } |
| 271 | // The SUM range's shape is checked too, and it was not. |
| 272 | // SUMIFS(A1:A2, B1:B3, "t*") answered 60 by reshaping A1:A2 up to |
| 273 | // the criteria range's height and reading A3 — a cell the formula |
| 274 | // never named. LibreOffice answers Err:502 and Excel #VALUE!. |
| 275 | // Caught by review, 2026-09-04. |
| 276 | // |
| 277 | // The singular forms are deliberately NOT checked this way: Excel's |
| 278 | // SUMIF really does read its third argument for its origin only, |
| 279 | // which is the documented asymmetry fn_cond.go's header describes. |
| 280 | if over.r1-over.r0 != pairs[0].over.r1-pairs[0].over.r0 || |
| 281 | over.c1-over.c0 != pairs[0].over.c1-pairs[0].over.c0 { |
| 282 | return rect{}, nil, errValue(ErrValue), false |
| 283 | } |
| 284 | return over, pairs, Value{}, true |
| 285 | } |
| 286 | |
amadan