chaffjs 0.19.0 → 0.21.0

package/dist/structure-shape/restatement.js

@@ -0,0 +1,32 @@
+/** Japanese is compared in runs of this many characters; four is past most particle-and-ending runs (ことが, します). */
+const CHAR_GRAM = 4;
+/** Punctuation, symbols and spaces end a run: a gram across them is shape, not content. */
+const BREAK = /[\s\p{P}\p{S}]/u;
+const WORD = /[\p{L}\p{N}]+/gu;
+/** Runs of code points, so a character outside the Basic Multilingual Plane counts as one. */
+const charGrams = (text) => text.split(BREAK).flatMap((run) => {
+    const characters = [...run];
+    return Array.from({ length: Math.max(0, characters.length - CHAR_GRAM + 1) }, (_, start) => characters.slice(start, start + CHAR_GRAM).join(""));
+});
+const wordGrams = (text) => {
+    const words = (text.toLowerCase().match(WORD) ?? []).filter((word) => word.length > 0);
+    return words.slice(1).map((word, index) => `${words[index] ?? ""} ${word}`);
+};
+/** The distinct grams of a text: four-character runs for a document measured in characters, word pairs otherwise. */
+export const gramsOf = (text, unit) => new Set(unit === "char" ? charGrams(text) : wordGrams(text));
+/** Fewer grams than this in the closing section and the share says more about its length than about what it says. */
+export const MIN_CLOSING_GRAMS = 10;
+const PERCENT = 100;
+/**
+ * The share of the closing section's grams already written in the body, as a rounded percentage: a まとめ that only
+ * restates scores high. Undefined when the closing is too short to measure.
+ */
+export const restatementPercent = (closing, body, unit) => {
+    const closingGrams = gramsOf(closing, unit);
+    if (closingGrams.size < MIN_CLOSING_GRAMS)
+        return undefined;
+    const bodyGrams = gramsOf(body, unit);
+    const repeated = [...closingGrams].filter((gram) => bodyGrams.has(gram)).length;
+    return Math.round((repeated / closingGrams.size) * PERCENT);
+};
+//# sourceMappingURL=restatement.js.map

package/dist/structure-shape/restatement.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"restatement.js","sourceRoot":"","sources":["../../src/structure-shape/restatement.ts"],"names":[],"mappings":"AAEA,mHAAmH;AACnH,MAAM,SAAS,GAAG,CAAC,CAAC;AAEpB,2FAA2F;AAC3F,MAAM,KAAK,GAAG,iBAAiB,CAAC;AAEhC,MAAM,IAAI,GAAG,iBAAiB,CAAC;AAE/B,8FAA8F;AAC9F,MAAM,SAAS,GAAG,CAAC,IAAY,EAAY,EAAE,CAC3C,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,OAAO,CAAC,CAAC,GAAG,EAAE,EAAE;IAChC,MAAM,UAAU,GAAG,CAAC,GAAG,GAAG,CAAC,CAAC;IAC5B,OAAO,KAAK,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,UAAU,CAAC,MAAM,GAAG,SAAS,GAAG,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,KAAK,EAAE,EAAE,CAAC,UAAU,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,GAAG,SAAS,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC;AACnJ,CAAC,CAAC,CAAC;AAEL,MAAM,SAAS,GAAG,CAAC,IAAY,EAAY,EAAE;IAC3C,MAAM,KAAK,GAAG,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;IACvF,OAAO,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,KAAK,EAAE,EAAE,CAAC,GAAG,KAAK,CAAC,KAAK,CAAC,IAAI,EAAE,IAAI,IAAI,EAAE,CAAC,CAAC;AAC9E,CAAC,CAAC;AAEF,qHAAqH;AACrH,MAAM,CAAC,MAAM,OAAO,GAAG,CAAC,IAAY,EAAE,IAAgB,EAAuB,EAAE,CAAC,IAAI,GAAG,CAAC,IAAI,KAAK,MAAM,CAAC,CAAC,CAAC,SAAS,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,IAAI,CAAC,CAAC,CAAC;AAE7I,qHAAqH;AACrH,MAAM,CAAC,MAAM,iBAAiB,GAAG,EAAE,CAAC;AAEpC,MAAM,OAAO,GAAG,GAAG,CAAC;AAEpB;;;GAGG;AACH,MAAM,CAAC,MAAM,kBAAkB,GAAG,CAAC,OAAe,EAAE,IAAY,EAAE,IAAgB,EAAsB,EAAE;IACxG,MAAM,YAAY,GAAG,OAAO,CAAC,OAAO,EAAE,IAAI,CAAC,CAAC;IAC5C,IAAI,YAAY,CAAC,IAAI,GAAG,iBAAiB;QAAE,OAAO,SAAS,CAAC;IAC5D,MAAM,SAAS,GAAG,OAAO,CAAC,IAAI,EAAE,IAAI,CAAC,CAAC;IACtC,MAAM,QAAQ,GAAG,CAAC,GAAG,YAAY,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,MAAM,CAAC;IAChF,OAAO,IAAI,CAAC,KAAK,CAAC,CAAC,QAAQ,GAAG,YAAY,CAAC,IAAI,CAAC,GAAG,OAAO,CAAC,CAAC;AAC9D,CAAC,CAAC"}

package/dist/structure-shape/score.d.ts

@@ -0,0 +1,30 @@
+import { type Direction, type FeatureValue } from "./features.ts";
+import type { FeatureBaseline, LanguageBaseline } from "./baseline.ts";
+/** A high measure counts past this percentile of human articles; a low one (uniform sections) below 100 minus it. */
+export declare const BEYOND_STEP = 90;
+/** One measure set against the human articles. */
+export type Placement = {
+    readonly feature: FeatureValue;
+    readonly direction: Direction;
+    /** Roughly what share of human articles the value is past, away from human writing (0 when it is not past any). */
+    readonly pastShare: number | undefined;
+    /** The value past which the measure counts: the human p90, or p10 for a low measure. */
+    readonly limit: number | undefined;
+    readonly median: number | undefined;
+    /** Whether the value is past the limit: one point of the structure score. */
+    readonly beyond: boolean;
+};
+/** The structure score: how many measures are past the human p90, out of how many could be compared. */
+export type StructureScore = {
+    readonly placements: readonly Placement[];
+    readonly score: number;
+    readonly compared: number;
+    /** How many human articles the baseline read, or undefined when there is no baseline for the language. */
+    readonly articles: number | undefined;
+};
+/** For a high measure: the highest step whose value is below this one. For a low one: the share whose value is above it. */
+export declare const pastShareOf: (value: number, baseline: FeatureBaseline, direction: Direction) => number;
+export declare const isBeyond: (value: number, limit: number, direction: Direction) => boolean;
+/** Each measure placed against the language's human baseline, and the count of those past the human p90. Never an opaque number. */
+export declare const structureScoreOf: (features: readonly FeatureValue[], baseline: LanguageBaseline | undefined) => StructureScore;
+//# sourceMappingURL=score.d.ts.map

package/dist/structure-shape/score.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"score.d.ts","sourceRoot":"","sources":["../../src/structure-shape/score.ts"],"names":[],"mappings":"AAAA,OAAO,EAAa,KAAK,SAAS,EAAE,KAAK,YAAY,EAAE,MAAM,eAAe,CAAC;AAC7E,OAAO,KAAK,EAAE,eAAe,EAAE,gBAAgB,EAAE,MAAM,eAAe,CAAC;AAEvE,qHAAqH;AACrH,eAAO,MAAM,WAAW,KAAK,CAAC;AAI9B,kDAAkD;AAClD,MAAM,MAAM,SAAS,GAAG;IACtB,QAAQ,CAAC,OAAO,EAAE,YAAY,CAAC;IAC/B,QAAQ,CAAC,SAAS,EAAE,SAAS,CAAC;IAC9B,mHAAmH;IACnH,QAAQ,CAAC,SAAS,EAAE,MAAM,GAAG,SAAS,CAAC;IACvC,wFAAwF;IACxF,QAAQ,CAAC,KAAK,EAAE,MAAM,GAAG,SAAS,CAAC;IACnC,QAAQ,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC;IACpC,6EAA6E;IAC7E,QAAQ,CAAC,MAAM,EAAE,OAAO,CAAC;CAC1B,CAAC;AAEF,wGAAwG;AACxG,MAAM,MAAM,cAAc,GAAG;IAC3B,QAAQ,CAAC,UAAU,EAAE,SAAS,SAAS,EAAE,CAAC;IAC1C,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,QAAQ,EAAE,MAAM,CAAC;IAC1B,0GAA0G;IAC1G,QAAQ,CAAC,QAAQ,EAAE,MAAM,GAAG,SAAS,CAAC;CACvC,CAAC;AAIF,4HAA4H;AAC5H,eAAO,MAAM,WAAW,GAAI,OAAO,MAAM,EAAE,UAAU,eAAe,EAAE,WAAW,SAAS,KAAG,MAK5F,CAAC;AAIF,eAAO,MAAM,QAAQ,GAAI,OAAO,MAAM,EAAE,OAAO,MAAM,EAAE,WAAW,SAAS,KAAG,OAAiE,CAAC;AAahJ,oIAAoI;AACpI,eAAO,MAAM,gBAAgB,GAAI,UAAU,SAAS,YAAY,EAAE,EAAE,UAAU,gBAAgB,GAAG,SAAS,KAAG,cAQ5G,CAAC"}

package/dist/structure-shape/score.js

@@ -0,0 +1,37 @@
+import { DIRECTION } from "./features.js";
+/** A high measure counts past this percentile of human articles; a low one (uniform sections) below 100 minus it. */
+export const BEYOND_STEP = 90;
+const MEDIAN_STEP = 50;
+const ALL = 100;
+const stepsOf = (baseline) => [...baseline.percentiles.entries()].toSorted((left, right) => left[0] - right[0]);
+/** For a high measure: the highest step whose value is below this one. For a low one: the share whose value is above it. */
+export const pastShareOf = (value, baseline, direction) => {
+    const steps = stepsOf(baseline);
+    if (direction === "high")
+        return steps.findLast(([, at]) => at < value)?.[0] ?? 0;
+    const above = steps.find(([, at]) => at > value)?.[0];
+    return above === undefined ? 0 : ALL - above;
+};
+const limitStep = (direction) => (direction === "high" ? BEYOND_STEP : ALL - BEYOND_STEP);
+export const isBeyond = (value, limit, direction) => (direction === "high" ? value > limit : value < limit);
+const placementOf = (feature, baseline) => {
+    const direction = DIRECTION[feature.id];
+    const limit = baseline?.percentiles.get(limitStep(direction));
+    const median = baseline?.percentiles.get(MEDIAN_STEP);
+    if (feature.value === undefined || baseline === undefined || limit === undefined) {
+        return { feature, direction, pastShare: undefined, limit, median, beyond: false };
+    }
+    const beyond = isBeyond(feature.value, limit, direction);
+    return { feature, direction, pastShare: pastShareOf(feature.value, baseline, direction), limit, median, beyond };
+};
+/** Each measure placed against the language's human baseline, and the count of those past the human p90. Never an opaque number. */
+export const structureScoreOf = (features, baseline) => {
+    const placements = features.map((feature) => placementOf(feature, baseline?.features[feature.id]));
+    return {
+        placements,
+        score: placements.filter((placement) => placement.beyond).length,
+        compared: placements.filter((placement) => placement.pastShare !== undefined).length,
+        articles: baseline?.articles,
+    };
+};
+//# sourceMappingURL=score.js.map

package/dist/structure-shape/score.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"score.js","sourceRoot":"","sources":["../../src/structure-shape/score.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAqC,MAAM,eAAe,CAAC;AAG7E,qHAAqH;AACrH,MAAM,CAAC,MAAM,WAAW,GAAG,EAAE,CAAC;AAC9B,MAAM,WAAW,GAAG,EAAE,CAAC;AACvB,MAAM,GAAG,GAAG,GAAG,CAAC;AAwBhB,MAAM,OAAO,GAAG,CAAC,QAAyB,EAAsB,EAAE,CAAC,CAAC,GAAG,QAAQ,CAAC,WAAW,CAAC,OAAO,EAAE,CAAC,CAAC,QAAQ,CAAC,CAAC,IAAI,EAAE,KAAK,EAAE,EAAE,CAAC,IAAI,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC;AAErJ,4HAA4H;AAC5H,MAAM,CAAC,MAAM,WAAW,GAAG,CAAC,KAAa,EAAE,QAAyB,EAAE,SAAoB,EAAU,EAAE;IACpG,MAAM,KAAK,GAAG,OAAO,CAAC,QAAQ,CAAC,CAAC;IAChC,IAAI,SAAS,KAAK,MAAM;QAAE,OAAO,KAAK,CAAC,QAAQ,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,EAAE,EAAE,CAAC,EAAE,GAAG,KAAK,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;IAClF,MAAM,KAAK,GAAG,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,EAAE,EAAE,CAAC,EAAE,GAAG,KAAK,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC;IACtD,OAAO,KAAK,KAAK,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,GAAG,KAAK,CAAC;AAC/C,CAAC,CAAC;AAEF,MAAM,SAAS,GAAG,CAAC,SAAoB,EAAU,EAAE,CAAC,CAAC,SAAS,KAAK,MAAM,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC,CAAC,GAAG,GAAG,WAAW,CAAC,CAAC;AAE7G,MAAM,CAAC,MAAM,QAAQ,GAAG,CAAC,KAAa,EAAE,KAAa,EAAE,SAAoB,EAAW,EAAE,CAAC,CAAC,SAAS,KAAK,MAAM,CAAC,CAAC,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,KAAK,GAAG,KAAK,CAAC,CAAC;AAEhJ,MAAM,WAAW,GAAG,CAAC,OAAqB,EAAE,QAAqC,EAAa,EAAE;IAC9F,MAAM,SAAS,GAAG,SAAS,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC;IACxC,MAAM,KAAK,GAAG,QAAQ,EAAE,WAAW,CAAC,GAAG,CAAC,SAAS,CAAC,SAAS,CAAC,CAAC,CAAC;IAC9D,MAAM,MAAM,GAAG,QAAQ,EAAE,WAAW,CAAC,GAAG,CAAC,WAAW,CAAC,CAAC;IACtD,IAAI,OAAO,CAAC,KAAK,KAAK,SAAS,IAAI,QAAQ,KAAK,SAAS,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;QACjF,OAAO,EAAE,OAAO,EAAE,SAAS,EAAE,SAAS,EAAE,SAAS,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC;IACpF,CAAC;IACD,MAAM,MAAM,GAAG,QAAQ,CAAC,OAAO,CAAC,KAAK,EAAE,KAAK,EAAE,SAAS,CAAC,CAAC;IACzD,OAAO,EAAE,OAAO,EAAE,SAAS,EAAE,SAAS,EAAE,WAAW,CAAC,OAAO,CAAC,KAAK,EAAE,QAAQ,EAAE,SAAS,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,CAAC;AACnH,CAAC,CAAC;AAEF,oIAAoI;AACpI,MAAM,CAAC,MAAM,gBAAgB,GAAG,CAAC,QAAiC,EAAE,QAAsC,EAAkB,EAAE;IAC5H,MAAM,UAAU,GAAG,QAAQ,CAAC,GAAG,CAAC,CAAC,OAAO,EAAE,EAAE,CAAC,WAAW,CAAC,OAAO,EAAE,QAAQ,EAAE,QAAQ,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC;IACnG,OAAO;QACL,UAAU;QACV,KAAK,EAAE,UAAU,CAAC,MAAM,CAAC,CAAC,SAAS,EAAE,EAAE,CAAC,SAAS,CAAC,MAAM,CAAC,CAAC,MAAM;QAChE,QAAQ,EAAE,UAAU,CAAC,MAAM,CAAC,CAAC,SAAS,EAAE,EAAE,CAAC,SAAS,CAAC,SAAS,KAAK,SAAS,CAAC,CAAC,MAAM;QACpF,QAAQ,EAAE,QAAQ,EAAE,QAAQ;KAC7B,CAAC;AACJ,CAAC,CAAC"}

package/dist/yaml-values.d.ts

@@ -0,0 +1,13 @@
+import type { Span } from "./plugin.ts";
+/** A YAML file is checked by its string values: keys, quotes, comments and punctuation are not prose. */
+export declare const isYamlPath: (path: string) => boolean;
+/**
+ * Every string value's text in a YAML file, keys left out, in document order. Each is one paragraph: two values are never one
+ * sentence. Undefined when the file is not valid YAML, which is then read as plain text.
+ */
+export declare const yamlValueSpans: (source: string) => Span[] | undefined;
+/** What lies outside the values, in order: keys, quotes, comments, punctuation. values: sorted and apart, as yamlValueSpans gives them. */
+export declare const outsideValues: (values: readonly Span[], length: number) => Span[];
+/** The file with only its values left where they stand: everything else blanked to spaces, line breaks kept. */
+export declare const valuesText: (source: string, values: readonly Span[]) => string;
+//# sourceMappingURL=yaml-values.d.ts.map

package/dist/yaml-values.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"yaml-values.d.ts","sourceRoot":"","sources":["../src/yaml-values.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,EAAE,IAAI,EAAE,MAAM,aAAa,CAAC;AAIxC,yGAAyG;AACzG,eAAO,MAAM,UAAU,GAAI,MAAM,MAAM,KAAG,OAAgD,CAAC;AAoB3F;;;GAGG;AACH,eAAO,MAAM,cAAc,GAAI,QAAQ,MAAM,KAAG,IAAI,EAAE,GAAG,SAcxD,CAAC;AAEF,2IAA2I;AAC3I,eAAO,MAAM,aAAa,GAAI,QAAQ,SAAS,IAAI,EAAE,EAAE,QAAQ,MAAM,KAAG,IAAI,EAQ3E,CAAC;AAEF,gHAAgH;AAChH,eAAO,MAAM,UAAU,GAAI,QAAQ,MAAM,EAAE,QAAQ,SAAS,IAAI,EAAE,KAAG,MAAiE,CAAC"}

package/dist/yaml-values.js

@@ -0,0 +1,58 @@
+import { extname } from "node:path";
+import { isScalar, parseAllDocuments, visit } from "yaml";
+import { maskSpans } from "./mask.js";
+const YAML = new Set([".yaml", ".yml"]);
+/** A YAML file is checked by its string values: keys, quotes, comments and punctuation are not prose. */
+export const isYamlPath = (path) => YAML.has(extname(path).toLowerCase());
+/** The characters of a string scalar that are its text: inside the quotes, below a block scalar's header line. */
+const textOf = (node, source) => {
+    const [start, end] = node.range ?? [];
+    if (start === undefined || end === undefined || typeof node.value !== "string")
+        return undefined;
+    if (node.type === "QUOTE_DOUBLE" || node.type === "QUOTE_SINGLE")
+        return { start: start + 1, end: end - 1 };
+    if (node.type === "BLOCK_LITERAL" || node.type === "BLOCK_FOLDED") {
+        const body = source.indexOf("\n", start);
+        return body === -1 || body >= end ? undefined : { start: body + 1, end };
+    }
+    return { start, end };
+};
+/** Trailing line breaks and spaces are not the value's text: a block scalar's range runs to the next key. */
+const trimmed = (span, source) => {
+    const text = source.slice(span.start, span.end);
+    return { start: span.start, end: span.start + text.trimEnd().length };
+};
+/**
+ * Every string value's text in a YAML file, keys left out, in document order. Each is one paragraph: two values are never one
+ * sentence. Undefined when the file is not valid YAML, which is then read as plain text.
+ */
+export const yamlValueSpans = (source) => {
+    const documents = parseAllDocuments(source);
+    if (!Array.isArray(documents) || documents.some((document) => document.errors.length > 0))
+        return undefined;
+    const spans = [];
+    documents.forEach((document) => visit(document, {
+        Scalar: (key, node) => {
+            const span = key === "key" || !isScalar(node) ? undefined : textOf(node, source);
+            const text = span === undefined ? undefined : trimmed(span, source);
+            if (text !== undefined && text.end > text.start)
+                spans.push(text);
+        },
+    }));
+    return spans.toSorted((left, right) => left.start - right.start);
+};
+/** What lies outside the values, in order: keys, quotes, comments, punctuation. values: sorted and apart, as yamlValueSpans gives them. */
+export const outsideValues = (values, length) => {
+    const gaps = [];
+    const cursor = values.reduce((at, span) => {
+        if (span.start > at)
+            gaps.push({ start: at, end: span.start });
+        return Math.max(at, span.end);
+    }, 0);
+    if (cursor < length)
+        gaps.push({ start: cursor, end: length });
+    return gaps;
+};
+/** The file with only its values left where they stand: everything else blanked to spaces, line breaks kept. */
+export const valuesText = (source, values) => maskSpans(source, outsideValues(values, source.length));
+//# sourceMappingURL=yaml-values.js.map

package/dist/yaml-values.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"yaml-values.js","sourceRoot":"","sources":["../src/yaml-values.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,OAAO,EAAE,MAAM,WAAW,CAAC;AACpC,OAAO,EAAE,QAAQ,EAAE,iBAAiB,EAAE,KAAK,EAAe,MAAM,MAAM,CAAC;AACvE,OAAO,EAAE,SAAS,EAAE,MAAM,WAAW,CAAC;AAGtC,MAAM,IAAI,GAAwB,IAAI,GAAG,CAAC,CAAC,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC;AAE7D,yGAAyG;AACzG,MAAM,CAAC,MAAM,UAAU,GAAG,CAAC,IAAY,EAAW,EAAE,CAAC,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC;AAE3F,kHAAkH;AAClH,MAAM,MAAM,GAAG,CAAC,IAAY,EAAE,MAAc,EAAoB,EAAE;IAChE,MAAM,CAAC,KAAK,EAAE,GAAG,CAAC,GAAG,IAAI,CAAC,KAAK,IAAI,EAAE,CAAC;IACtC,IAAI,KAAK,KAAK,SAAS,IAAI,GAAG,KAAK,SAAS,IAAI,OAAO,IAAI,CAAC,KAAK,KAAK,QAAQ;QAAE,OAAO,SAAS,CAAC;IACjG,IAAI,IAAI,CAAC,IAAI,KAAK,cAAc,IAAI,IAAI,CAAC,IAAI,KAAK,cAAc;QAAE,OAAO,EAAE,KAAK,EAAE,KAAK,GAAG,CAAC,EAAE,GAAG,EAAE,GAAG,GAAG,CAAC,EAAE,CAAC;IAC5G,IAAI,IAAI,CAAC,IAAI,KAAK,eAAe,IAAI,IAAI,CAAC,IAAI,KAAK,cAAc,EAAE,CAAC;QAClE,MAAM,IAAI,GAAG,MAAM,CAAC,OAAO,CAAC,IAAI,EAAE,KAAK,CAAC,CAAC;QACzC,OAAO,IAAI,KAAK,CAAC,CAAC,IAAI,IAAI,IAAI,GAAG,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,EAAE,KAAK,EAAE,IAAI,GAAG,CAAC,EAAE,GAAG,EAAE,CAAC;IAC3E,CAAC;IACD,OAAO,EAAE,KAAK,EAAE,GAAG,EAAE,CAAC;AACxB,CAAC,CAAC;AAEF,6GAA6G;AAC7G,MAAM,OAAO,GAAG,CAAC,IAAU,EAAE,MAAc,EAAQ,EAAE;IACnD,MAAM,IAAI,GAAG,MAAM,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC;IAChD,OAAO,EAAE,KAAK,EAAE,IAAI,CAAC,KAAK,EAAE,GAAG,EAAE,IAAI,CAAC,KAAK,GAAG,IAAI,CAAC,OAAO,EAAE,CAAC,MAAM,EAAE,CAAC;AACxE,CAAC,CAAC;AAEF;;;GAGG;AACH,MAAM,CAAC,MAAM,cAAc,GAAG,CAAC,MAAc,EAAsB,EAAE;IACnE,MAAM,SAAS,GAAG,iBAAiB,CAAC,MAAM,CAAC,CAAC;IAC5C,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,SAAS,CAAC,IAAI,SAAS,CAAC,IAAI,CAAC,CAAC,QAAQ,EAAE,EAAE,CAAC,QAAQ,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC;QAAE,OAAO,SAAS,CAAC;IAC5G,MAAM,KAAK,GAAW,EAAE,CAAC;IACzB,SAAS,CAAC,OAAO,CAAC,CAAC,QAAQ,EAAE,EAAE,CAC7B,KAAK,CAAC,QAAQ,EAAE;QACd,MAAM,EAAE,CAAC,GAAG,EAAE,IAAI,EAAE,EAAE;YACpB,MAAM,IAAI,GAAG,GAAG,KAAK,KAAK,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,EAAE,MAAM,CAAC,CAAC;YACjF,MAAM,IAAI,GAAG,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,OAAO,CAAC,IAAI,EAAE,MAAM,CAAC,CAAC;YACpE,IAAI,IAAI,KAAK,SAAS,IAAI,IAAI,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK;gBAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACpE,CAAC;KACF,CAAC,CACH,CAAC;IACF,OAAO,KAAK,CAAC,QAAQ,CAAC,CAAC,IAAI,EAAE,KAAK,EAAE,EAAE,CAAC,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC;AACnE,CAAC,CAAC;AAEF,2IAA2I;AAC3I,MAAM,CAAC,MAAM,aAAa,GAAG,CAAC,MAAuB,EAAE,MAAc,EAAU,EAAE;IAC/E,MAAM,IAAI,GAAW,EAAE,CAAC;IACxB,MAAM,MAAM,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,EAAE,EAAE,IAAI,EAAE,EAAE;QACxC,IAAI,IAAI,CAAC,KAAK,GAAG,EAAE;YAAE,IAAI,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,IAAI,CAAC,KAAK,EAAE,CAAC,CAAC;QAC/D,OAAO,IAAI,CAAC,GAAG,CAAC,EAAE,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC;IAChC,CAAC,EAAE,CAAC,CAAC,CAAC;IACN,IAAI,MAAM,GAAG,MAAM;QAAE,IAAI,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,EAAE,CAAC,CAAC;IAC/D,OAAO,IAAI,CAAC;AACd,CAAC,CAAC;AAEF,gHAAgH;AAChH,MAAM,CAAC,MAAM,UAAU,GAAG,CAAC,MAAc,EAAE,MAAuB,EAAU,EAAE,CAAC,SAAS,CAAC,MAAM,EAAE,aAAa,CAAC,MAAM,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC"}

package/genres.yaml

@@ -1,15 +1,19 @@
 # The genres chaff knows, in the order `chaff genres` lists them.
 #
 # A genre is a preset: pick the kind of document and chaff checks it the way that kind is written.
-#   rules:   the level a rule runs at for this genre (strict / normal / relaxed / off), under chaff.yaml's rules.
+#   rules:   the level a rule runs at for this genre (strict / normal / relaxed), under chaff.yaml's rules.
 #            A group's rules apply to every genre in it; a genre's own rules win over its group's.
-#            A rule turned off here is listed under "did not run" with the genre as the reason.
+#            A rule a group or genre turns off says so in its own file (off_for, with why), not here; chaff reads it
+#            as an off in that group's or genre's rules. It is listed under "did not run" with the genre as the reason.
+#            An off_for entry "<group>: measured by yarn rules:measure" is written by `yarn rules:measure --apply`: the
+#            rule reports on most human documents of the group (chaff-spec.md §21.1). Edit the measurement, not these.
 #   profile: the document profile (profiles/*.yaml) this genre reads with, when chaff.yaml names none.
 #   suggest: when no genre is set, what makes a file look like this genre, strongest first:
 #            paths (the file's path), then per language an opening (the body's first line, past front matter and
 #            headings) or a line with min_lines (how many lines must have that shape). A wrong suggestion misleads
 #            more than none, so a cue is added only where the corpus shows it does not fire on other kinds.
-# Which genres a rule applies to at all is the rule's use_for; the numbers behind a level are the rule's by_genre.
+# Which genres a rule applies to at all is the rule's use_for (a rule it leaves out is listed under "did not run" the same
+# way); the numbers behind a level are the rule's by_genre.
 
 groups:
   - id: technical
@@ -22,179 +26,19 @@ groups:
   # (shall be deemed), and joins long noun compounds (個人情報取扱事業者). Those rules would only flag the form.
   - id: legal
     name: { ja: 法務, en: Legal }
-    rules:
-      agentless-passive: off
-      max-kanji-continuous: off
-      no-doubled-joshi: off
-      ngram-repetition: off
-      repeated-sentence-head: off
-      heading-echo: off
-      taigen-dome-in-prose: off
-      unqualified-superlative: off
-      excessive-hedging: off
-      proper-noun-density: off
-      expletive-construction: off
-      adverb-overuse: off
-      concrete-evidence-density: off
-      closing-cliche: off
-      padded-intro: off
-      preamble-length: off
-      cushion-phrase-density: off
-      sasete-itadaku: off
-      sentence-rhythm: off
-      rule-of-three: off
-      section-length-uniformity: off
-      paragraph-length-variance: off
-      ai-tell: off
-      ai-generated-composite: off
-      contrast-framing: off
-      stock-transition: off
-      announcing-opener: off
-      colon-lead-in: off
-      bold-label-list: off
-      # Statutes and opinions write "not inconsistent with" on purpose.
-      double-negative: off
-      # Legal drafting writes 「…を行う」 and "make a payment" as its register; the corpus statutes are full of them.
-      nominalization: off
   # A help page repeats its heading in its first line (To upload a file, ...) and states no claims to back up.
   - id: docs
     name: { ja: 説明書, en: Documentation }
-    rules:
-      heading-echo: off
-      agentless-passive: off
-      concrete-evidence-density: off
-      preamble-length: off
-      proper-noun-density: off
-      unqualified-superlative: off
-      expletive-construction: off
-      adverb-overuse: off
-      cushion-phrase-density: off
-      double-keigo: off
-      sasete-itadaku: off
-      sentence-initial-conjunction-run: off
-      closing-cliche: off
-      padded-intro: off
-      sentence-rhythm: off
-      rule-of-three: off
-      section-length-uniformity: off
-      paragraph-length-variance: off
-      ai-tell: off
-      ai-generated-composite: off
-      contrast-framing: off
-      stock-transition: off
-      announcing-opener: off
-      colon-lead-in: off
-      bold-label-list: off
   # Science writes in the passive, hedges its claims on purpose (may, suggest) and cites numbers as its evidence.
   - id: academic
     name: { ja: 学術, en: Academic }
-    rules:
-      agentless-passive: off
-      excessive-hedging: off
-      expletive-construction: off
-      concrete-evidence-density: off
-      proper-noun-density: off
-      adverb-overuse: off
-      closing-cliche: off
-      padded-intro: off
-      preamble-length: off
-      cushion-phrase-density: off
-      sasete-itadaku: off
-      double-keigo: off
-      sentence-rhythm: off
-      rule-of-three: off
-      section-length-uniformity: off
-      paragraph-length-variance: off
-      stock-transition: off
-      bold-label-list: off
-      requirement-smell: off
   # Length, rhythm, repetition and register are the author's choices in literature, and dialogue mixes styles.
   # Only what one work can hold itself to runs: consistent spacing and commas, and numbering without gaps.
   - id: literature
     name: { ja: 文学, en: Literature }
-    rules:
-      max-sentence-length: off
-      max-paragraph-length: off
-      ngram-repetition: off
-      repeated-sentence-head: off
-      heading-echo: off
-      empty-intensifier: off
-      no-doubled-joshi: off
-      no-nakaguro-parallel: off
-      taigen-dome-in-prose: off
-      max-kanji-continuous: off
-      hiragana-fukushi: off
-      no-mixed-desumasu: off
-      contraction-consistency: off
-      doubled-word: off
-      agreement-slip: off
-      agentless-passive: off
-      expletive-construction: off
-      excessive-hedging: off
-      unqualified-superlative: off
-      adverb-overuse: off
-      sentence-initial-conjunction-run: off
-      repeated-conjunction: off
-      undefined-acronym: off
-      proper-noun-density: off
-      concrete-evidence-density: off
-      closing-cliche: off
-      padded-intro: off
-      preamble-length: off
-      cushion-phrase-density: off
-      sasete-itadaku: off
-      double-keigo: off
-      no-em-dash: off
-      emoji-density: off
-      sentence-rhythm: off
-      rule-of-three: off
-      section-length-uniformity: off
-      paragraph-length-variance: off
-      ai-tell: off
-      ai-generated-composite: off
-      contrast-framing: off
-      stock-transition: off
-      announcing-opener: off
-      colon-lead-in: off
-      bold-label-list: off
-      nominalization: off
-      dangling-reference: off
-      duplicate-definition: off
-      date-order: off
-      date-weekday-mismatch: off
-      total-mismatch: off
-      announced-count-mismatch: off
-      dangling-figure-reference: off
-      date-range-reversed: off
-      percent-sum-mismatch: off
-      vague-figure-reference: off
-      # A narrator may hedge, speak with ら dropped, and spell as the work's time and place did.
-      double-negative: off
-      ra-nuki: off
-      spelling-consistency: off
-      requirement-smell: off
   # A speech opens with greetings and closes with thanks: that is its form, not padding.
   - id: speech
     name: { ja: 話し言葉, en: Speech }
-    rules:
-      agentless-passive: off
-      preamble-length: off
-      padded-intro: off
-      closing-cliche: off
-      concrete-evidence-density: off
-      proper-noun-density: off
-      sentence-rhythm: off
-      rule-of-three: off
-      section-length-uniformity: off
-      paragraph-length-variance: off
-      contrast-framing: off
-      stock-transition: off
-      announcing-opener: off
-      colon-lead-in: off
-      # A transcript keeps the speaker's words: hedges and ら dropped as spoken.
-      double-negative: off
-      ra-nuki: off
-      requirement-smell: off
 
 genres:
   - id: technical/spec
@@ -247,21 +91,11 @@ genres:
     summary:
       ja: プレスリリースと、お知らせ
       en: Press releases and public notices
-    # A release names the agency, the people quoted, the mission and the place: that is its news, not a crowd of
-    # names to thin out. Every English release in the corpus (GSA, NASA, NOAA) is over the limit.
-    rules:
-      proper-noun-density: off
   - id: business/meeting-notes
     name: { ja: 議事録, en: Meeting notes }
     summary:
       ja: 議事録・議事要旨
       en: Minutes and meeting notes
-    # Minutes record what the meeting did (the meeting was held, 了承された): the actor is the meeting, by the form.
-    # In the corpus the rule fires on nearly every set of minutes, English and Japanese.
-    # A header block of labelled fields (- **日時**：…, - **出席者**：…) is the form of minutes, not a list of captions.
-    rules:
-      agentless-passive: off
-      bold-label-list: off
   - id: legal/contract
     name: { ja: 契約書・規約, en: Contract and terms }
     summary:
@@ -284,6 +118,9 @@ genres:
       total-mismatch: normal
       dangling-figure-reference: normal
       date-range-reversed: normal
+      # A contract names each party and term once and uses that name: a term quoted again or a party written out in full is a slip.
+      defined-term-form: normal
+      defined-name-repeated: normal
   - id: legal/statute
     name: { ja: 法令・規程, en: Statute and regulation }
     summary:
@@ -305,8 +142,6 @@ genres:
       total-mismatch: normal
       dangling-figure-reference: normal
       date-range-reversed: normal
-      # A statute defines its 等 terms (以下「個人情報等」という) and uses them in its obligations.
-      requirement-smell: off
   - id: legal/judgment
     name: { ja: 判決, en: Court decision }
     summary:
@@ -315,9 +150,6 @@ genres:
     suggest:
       ja: { line: "^(?:#{1,6}\\s*)?(?:主\\s*文|理\\s*由|判\\s*決)\\s*$", min_lines: 2 }
       en: { line: "delivered the opinion of the Court|^(?:Reversed|Affirmed)\\.?$", min_lines: 1 }
-    # requirement-smell is for the documents that state requirements: specifications, contracts and documentation.
-    rules:
-      requirement-smell: off
   - id: legal/patent
     name: { ja: 特許明細書, en: Patent }
     summary:
@@ -326,9 +158,6 @@ genres:
     suggest:
       ja: { line: "【(?:課題|解決手段|請求項[0-9０-９]+|選択図|技術分野|背景技術|発明の詳細な説明|特許請求の範囲)】", min_lines: 2 }
       en: { line: "^#{1,6}\\s*(?:BACKGROUND OF THE INVENTION|SUMMARY OF THE INVENTION|Claims\\b)|\\bof claim \\d+", min_lines: 3 }
-    # requirement-smell is for the documents that state requirements: specifications, contracts and documentation.
-    rules:
-      requirement-smell: off
   - id: docs/manual
     name: { ja: マニュアル・手順書, en: Manual and how-to }
     summary:
@@ -337,6 +166,12 @@ genres:
     # Path only: step and procedure headings are as common in tech blogs and specifications as in manuals.
     suggest:
       paths: ["(^|/)(manuals?|guides?|how-?tos?|help|tutorials?|user-?guides?)(/|[-_.])"]
+    # A manual is followed command by command: its code blocks, prompts, command headings and steps are what the reader acts on.
+    rules:
+      code-fence-language-mix: normal
+      shell-prompt-mix: normal
+      heading-command-missing: normal
+      step-statement-mix: normal
   - id: docs/faq
     name: { ja: よくある質問, en: FAQ }
     summary:
@@ -355,13 +190,6 @@ genres:
       paths: ["(^|/)(glossary|glossaries|yogo|用語集)(/|[-_.])"]
       ja: { line: "^#{1,6}\\s*用語集", min_lines: 1 }
       en: { line: "^#{1,6}\\s*Glossary\\b", min_lines: 1 }
-    # A reader looks up one entry and reads only that. A definition is a noun phrase (風の吹いてくる方向。), and sibling
-    # entries share one wording on purpose (Portion of labor productivity change attributed to …), which no reader sees twice.
-    rules:
-      taigen-dome-in-prose: off
-      ngram-repetition: off
-      # An entry that ends 〜すること or 〜のこと defines a word; it is no requirement.
-      requirement-smell: off
   - id: academic/paper
     name: { ja: 論文, en: Academic paper }
     summary:
@@ -375,6 +203,13 @@ genres:
           line: "^(?:#{1,6}\\s*)?(?:\\d+\\.?\\s+)?(?:Abstract|Keywords|Methods?|Methodology|Results|Discussion|Conclusions?|Related [Ww]ork|Bibliography|References)\\s*$",
           min_lines: 4,
         }
+    # A paper is read against its own apparatus: one citation style, citations that match the reference list, figures
+    # numbered in the order they are referred to, and an abstract of a fixed length.
+    rules:
+      citation-style-mix: normal
+      citation-reference-mismatch: normal
+      figure-reference-order: normal
+      abstract-length: normal
   - id: literature/fiction
     name: { ja: 小説, en: Fiction }
     summary:
@@ -429,35 +264,3 @@ genres:
       paths: ["(^|/)(transcripts?|kaigiroku)(/|[-_.])"]
       ja: { line: "^○[^　\\s]{1,30}　|^（(?:問|答)）|^【(?:記者|大臣|質問|回答)】", min_lines: 3 }
       en: { line: "^(?:Mr|Ms|Mrs|Dr)\\. [A-Z][A-Z-]{2,}(?: of [A-Z][a-z]+)?\\. ", min_lines: 3 }
-    # A transcript records what was said: its sentences, repeats and register are the speaker's, not the writer's to fix.
-    rules:
-      max-sentence-length: off
-      max-paragraph-length: off
-      ngram-repetition: off
-      repeated-sentence-head: off
-      heading-echo: off
-      empty-intensifier: off
-      no-doubled-joshi: off
-      no-nakaguro-parallel: off
-      taigen-dome-in-prose: off
-      max-kanji-continuous: off
-      hiragana-fukushi: off
-      no-mixed-desumasu: off
-      contraction-consistency: off
-      doubled-word: off
-      agreement-slip: off
-      sasete-itadaku: off
-      double-keigo: off
-      # Labelled speaker turns (- **記者**：…) are how a transcript is laid out.
-      bold-label-list: off
-      cushion-phrase-density: off
-      excessive-hedging: off
-      unqualified-superlative: off
-      expletive-construction: off
-      adverb-overuse: off
-      sentence-initial-conjunction-run: off
-      repeated-conjunction: off
-      undefined-acronym: off
-      ai-tell: off
-      ai-generated-composite: off
-      assistant-residue: off

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "chaffjs",
-  "version": "0.19.0",
+  "version": "0.21.0",
   "description": "Prose validation harness. Finds what is hard to read, and never rewrites it for you.",
   "keywords": [
     "chaff",
@@ -37,6 +37,10 @@
     "./api": {
       "types": "./src/api.ts",
       "import": "./dist/api.js"
+    },
+    "./grade": {
+      "types": "./src/grade-api.ts",
+      "import": "./dist/grade-api.js"
     }
   },
   "files": [
@@ -47,6 +51,7 @@
     "genres.yaml",
     "profiles",
     "reference-headings.yaml",
+    "structure-baseline.yaml",
     "rules",
     "skills",
     "styles",
@@ -61,8 +66,8 @@
   },
   "dependencies": {
     "@anthropic-ai/sdk": "^0.125.0",
-    "@chaffjs/lang-en": "0.17.0",
-    "@chaffjs/lang-ja": "0.18.0",
+    "@chaffjs/lang-en": "0.19.0",
+    "@chaffjs/lang-ja": "0.20.0",
     "mdast-util-from-markdown": "^2.0.3",
     "mdast-util-frontmatter": "^2.0.1",
     "mdast-util-gfm-table": "^2.0.0",