@lzehrung/codegraph-core 2.1.0

package/dist/chunking/chunkSplit.js

@@ -0,0 +1,170 @@
+export function splitLargeBlockSimple(block, source, tokenizer, maxTokens, out, languageId, filePath) {
+    let currentStart = -1;
+    let currentEnd = -1;
+    let currentStartLine = block.startLine;
+    let currentTokens = 0;
+    const flush = () => {
+        if (currentStart === -1)
+            return;
+        emitRangeWithinBudget(block, source, currentStart, currentEnd, currentStartLine, tokenizer, maxTokens, out, languageId, filePath);
+        currentStart = -1;
+        currentEnd = -1;
+        currentTokens = 0;
+    };
+    for (const line of lineRanges(source, block.startByte, block.endByte, block.startLine)) {
+        const lineText = source.slice(line.start, line.end);
+        const lineTokens = tokenizer(lineText);
+        if (lineTokens > maxTokens) {
+            flush();
+            emitRangeWithinBudget(block, source, line.start, line.end, line.startLine, tokenizer, maxTokens, out, languageId, filePath);
+            continue;
+        }
+        if (currentStart !== -1 && currentTokens + lineTokens > maxTokens) {
+            flush();
+        }
+        if (currentStart === -1) {
+            currentStart = line.start;
+            currentStartLine = line.startLine;
+        }
+        currentEnd = line.end;
+        currentTokens += lineTokens;
+    }
+    flush();
+}
+export function splitLargeBlockUsingInnerBlocks(block, _innerBlocks, source, tokenizer, maxTokens, out, languageId, _newlineOffsets, filePath) {
+    // Semantic boundaries are useful for selecting a parent chunk, but splitting
+    // that parent at arbitrary AST offsets can make sibling line ranges overlap.
+    // Pack complete source lines here; split within a line only when that line
+    // alone exceeds the token budget.
+    splitLargeBlockSimple(block, source, tokenizer, maxTokens, out, languageId, filePath);
+}
+/** Splits an oversized string on code-point boundaries using the configured tokenizer. */
+export function splitTextWithinTokenBudget(text, tokenizer, maxTokens) {
+    if (!text.length)
+        return [];
+    if (tokenizer(text) <= maxTokens)
+        return [text];
+    const segments = [];
+    let start = 0;
+    while (start < text.length) {
+        if (tokenizer(text.slice(start)) <= maxTokens) {
+            segments.push(text.slice(start));
+            break;
+        }
+        let low = nextCodePointBoundary(text, start);
+        let high = text.length;
+        let end = start;
+        while (low <= high) {
+            const candidate = previousCodePointBoundary(text, low + ((high - low) >>> 1));
+            if (candidate <= start) {
+                low = nextCodePointBoundary(text, start);
+                continue;
+            }
+            if (tokenizer(text.slice(start, candidate)) <= maxTokens) {
+                end = candidate;
+                low = candidate + 1;
+            }
+            else {
+                high = candidate - 1;
+            }
+        }
+        if (end === start) {
+            const next = nextCodePointBoundary(text, start);
+            const singleCodePoint = text.slice(start, next);
+            if (tokenizer(singleCodePoint) > maxTokens) {
+                throw new RangeError("maxTokens is smaller than the configured tokenizer's smallest unit");
+            }
+            end = next;
+        }
+        segments.push(text.slice(start, end));
+        start = end;
+    }
+    return segments;
+}
+function emitRangeWithinBudget(block, source, start, end, startLine, tokenizer, maxTokens, out, languageId, filePath) {
+    const text = source.slice(start, end);
+    const tokenCount = tokenizer(text);
+    if (tokenCount <= maxTokens) {
+        pushRangedChunk(out, block, languageId, filePath, start, end, startLine, endLineForText(startLine, text), text, tokenCount);
+        return;
+    }
+    let segmentStart = start;
+    let segmentStartLine = startLine;
+    for (const segment of splitTextWithinTokenBudget(text, tokenizer, maxTokens)) {
+        const segmentEnd = segmentStart + segment.length;
+        pushRangedChunk(out, block, languageId, filePath, segmentStart, segmentEnd, segmentStartLine, endLineForText(segmentStartLine, segment), segment, tokenizer(segment));
+        segmentStartLine += countLineBreaks(segment);
+        segmentStart = segmentEnd;
+    }
+}
+function pushRangedChunk(out, block, languageId, filePath, sourceStart, sourceEnd, startLine, endLine, text, tokenCount) {
+    out.push({
+        id: "",
+        languageId,
+        type: block.kind,
+        startLine,
+        endLine,
+        text,
+        tokenCount,
+        sourceStart,
+        sourceEnd,
+        ...(filePath !== undefined ? { filePath } : {}),
+        ...(block.name !== undefined ? { name: block.name } : {}),
+    });
+}
+function lineRanges(source, start, end, startLine) {
+    const ranges = [];
+    let rangeStart = start;
+    let line = startLine;
+    for (let offset = start; offset < end; offset++) {
+        if (source[offset] !== "\n")
+            continue;
+        ranges.push({ start: rangeStart, end: offset + 1, startLine: line });
+        rangeStart = offset + 1;
+        line++;
+    }
+    if (rangeStart < end) {
+        ranges.push({ start: rangeStart, end, startLine: line });
+    }
+    return ranges;
+}
+function countLineBreaks(text) {
+    let count = 0;
+    for (let index = 0; index < text.length; index++) {
+        if (text[index] === "\n")
+            count++;
+    }
+    return count;
+}
+function endLineForText(startLine, text) {
+    const lineBreaks = countLineBreaks(text);
+    return startLine + lineBreaks - (text.endsWith("\n") ? 1 : 0);
+}
+function nextCodePointBoundary(text, index) {
+    if (index >= text.length)
+        return text.length;
+    const codePoint = text.codePointAt(index);
+    return index + (codePoint !== undefined && codePoint > 0xffff ? 2 : 1);
+}
+function previousCodePointBoundary(text, index) {
+    if (index <= 0 || index >= text.length || text.charCodeAt(index) < 0xdc00 || text.charCodeAt(index) > 0xdfff) {
+        return index;
+    }
+    return index - 1;
+}
+function locateLineAndColFromByte(newlineOffsets, byteOffset) {
+    let low = 0;
+    let high = newlineOffsets.length;
+    while (low < high) {
+        const mid = (low + high) >>> 1;
+        if (newlineOffsets[mid] < byteOffset)
+            low = mid + 1;
+        else
+            high = mid;
+    }
+    const line = low;
+    const prevNewline = low > 0 ? newlineOffsets[low - 1] : -1;
+    const col = byteOffset - prevNewline - 1;
+    return [line, col];
+}
+//# sourceMappingURL=chunkSplit.js.map

package/dist/chunking/chunkSplit.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"chunkSplit.js","sourceRoot":"","sources":["../../src/chunking/chunkSplit.ts"],"names":[],"mappings":"AAEA,MAAM,UAAU,qBAAqB,CACnC,KAAqB,EACrB,MAAc,EACd,SAAyB,EACzB,SAAiB,EACjB,GAAkB,EAClB,UAAkB,EAClB,QAAiB;IAEjB,IAAI,YAAY,GAAG,CAAC,CAAC,CAAC;IACtB,IAAI,UAAU,GAAG,CAAC,CAAC,CAAC;IACpB,IAAI,gBAAgB,GAAG,KAAK,CAAC,SAAS,CAAC;IACvC,IAAI,aAAa,GAAG,CAAC,CAAC;IAEtB,MAAM,KAAK,GAAG,GAAG,EAAE;QACjB,IAAI,YAAY,KAAK,CAAC,CAAC;YAAE,OAAO;QAChC,qBAAqB,CACnB,KAAK,EACL,MAAM,EACN,YAAY,EACZ,UAAU,EACV,gBAAgB,EAChB,SAAS,EACT,SAAS,EACT,GAAG,EACH,UAAU,EACV,QAAQ,CACT,CAAC;QACF,YAAY,GAAG,CAAC,CAAC,CAAC;QAClB,UAAU,GAAG,CAAC,CAAC,CAAC;QAChB,aAAa,GAAG,CAAC,CAAC;IACpB,CAAC,CAAC;IAEF,KAAK,MAAM,IAAI,IAAI,UAAU,CAAC,MAAM,EAAE,KAAK,CAAC,SAAS,EAAE,KAAK,CAAC,OAAO,EAAE,KAAK,CAAC,SAAS,CAAC,EAAE,CAAC;QACvF,MAAM,QAAQ,GAAG,MAAM,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC;QACpD,MAAM,UAAU,GAAG,SAAS,CAAC,QAAQ,CAAC,CAAC;QAEvC,IAAI,UAAU,GAAG,SAAS,EAAE,CAAC;YAC3B,KAAK,EAAE,CAAC;YACR,qBAAqB,CACnB,KAAK,EACL,MAAM,EACN,IAAI,CAAC,KAAK,EACV,IAAI,CAAC,GAAG,EACR,IAAI,CAAC,SAAS,EACd,SAAS,EACT,SAAS,EACT,GAAG,EACH,UAAU,EACV,QAAQ,CACT,CAAC;YACF,SAAS;QACX,CAAC;QAED,IAAI,YAAY,KAAK,CAAC,CAAC,IAAI,aAAa,GAAG,UAAU,GAAG,SAAS,EAAE,CAAC;YAClE,KAAK,EAAE,CAAC;QACV,CAAC;QAED,IAAI,YAAY,KAAK,CAAC,CAAC,EAAE,CAAC;YACxB,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC;YAC1B,gBAAgB,GAAG,IAAI,CAAC,SAAS,CAAC;QACpC,CAAC;QACD,UAAU,GAAG,IAAI,CAAC,GAAG,CAAC;QACtB,aAAa,IAAI,UAAU,CAAC;IAC9B,CAAC;IAED,KAAK,EAAE,CAAC;AACV,CAAC;AAED,MAAM,UAAU,+BAA+B,CAC7C,KAAqB,EACrB,YAA8B,EAC9B,MAAc,EACd,SAAyB,EACzB,SAAiB,EACjB,GAAkB,EAClB,UAAkB,EAClB,eAAyB,EACzB,QAAiB;IAEjB,6EAA6E;IAC7E,6EAA6E;IAC7E,2EAA2E;IAC3E,kCAAkC;IAClC,qBAAqB,CAAC,KAAK,EAAE,MAAM,EAAE,SAAS,EAAE,SAAS,EAAE,GAAG,EAAE,UAAU,EAAE,QAAQ,CAAC,CAAC;AACxF,CAAC;AAED,0FAA0F;AAC1F,MAAM,UAAU,0BAA0B,CAAC,IAAY,EAAE,SAAyB,EAAE,SAAiB;IACnG,IAAI,CAAC,IAAI,CAAC,MAAM;QAAE,OAAO,EAAE,CAAC;IAC5B,IAAI,SAAS,CAAC,IAAI,CAAC,IAAI,SAAS;QAAE,OAAO,CAAC,IAAI,CAAC,CAAC;IAEhD,MAAM,QAAQ,GAAa,EAAE,CAAC;IAC9B,IAAI,KAAK,GAAG,CAAC,CAAC;IAEd,OAAO,KAAK,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC;QAC3B,IAAI,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,IAAI,SAAS,EAAE,CAAC;YAC9C,QAAQ,CAAC,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC;YACjC,MAAM;QACR,CAAC;QAED,IAAI,GAAG,GAAG,qBAAqB,CAAC,IAAI,EAAE,KAAK,CAAC,CAAC;QAC7C,IAAI,IAAI,GAAG,IAAI,CAAC,MAAM,CAAC;QACvB,IAAI,GAAG,GAAG,KAAK,CAAC;QAEhB,OAAO,GAAG,IAAI,IAAI,EAAE,CAAC;YACnB,MAAM,SAAS,GAAG,yBAAyB,CAAC,IAAI,EAAE,GAAG,GAAG,CAAC,CAAC,IAAI,GAAG,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC;YAC9E,IAAI,SAAS,IAAI,KAAK,EAAE,CAAC;gBACvB,GAAG,GAAG,qBAAqB,CAAC,IAAI,EAAE,KAAK,CAAC,CAAC;gBACzC,SAAS;YACX,CAAC;YAED,IAAI,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,EAAE,SAAS,CAAC,CAAC,IAAI,SAAS,EAAE,CAAC;gBACzD,GAAG,GAAG,SAAS,CAAC;gBAChB,GAAG,GAAG,SAAS,GAAG,CAAC,CAAC;YACtB,CAAC;iBAAM,CAAC;gBACN,IAAI,GAAG,SAAS,GAAG,CAAC,CAAC;YACvB,CAAC;QACH,CAAC;QAED,IAAI,GAAG,KAAK,KAAK,EAAE,CAAC;YAClB,MAAM,IAAI,GAAG,qBAAqB,CAAC,IAAI,EAAE,KAAK,CAAC,CAAC;YAChD,MAAM,eAAe,GAAG,IAAI,CAAC,KAAK,CAAC,KAAK,EAAE,IAAI,CAAC,CAAC;YAChD,IAAI,SAAS,CAAC,eAAe,CAAC,GAAG,SAAS,EAAE,CAAC;gBAC3C,MAAM,IAAI,UAAU,CAAC,oEAAoE,CAAC,CAAC;YAC7F,CAAC;YACD,GAAG,GAAG,IAAI,CAAC;QACb,CAAC;QAED,QAAQ,CAAC,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC;QACtC,KAAK,GAAG,GAAG,CAAC;IACd,CAAC;IAED,OAAO,QAAQ,CAAC;AAClB,CAAC;AAED,SAAS,qBAAqB,CAC5B,KAAqB,EACrB,MAAc,EACd,KAAa,EACb,GAAW,EACX,SAAiB,EACjB,SAAyB,EACzB,SAAiB,EACjB,GAAkB,EAClB,UAAkB,EAClB,QAA4B;IAE5B,MAAM,IAAI,GAAG,MAAM,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC;IACtC,MAAM,UAAU,GAAG,SAAS,CAAC,IAAI,CAAC,CAAC;IACnC,IAAI,UAAU,IAAI,SAAS,EAAE,CAAC;QAC5B,eAAe,CACb,GAAG,EACH,KAAK,EACL,UAAU,EACV,QAAQ,EACR,KAAK,EACL,GAAG,EACH,SAAS,EACT,cAAc,CAAC,SAAS,EAAE,IAAI,CAAC,EAC/B,IAAI,EACJ,UAAU,CACX,CAAC;QACF,OAAO;IACT,CAAC;IAED,IAAI,YAAY,GAAG,KAAK,CAAC;IACzB,IAAI,gBAAgB,GAAG,SAAS,CAAC;IACjC,KAAK,MAAM,OAAO,IAAI,0BAA0B,CAAC,IAAI,EAAE,SAAS,EAAE,SAAS,CAAC,EAAE,CAAC;QAC7E,MAAM,UAAU,GAAG,YAAY,GAAG,OAAO,CAAC,MAAM,CAAC;QACjD,eAAe,CACb,GAAG,EACH,KAAK,EACL,UAAU,EACV,QAAQ,EACR,YAAY,EACZ,UAAU,EACV,gBAAgB,EAChB,cAAc,CAAC,gBAAgB,EAAE,OAAO,CAAC,EACzC,OAAO,EACP,SAAS,CAAC,OAAO,CAAC,CACnB,CAAC;QACF,gBAAgB,IAAI,eAAe,CAAC,OAAO,CAAC,CAAC;QAC7C,YAAY,GAAG,UAAU,CAAC;IAC5B,CAAC;AACH,CAAC;AAED,SAAS,eAAe,CACtB,GAAkB,EAClB,KAAqB,EACrB,UAAkB,EAClB,QAA4B,EAC5B,WAAmB,EACnB,SAAiB,EACjB,SAAiB,EACjB,OAAe,EACf,IAAY,EACZ,UAAkB;IAElB,GAAG,CAAC,IAAI,CAAC;QACP,EAAE,EAAE,EAAE;QACN,UAAU;QACV,IAAI,EAAE,KAAK,CAAC,IAAI;QAChB,SAAS;QACT,OAAO;QACP,IAAI;QACJ,UAAU;QACV,WAAW;QACX,SAAS;QACT,GAAG,CAAC,QAAQ,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;QAC/C,GAAG,CAAC,KAAK,CAAC,IAAI,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;KAC1D,CAAC,CAAC;AACL,CAAC;AAED,SAAS,UAAU,CACjB,MAAc,EACd,KAAa,EACb,GAAW,EACX,SAAiB;IAEjB,MAAM,MAAM,GAA6D,EAAE,CAAC;IAC5E,IAAI,UAAU,GAAG,KAAK,CAAC;IACvB,IAAI,IAAI,GAAG,SAAS,CAAC;IAErB,KAAK,IAAI,MAAM,GAAG,KAAK,EAAE,MAAM,GAAG,GAAG,EAAE,MAAM,EAAE,EAAE,CAAC;QAChD,IAAI,MAAM,CAAC,MAAM,CAAC,KAAK,IAAI;YAAE,SAAS;QACtC,MAAM,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,UAAU,EAAE,GAAG,EAAE,MAAM,GAAG,CAAC,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;QACrE,UAAU,GAAG,MAAM,GAAG,CAAC,CAAC;QACxB,IAAI,EAAE,CAAC;IACT,CAAC;IAED,IAAI,UAAU,GAAG,GAAG,EAAE,CAAC;QACrB,MAAM,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,UAAU,EAAE,GAAG,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;IAC3D,CAAC;IAED,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,eAAe,CAAC,IAAY;IACnC,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,IAAI,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC;QACjD,IAAI,IAAI,CAAC,KAAK,CAAC,KAAK,IAAI;YAAE,KAAK,EAAE,CAAC;IACpC,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAED,SAAS,cAAc,CAAC,SAAiB,EAAE,IAAY;IACrD,MAAM,UAAU,GAAG,eAAe,CAAC,IAAI,CAAC,CAAC;IACzC,OAAO,SAAS,GAAG,UAAU,GAAG,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;AAChE,CAAC;AAED,SAAS,qBAAqB,CAAC,IAAY,EAAE,KAAa;IACxD,IAAI,KAAK,IAAI,IAAI,CAAC,MAAM;QAAE,OAAO,IAAI,CAAC,MAAM,CAAC;IAC7C,MAAM,SAAS,GAAG,IAAI,CAAC,WAAW,CAAC,KAAK,CAAC,CAAC;IAC1C,OAAO,KAAK,GAAG,CAAC,SAAS,KAAK,SAAS,IAAI,SAAS,GAAG,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;AACzE,CAAC;AAED,SAAS,yBAAyB,CAAC,IAAY,EAAE,KAAa;IAC5D,IAAI,KAAK,IAAI,CAAC,IAAI,KAAK,IAAI,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,UAAU,CAAC,KAAK,CAAC,GAAG,MAAM,IAAI,IAAI,CAAC,UAAU,CAAC,KAAK,CAAC,GAAG,MAAM,EAAE,CAAC;QAC7G,OAAO,KAAK,CAAC;IACf,CAAC;IACD,OAAO,KAAK,GAAG,CAAC,CAAC;AACnB,CAAC;AAED,SAAS,wBAAwB,CAAC,cAAwB,EAAE,UAAkB;IAC5E,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,IAAI,IAAI,GAAG,cAAc,CAAC,MAAM,CAAC;IACjC,OAAO,GAAG,GAAG,IAAI,EAAE,CAAC;QAClB,MAAM,GAAG,GAAG,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC;QAC/B,IAAI,cAAc,CAAC,GAAG,CAAE,GAAG,UAAU;YAAE,GAAG,GAAG,GAAG,GAAG,CAAC,CAAC;;YAChD,IAAI,GAAG,GAAG,CAAC;IAClB,CAAC;IACD,MAAM,IAAI,GAAG,GAAG,CAAC;IACjB,MAAM,WAAW,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,CAAC,cAAc,CAAC,GAAG,GAAG,CAAC,CAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IAC5D,MAAM,GAAG,GAAG,UAAU,GAAG,WAAW,GAAG,CAAC,CAAC;IACzC,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC;AACrB,CAAC"}

package/dist/chunking/chunkTextFile.d.ts

@@ -0,0 +1,28 @@
+import type { Chunk } from "./chunkFile.js";
+import type { ChunkTokenizer } from "./types.js";
+/**
+ * Options for text file chunking (JSON, YAML, config files, etc.).
+ */
+export interface TextChunkOptions {
+    /** Text content to chunk */
+    source: string;
+    /** Optional source file path for chunk IDs */
+    filePath?: string | undefined;
+    /** Language identifier (e.g., "json", "yaml", "text") */
+    languageId?: string;
+    /** Minimum tokens per chunk (default: 150). Smaller chunks are merged. */
+    minTokens?: number;
+    /** Maximum tokens per chunk (default: 400). Larger chunks are split. */
+    maxTokens?: number;
+    /** Custom token counting function (default: whitespace-based) */
+    tokenizer?: ChunkTokenizer | undefined;
+}
+/**
+ * Splits text files into chunks respecting token budgets.
+ * Useful for JSON, YAML, configuration files, and other non-code text content.
+ *
+ * @param opts Text chunking options
+ * @returns Array of text chunks
+ */
+export declare function chunkTextFile(opts: TextChunkOptions): Chunk[];
+//# sourceMappingURL=chunkTextFile.d.ts.map

package/dist/chunking/chunkTextFile.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"chunkTextFile.d.ts","sourceRoot":"","sources":["../../src/chunking/chunkTextFile.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,KAAK,EAAE,MAAM,gBAAgB,CAAC;AAI5C,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,YAAY,CAAC;AAEjD;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAC/B,4BAA4B;IAC5B,MAAM,EAAE,MAAM,CAAC;IACf,8CAA8C;IAC9C,QAAQ,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;IAC9B,yDAAyD;IACzD,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,0EAA0E;IAC1E,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,wEAAwE;IACxE,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,iEAAiE;IACjE,SAAS,CAAC,EAAE,cAAc,GAAG,SAAS,CAAC;CACxC;AAED;;;;;;GAMG;AACH,wBAAgB,aAAa,CAAC,IAAI,EAAE,gBAAgB,GAAG,KAAK,EAAE,CA4D7D"}

package/dist/chunking/chunkTextFile.js

@@ -0,0 +1,107 @@
+import { withStableChunkIds } from "./chunkId.js";
+import { splitTextWithinTokenBudget } from "./chunkSplit.js";
+import { countWhitespaceTokens } from "./tokenizer.js";
+/**
+ * Splits text files into chunks respecting token budgets.
+ * Useful for JSON, YAML, configuration files, and other non-code text content.
+ *
+ * @param opts Text chunking options
+ * @returns Array of text chunks
+ */
+export function chunkTextFile(opts) {
+    const { source, filePath, languageId = "text", maxTokens = 400, tokenizer = countWhitespaceTokens } = opts;
+    if (!source.length)
+        return [];
+    const chunks = [];
+    let currentStart = -1;
+    let currentEnd = -1;
+    let currentStartLine = 1;
+    let currentTokens = 0;
+    const flush = () => {
+        if (currentStart === -1)
+            return;
+        pushBoundedRange(chunks, source, currentStart, currentEnd, currentStartLine, languageId, filePath, tokenizer, maxTokens);
+        currentStart = -1;
+        currentEnd = -1;
+        currentTokens = 0;
+    };
+    for (const line of lineRanges(source)) {
+        const lineTokens = tokenizer(source.slice(line.start, line.end));
+        if (lineTokens > maxTokens) {
+            flush();
+            pushBoundedRange(chunks, source, line.start, line.end, line.startLine, languageId, filePath, tokenizer, maxTokens);
+            continue;
+        }
+        if (currentStart !== -1 && currentTokens + lineTokens > maxTokens) {
+            flush();
+        }
+        if (currentStart === -1) {
+            currentStart = line.start;
+            currentStartLine = line.startLine;
+        }
+        currentEnd = line.end;
+        currentTokens += lineTokens;
+    }
+    flush();
+    return withStableChunkIds(chunks, languageId, filePath);
+}
+function pushBoundedRange(chunks, source, start, end, startLine, languageId, filePath, tokenizer, maxTokens) {
+    const text = source.slice(start, end);
+    const tokenCount = tokenizer(text);
+    if (tokenCount <= maxTokens) {
+        chunks.push({
+            id: "",
+            languageId,
+            type: "text",
+            startLine,
+            endLine: endLineForText(startLine, text),
+            text,
+            tokenCount,
+            ...(filePath !== undefined ? { filePath } : {}),
+        });
+        return;
+    }
+    let segmentStartLine = startLine;
+    for (const segment of splitTextWithinTokenBudget(text, tokenizer, maxTokens)) {
+        chunks.push({
+            id: "",
+            languageId,
+            type: "text",
+            startLine: segmentStartLine,
+            endLine: endLineForText(segmentStartLine, segment),
+            text: segment,
+            tokenCount: tokenizer(segment),
+            ...(filePath !== undefined ? { filePath } : {}),
+        });
+        segmentStartLine += countLineBreaks(segment);
+    }
+}
+function lineRanges(source) {
+    const ranges = [];
+    let start = 0;
+    let line = 1;
+    for (let offset = 0; offset < source.length; offset++) {
+        if (source[offset] !== "\n")
+            continue;
+        ranges.push({ start, end: offset + 1, startLine: line });
+        start = offset + 1;
+        line++;
+    }
+    if (start < source.length) {
+        ranges.push({ start, end: source.length, startLine: line });
+    }
+    return ranges;
+}
+function countLineBreaks(text) {
+    let count = 0;
+    for (let index = 0; index < text.length; index++) {
+        if (text[index] === "\n")
+            count++;
+    }
+    return count;
+}
+function endLineForText(startLine, text) {
+    const lineBreaks = countLineBreaks(text);
+    return startLine + lineBreaks - (text.endsWith("\n") ? 1 : 0);
+}
+//# sourceMappingURL=chunkTextFile.js.map

package/dist/chunking/chunkTextFile.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"chunkTextFile.js","sourceRoot":"","sources":["../../src/chunking/chunkTextFile.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,kBAAkB,EAAE,MAAM,cAAc,CAAC;AAClD,OAAO,EAAE,0BAA0B,EAAE,MAAM,iBAAiB,CAAC;AAC7D,OAAO,EAAE,qBAAqB,EAAE,MAAM,gBAAgB,CAAC;AAqBvD;;;;;;GAMG;AACH,MAAM,UAAU,aAAa,CAAC,IAAsB;IAClD,MAAM,EAAE,MAAM,EAAE,QAAQ,EAAE,UAAU,GAAG,MAAM,EAAE,SAAS,GAAG,GAAG,EAAE,SAAS,GAAG,qBAAqB,EAAE,GAAG,IAAI,CAAC;IAC3G,IAAI,CAAC,MAAM,CAAC,MAAM;QAAE,OAAO,EAAE,CAAC;IAE9B,MAAM,MAAM,GAAY,EAAE,CAAC;IAC3B,IAAI,YAAY,GAAG,CAAC,CAAC,CAAC;IACtB,IAAI,UAAU,GAAG,CAAC,CAAC,CAAC;IACpB,IAAI,gBAAgB,GAAG,CAAC,CAAC;IACzB,IAAI,aAAa,GAAG,CAAC,CAAC;IAEtB,MAAM,KAAK,GAAG,GAAG,EAAE;QACjB,IAAI,YAAY,KAAK,CAAC,CAAC;YAAE,OAAO;QAChC,gBAAgB,CACd,MAAM,EACN,MAAM,EACN,YAAY,EACZ,UAAU,EACV,gBAAgB,EAChB,UAAU,EACV,QAAQ,EACR,SAAS,EACT,SAAS,CACV,CAAC;QACF,YAAY,GAAG,CAAC,CAAC,CAAC;QAClB,UAAU,GAAG,CAAC,CAAC,CAAC;QAChB,aAAa,GAAG,CAAC,CAAC;IACpB,CAAC,CAAC;IAEF,KAAK,MAAM,IAAI,IAAI,UAAU,CAAC,MAAM,CAAC,EAAE,CAAC;QACtC,MAAM,UAAU,GAAG,SAAS,CAAC,MAAM,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC;QACjE,IAAI,UAAU,GAAG,SAAS,EAAE,CAAC;YAC3B,KAAK,EAAE,CAAC;YACR,gBAAgB,CACd,MAAM,EACN,MAAM,EACN,IAAI,CAAC,KAAK,EACV,IAAI,CAAC,GAAG,EACR,IAAI,CAAC,SAAS,EACd,UAAU,EACV,QAAQ,EACR,SAAS,EACT,SAAS,CACV,CAAC;YACF,SAAS;QACX,CAAC;QAED,IAAI,YAAY,KAAK,CAAC,CAAC,IAAI,aAAa,GAAG,UAAU,GAAG,SAAS,EAAE,CAAC;YAClE,KAAK,EAAE,CAAC;QACV,CAAC;QAED,IAAI,YAAY,KAAK,CAAC,CAAC,EAAE,CAAC;YACxB,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC;YAC1B,gBAAgB,GAAG,IAAI,CAAC,SAAS,CAAC;QACpC,CAAC;QACD,UAAU,GAAG,IAAI,CAAC,GAAG,CAAC;QACtB,aAAa,IAAI,UAAU,CAAC;IAC9B,CAAC;IAED,KAAK,EAAE,CAAC;IACR,OAAO,kBAAkB,CAAC,MAAM,EAAE,UAAU,EAAE,QAAQ,CAAC,CAAC;AAC1D,CAAC;AAED,SAAS,gBAAgB,CACvB,MAAe,EACf,MAAc,EACd,KAAa,EACb,GAAW,EACX,SAAiB,EACjB,UAAkB,EAClB,QAA4B,EAC5B,SAAyB,EACzB,SAAiB;IAEjB,MAAM,IAAI,GAAG,MAAM,CAAC,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC;IACtC,MAAM,UAAU,GAAG,SAAS,CAAC,IAAI,CAAC,CAAC;IACnC,IAAI,UAAU,IAAI,SAAS,EAAE,CAAC;QAC5B,MAAM,CAAC,IAAI,CAAC;YACV,EAAE,EAAE,EAAE;YACN,UAAU;YACV,IAAI,EAAE,MAAM;YACZ,SAAS;YACT,OAAO,EAAE,cAAc,CAAC,SAAS,EAAE,IAAI,CAAC;YACxC,IAAI;YACJ,UAAU;YACV,GAAG,CAAC,QAAQ,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SAChD,CAAC,CAAC;QACH,OAAO;IACT,CAAC;IAED,IAAI,gBAAgB,GAAG,SAAS,CAAC;IACjC,KAAK,MAAM,OAAO,IAAI,0BAA0B,CAAC,IAAI,EAAE,SAAS,EAAE,SAAS,CAAC,EAAE,CAAC;QAC7E,MAAM,CAAC,IAAI,CAAC;YACV,EAAE,EAAE,EAAE;YACN,UAAU;YACV,IAAI,EAAE,MAAM;YACZ,SAAS,EAAE,gBAAgB;YAC3B,OAAO,EAAE,cAAc,CAAC,gBAAgB,EAAE,OAAO,CAAC;YAClD,IAAI,EAAE,OAAO;YACb,UAAU,EAAE,SAAS,CAAC,OAAO,CAAC;YAC9B,GAAG,CAAC,QAAQ,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SAChD,CAAC,CAAC;QACH,gBAAgB,IAAI,eAAe,CAAC,OAAO,CAAC,CAAC;IAC/C,CAAC;AACH,CAAC;AAED,SAAS,UAAU,CAAC,MAAc;IAChC,MAAM,MAAM,GAA6D,EAAE,CAAC;IAC5E,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,IAAI,GAAG,CAAC,CAAC;IAEb,KAAK,IAAI,MAAM,GAAG,CAAC,EAAE,MAAM,GAAG,MAAM,CAAC,MAAM,EAAE,MAAM,EAAE,EAAE,CAAC;QACtD,IAAI,MAAM,CAAC,MAAM,CAAC,KAAK,IAAI;YAAE,SAAS;QACtC,MAAM,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,GAAG,CAAC,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;QACzD,KAAK,GAAG,MAAM,GAAG,CAAC,CAAC;QACnB,IAAI,EAAE,CAAC;IACT,CAAC;IAED,IAAI,KAAK,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC;QAC1B,MAAM,CAAC,IAAI,CAAC,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,CAAC,MAAM,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;IAC9D,CAAC;IAED,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,eAAe,CAAC,IAAY;IACnC,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,IAAI,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC;QACjD,IAAI,IAAI,CAAC,KAAK,CAAC,KAAK,IAAI;YAAE,KAAK,EAAE,CAAC;IACpC,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAED,SAAS,cAAc,CAAC,SAAiB,EAAE,IAAY;IACrD,MAAM,UAAU,GAAG,eAAe,CAAC,IAAI,CAAC,CAAC;IACzC,OAAO,SAAS,GAAG,UAAU,GAAG,CAAC,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;AAChE,CAAC"}

package/dist/chunking/languageConfig.d.ts

@@ -0,0 +1,35 @@
+import type { LanguageDefinition } from "../languages/types.js";
+/**
+ * Configuration for a specific language's chunking behavior.
+ * Contains Tree-sitter parser, query, and capture group definitions.
+ */
+export interface LanguageConfig {
+    /** Language identifier */
+    id: string;
+    /** Codegraph language id used by native query execution */
+    supportId: string;
+    /** Generated query text used for chunk extraction */
+    queryText: string;
+    /** Original language definition used for JS fallback */
+    definition: LanguageDefinition;
+    /** Capture group names used in queries */
+    captures: {
+        /** Capture name for symbol names (e.g., function names) */
+        name: string;
+        /** Prefix for semantic block captures */
+        blockPrefix: string;
+        /** Capture name for inner control-flow blocks */
+        innerBlock: string;
+        /** Capture names for comments and docstrings */
+        comments: string[];
+    };
+}
+/**
+ * Creates a language configuration for semantic chunking.
+ *
+ * @param def Language definition
+ * @param publicId Public language id exposed by chunking output and lookup tables
+ * @returns Language configuration object
+ */
+export declare function makeLanguageConfig(def: LanguageDefinition, publicId?: string): LanguageConfig;
+//# sourceMappingURL=languageConfig.d.ts.map

package/dist/chunking/languageConfig.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"languageConfig.d.ts","sourceRoot":"","sources":["../../src/chunking/languageConfig.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,uBAAuB,CAAC;AAGhE;;;GAGG;AACH,MAAM,WAAW,cAAc;IAC7B,0BAA0B;IAC1B,EAAE,EAAE,MAAM,CAAC;IACX,2DAA2D;IAC3D,SAAS,EAAE,MAAM,CAAC;IAClB,qDAAqD;IACrD,SAAS,EAAE,MAAM,CAAC;IAClB,wDAAwD;IACxD,UAAU,EAAE,kBAAkB,CAAC;IAC/B,0CAA0C;IAC1C,QAAQ,EAAE;QACR,2DAA2D;QAC3D,IAAI,EAAE,MAAM,CAAC;QACb,yCAAyC;QACzC,WAAW,EAAE,MAAM,CAAC;QACpB,iDAAiD;QACjD,UAAU,EAAE,MAAM,CAAC;QACnB,gDAAgD;QAChD,QAAQ,EAAE,MAAM,EAAE,CAAC;KACpB,CAAC;CACH;AAED;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,GAAG,EAAE,kBAAkB,EAAE,QAAQ,SAAS,GAAG,cAAc,CAe7F"}

package/dist/chunking/languageConfig.js

@@ -0,0 +1,24 @@
+import { generateChunkingQuery } from "../languages/queryGenerator.js";
+/**
+ * Creates a language configuration for semantic chunking.
+ *
+ * @param def Language definition
+ * @param publicId Public language id exposed by chunking output and lookup tables
+ * @returns Language configuration object
+ */
+export function makeLanguageConfig(def, publicId = def.id) {
+    const queryText = generateChunkingQuery(def);
+    return {
+        id: publicId,
+        supportId: def.id,
+        queryText,
+        definition: def,
+        captures: {
+            name: "chunk.name",
+            blockPrefix: "chunk.block.",
+            innerBlock: "chunk.block.inner",
+            comments: ["chunk.comment", "chunk.docstring"],
+        },
+    };
+}
+//# sourceMappingURL=languageConfig.js.map

package/dist/chunking/languageConfig.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"languageConfig.js","sourceRoot":"","sources":["../../src/chunking/languageConfig.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,qBAAqB,EAAE,MAAM,gCAAgC,CAAC;AA4BvE;;;;;;GAMG;AACH,MAAM,UAAU,kBAAkB,CAAC,GAAuB,EAAE,QAAQ,GAAG,GAAG,CAAC,EAAE;IAC3E,MAAM,SAAS,GAAG,qBAAqB,CAAC,GAAG,CAAC,CAAC;IAE7C,OAAO;QACL,EAAE,EAAE,QAAQ;QACZ,SAAS,EAAE,GAAG,CAAC,EAAE;QACjB,SAAS;QACT,UAAU,EAAE,GAAG;QACf,QAAQ,EAAE;YACR,IAAI,EAAE,YAAY;YAClB,WAAW,EAAE,cAAc;YAC3B,UAAU,EAAE,mBAAmB;YAC/B,QAAQ,EAAE,CAAC,eAAe,EAAE,iBAAiB,CAAC;SAC/C;KACF,CAAC;AACJ,CAAC"}

package/dist/chunking/tokenizer.d.ts

@@ -0,0 +1,2 @@
+export declare function countWhitespaceTokens(text: string): number;
+//# sourceMappingURL=tokenizer.d.ts.map

package/dist/chunking/tokenizer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokenizer.d.ts","sourceRoot":"","sources":["../../src/chunking/tokenizer.ts"],"names":[],"mappings":"AAAA,wBAAgB,qBAAqB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAG1D"}

package/dist/chunking/tokenizer.js

@@ -0,0 +1,6 @@
+export function countWhitespaceTokens(text) {
+    if (!text.trim())
+        return 0;
+    return text.trim().split(/\s+/).length;
+}
+//# sourceMappingURL=tokenizer.js.map

package/dist/chunking/tokenizer.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokenizer.js","sourceRoot":"","sources":["../../src/chunking/tokenizer.ts"],"names":[],"mappings":"AAAA,MAAM,UAAU,qBAAqB,CAAC,IAAY;IAChD,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;QAAE,OAAO,CAAC,CAAC;IAC3B,OAAO,IAAI,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,MAAM,CAAC;AACzC,CAAC"}

package/dist/chunking/types.d.ts

@@ -0,0 +1,54 @@
+/**
+ * Represents a semantic chunk of code or text, ready for LLM processing or vector embeddings.
+ */
+export interface Chunk {
+    /** Unique identifier for the chunk */
+    id: string;
+    /** Language identifier (e.g., "javascript", "typescript", "python") */
+    languageId: string;
+    /** Optional source file path */
+    filePath?: string | undefined;
+    /** Chunk type (e.g., "function", "class", "method", "import", "misc") */
+    type: string;
+    /** Symbol name if applicable (e.g., function name, class name) */
+    name?: string;
+    /** 1-based start line number */
+    startLine: number;
+    /** 1-based end line number */
+    endLine: number;
+    /** The chunk content text */
+    text: string;
+    /** Estimated token count */
+    tokenCount: number;
+}
+export interface BlockCandidate {
+    kind: string;
+    name?: string;
+    /** Start offset in JS string index units (UTF-16 code units). */
+    startByte: number;
+    /** End offset in JS string index units (UTF-16 code units). */
+    endByte: number;
+    startLine: number;
+    endLine: number;
+}
+export type ChunkCapture = {
+    name: string;
+    text: string;
+    /** Start offset in JS string index units (UTF-16 code units). */
+    startByte: number;
+    /** End offset in JS string index units (UTF-16 code units). */
+    endByte: number;
+    startLine: number;
+    endLine: number;
+    nodeType: string;
+};
+export type ChunkMatch = {
+    captures: ChunkCapture[];
+};
+export type ChunkTokenizer = (text: string) => number;
+/** Internal source offsets used while assembling non-overlapping chunks. */
+export type RangedChunk = Chunk & {
+    sourceStart: number;
+    sourceEnd: number;
+};
+//# sourceMappingURL=types.d.ts.map

package/dist/chunking/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../src/chunking/types.ts"],"names":[],"mappings":"AAAA;;GAEG;AACH,MAAM,WAAW,KAAK;IACpB,sCAAsC;IACtC,EAAE,EAAE,MAAM,CAAC;IACX,uEAAuE;IACvE,UAAU,EAAE,MAAM,CAAC;IACnB,gCAAgC;IAChC,QAAQ,CAAC,EAAE,MAAM,GAAG,SAAS,CAAC;IAC9B,yEAAyE;IACzE,IAAI,EAAE,MAAM,CAAC;IACb,kEAAkE;IAClE,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,gCAAgC;IAChC,SAAS,EAAE,MAAM,CAAC;IAClB,8BAA8B;IAC9B,OAAO,EAAE,MAAM,CAAC;IAChB,6BAA6B;IAC7B,IAAI,EAAE,MAAM,CAAC;IACb,4BAA4B;IAC5B,UAAU,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,WAAW,cAAc;IAC7B,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,iEAAiE;IACjE,SAAS,EAAE,MAAM,CAAC;IAClB,+DAA+D;IAC/D,OAAO,EAAE,MAAM,CAAC;IAChB,SAAS,EAAE,MAAM,CAAC;IAClB,OAAO,EAAE,MAAM,CAAC;CACjB;AAED,MAAM,MAAM,YAAY,GAAG;IACzB,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,EAAE,MAAM,CAAC;IACb,iEAAiE;IACjE,SAAS,EAAE,MAAM,CAAC;IAClB,+DAA+D;IAC/D,OAAO,EAAE,MAAM,CAAC;IAChB,SAAS,EAAE,MAAM,CAAC;IAClB,OAAO,EAAE,MAAM,CAAC;IAChB,QAAQ,EAAE,MAAM,CAAC;CAClB,CAAC;AAEF,MAAM,MAAM,UAAU,GAAG;IACvB,QAAQ,EAAE,YAAY,EAAE,CAAC;CAC1B,CAAC;AAEF,MAAM,MAAM,cAAc,GAAG,CAAC,IAAI,EAAE,MAAM,KAAK,MAAM,CAAC;AAEtD,4EAA4E;AAC5E,MAAM,MAAM,WAAW,GAAG,KAAK,GAAG;IAChC,WAAW,EAAE,MAAM,CAAC;IACpB,SAAS,EAAE,MAAM,CAAC;CACnB,CAAC"}

package/dist/chunking/types.js

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=types.js.map

package/dist/chunking/types.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.js","sourceRoot":"","sources":["../../src/chunking/types.ts"],"names":[],"mappings":""}

package/dist/config.d.ts

@@ -0,0 +1,18 @@
+import type { LanguageExtensionMap } from "./languages.js";
+import type { GraphBuildOptions } from "./graphs/types.js";
+import { type ProjectFileDiscoveryOptions } from "./util/projectFiles.js";
+export declare const CODEGRAPH_CONFIG_FILE = "codegraph.config.json";
+export type CodegraphConfig = {
+    discovery?: ProjectFileDiscoveryOptions;
+    languages?: {
+        extensions?: LanguageExtensionMap;
+    };
+    graph?: {
+        resolutionHints?: string[];
+    };
+};
+export declare function hasDiscoveryOptions(discovery: ProjectFileDiscoveryOptions): boolean;
+export declare function mergeDiscoveryOptions(base: ProjectFileDiscoveryOptions | undefined, override: ProjectFileDiscoveryOptions | undefined): ProjectFileDiscoveryOptions;
+export declare function mergeGraphOptions(base: CodegraphConfig["graph"] | undefined, override: GraphBuildOptions | undefined): GraphBuildOptions;
+export declare function loadCodegraphConfig(projectRoot: string): Promise<CodegraphConfig>;
+//# sourceMappingURL=config.d.ts.map

package/dist/config.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"config.d.ts","sourceRoot":"","sources":["../src/config.ts"],"names":[],"mappings":"AASA,OAAO,KAAK,EAAE,oBAAoB,EAAE,MAAM,gBAAgB,CAAC;AAC3D,OAAO,KAAK,EAAE,iBAAiB,EAAE,MAAM,mBAAmB,CAAC;AAG3D,OAAO,EAAE,KAAK,2BAA2B,EAAE,MAAM,wBAAwB,CAAC;AAE1E,eAAO,MAAM,qBAAqB,0BAA0B,CAAC;AAiC7D,MAAM,MAAM,eAAe,GAAG;IAC5B,SAAS,CAAC,EAAE,2BAA2B,CAAC;IACxC,SAAS,CAAC,EAAE;QACV,UAAU,CAAC,EAAE,oBAAoB,CAAC;KACnC,CAAC;IACF,KAAK,CAAC,EAAE;QACN,eAAe,CAAC,EAAE,MAAM,EAAE,CAAC;KAC5B,CAAC;CACH,CAAC;AAWF,wBAAgB,mBAAmB,CAAC,SAAS,EAAE,2BAA2B,GAAG,OAAO,CASnF;AAED,wBAAgB,qBAAqB,CACnC,IAAI,EAAE,2BAA2B,GAAG,SAAS,EAC7C,QAAQ,EAAE,2BAA2B,GAAG,SAAS,GAChD,2BAA2B,CAe7B;AACD,wBAAgB,iBAAiB,CAC/B,IAAI,EAAE,eAAe,CAAC,OAAO,CAAC,GAAG,SAAS,EAC1C,QAAQ,EAAE,iBAAiB,GAAG,SAAS,GACtC,iBAAiB,CASnB;AAyCD,wBAAsB,mBAAmB,CAAC,WAAW,EAAE,MAAM,GAAG,OAAO,CAAC,eAAe,CAAC,CAiCvF"}

package/dist/config.js

@@ -0,0 +1,141 @@
+import fsp from "node:fs/promises";
+import path from "node:path";
+import { z } from "zod";
+import { isLiteralLanguageExtension, isRemappableLanguageExtension, normalizeLanguageExtensions, supportById, } from "./languages.js";
+import { normalizeResolutionHints } from "./util/paths.js";
+import { errorMessage } from "./util/errors.js";
+import {} from "./util/projectFiles.js";
+export const CODEGRAPH_CONFIG_FILE = "codegraph.config.json";
+const stringArraySchema = z.array(z.string().trim().min(1));
+const languageExtensionsSchema = z.record(z.string().trim().min(1), z.string().trim().min(1));
+const codegraphConfigSchema = z
+    .object({
+    discovery: z
+        .object({
+        includeGlobs: stringArraySchema.optional(),
+        ignoreGlobs: stringArraySchema.optional(),
+        useGitignore: z.boolean().optional(),
+    })
+        .strict()
+        .optional(),
+    languages: z
+        .object({
+        extensions: languageExtensionsSchema.optional(),
+    })
+        .strict()
+        .optional(),
+    graph: z
+        .object({
+        resolutionHints: stringArraySchema.optional(),
+    })
+        .strict()
+        .optional(),
+})
+    .strict();
+function uniq(values) {
+    return Array.from(new Set(values.map((value) => value.trim().replace(/\\/g, "/")).filter(Boolean)));
+}
+function normalizeDiscoveryRoot(root) {
+    const normalized = root?.trim().replace(/\\/g, "/");
+    return normalized ? normalized : undefined;
+}
+export function hasDiscoveryOptions(discovery) {
+    return Boolean(discovery.includeGlobs?.length ||
+        discovery.ignoreGlobs?.length ||
+        discovery.useGitignore !== undefined ||
+        normalizeDiscoveryRoot(discovery.globRoot) ||
+        normalizeDiscoveryRoot(discovery.gitignoreRoot) ||
+        discovery.logLevel);
+}
+export function mergeDiscoveryOptions(base, override) {
+    const includeGlobs = uniq([...(base?.includeGlobs ?? []), ...(override?.includeGlobs ?? [])]);
+    const ignoreGlobs = uniq([...(base?.ignoreGlobs ?? []), ...(override?.ignoreGlobs ?? [])]);
+    const useGitignore = override?.useGitignore ?? base?.useGitignore;
+    const globRoot = normalizeDiscoveryRoot(override?.globRoot) ?? normalizeDiscoveryRoot(base?.globRoot);
+    const gitignoreRoot = normalizeDiscoveryRoot(override?.gitignoreRoot) ?? normalizeDiscoveryRoot(base?.gitignoreRoot);
+    const logLevel = override?.logLevel ?? base?.logLevel;
+    return {
+        ...(includeGlobs.length ? { includeGlobs } : {}),
+        ...(ignoreGlobs.length ? { ignoreGlobs } : {}),
+        ...(useGitignore !== undefined ? { useGitignore } : {}),
+        ...(globRoot !== undefined ? { globRoot } : {}),
+        ...(gitignoreRoot !== undefined ? { gitignoreRoot } : {}),
+        ...(logLevel !== undefined ? { logLevel } : {}),
+    };
+}
+export function mergeGraphOptions(base, override) {
+    const resolutionHints = normalizeResolutionHints([
+        ...(base?.resolutionHints ?? []),
+        ...(override?.resolutionHints ?? []),
+    ]);
+    return {
+        ...override,
+        ...(resolutionHints.length ? { resolutionHints } : {}),
+    };
+}
+function normalizeDiscoveryConfig(discovery) {
+    if (!discovery)
+        return undefined;
+    const normalized = {
+        ...(discovery.includeGlobs !== undefined ? { includeGlobs: discovery.includeGlobs } : {}),
+        ...(discovery.ignoreGlobs !== undefined ? { ignoreGlobs: discovery.ignoreGlobs } : {}),
+        ...(discovery.useGitignore !== undefined ? { useGitignore: discovery.useGitignore } : {}),
+    };
+    return hasDiscoveryOptions(normalized) ? normalized : undefined;
+}
+function normalizeConfigLanguageExtensions(extensions) {
+    if (!extensions)
+        return undefined;
+    for (const [rawKey, rawLanguageId] of Object.entries(extensions)) {
+        const key = rawKey.trim().toLowerCase();
+        const languageId = rawLanguageId.trim().toLowerCase();
+        if (!key.startsWith(".")) {
+            throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: languages.extensions key "${rawKey}" must start with ".".`);
+        }
+        if (!isLiteralLanguageExtension(key)) {
+            throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: languages.extensions key "${rawKey}" must be a literal suffix containing only letters, digits, ".", "_", "+", or "-".`);
+        }
+        if (!isRemappableLanguageExtension(key)) {
+            throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: languages.extensions key "${rawKey}" cannot be remapped.`);
+        }
+        if (!supportById(languageId)) {
+            throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: languages.extensions["${rawKey}"] references unknown language "${languageId}".`);
+        }
+    }
+    return normalizeLanguageExtensions(extensions);
+}
+export async function loadCodegraphConfig(projectRoot) {
+    const configPath = path.join(projectRoot, CODEGRAPH_CONFIG_FILE);
+    let raw;
+    try {
+        raw = await fsp.readFile(configPath, "utf8");
+    }
+    catch (error) {
+        if (error instanceof Error && "code" in error && error.code === "ENOENT") {
+            return {};
+        }
+        throw error;
+    }
+    let parsedJson;
+    try {
+        parsedJson = JSON.parse(raw);
+    }
+    catch (error) {
+        const message = errorMessage(error);
+        throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: ${message}`);
+    }
+    const parsed = codegraphConfigSchema.safeParse(parsedJson);
+    if (!parsed.success) {
+        throw new Error(`Invalid ${CODEGRAPH_CONFIG_FILE}: ${z.prettifyError(parsed.error)}`);
+    }
+    const languageExtensions = normalizeConfigLanguageExtensions(parsed.data.languages?.extensions);
+    const discovery = normalizeDiscoveryConfig(parsed.data.discovery);
+    const resolutionHints = normalizeResolutionHints(parsed.data.graph?.resolutionHints);
+    const graph = resolutionHints.length ? { resolutionHints } : undefined;
+    return {
+        ...(discovery ? { discovery } : {}),
+        ...(graph ? { graph } : {}),
+        ...(languageExtensions ? { languages: { extensions: languageExtensions } } : {}),
+    };
+}
+//# sourceMappingURL=config.js.map