opkg 0.6.1 → 0.7.0

package/dist/core/flows/flow-executor.js

@@ -0,0 +1,974 @@
+/**
+ * Flow Executor
+ *
+ * Executes flows through a multi-stage pipeline:
+ * 1. Load source file and parse format
+ * 2. Extract JSONPath (if specified)
+ * 3. Pick/omit keys
+ * 4. Map keys (with transforms)
+ * 5. Apply pipe transforms
+ * 6. Embed in target structure
+ * 7. Merge with existing target (priority-based)
+ * 8. Write to target file
+ */
+import fsSync from 'fs';
+import path from 'path';
+import yaml from 'js-yaml';
+import * as TOML from 'smol-toml';
+import { parse as parseJsonc } from 'jsonc-parser';
+import { JSONPath } from 'jsonpath-plus';
+import * as fsUtils from '../../utils/fs.js';
+import { mergePackageContentIntoRootFile } from '../../utils/root-file-merger.js';
+import { logger } from '../../utils/logger.js';
+import { defaultTransformRegistry, } from './flow-transforms.js';
+import { mergeInlinePlatformOverride } from '../../utils/platform-yaml-merge.js';
+import { getNestedValue, setNestedValue, deleteNestedValue } from './flow-key-mapper.js';
+import { extractAllKeys } from './flow-key-extractor.js';
+import { applyMapPipeline, createMapContext, validateMapPipeline } from './map-pipeline/index.js';
+import { SourcePatternResolver } from './source-resolver.js';
+/**
+ * Default flow executor implementation
+ */
+export class DefaultFlowExecutor {
+    constructor(transformRegistry) {
+        this.transformRegistry = transformRegistry || defaultTransformRegistry;
+        this.sourceResolver = new SourcePatternResolver();
+    }
+    /**
+     * Execute a single flow (now supports glob patterns)
+     */
+    async executeFlow(flow, context) {
+        const startTime = Date.now();
+        try {
+            // Check if this is a multi-target flow
+            if (typeof flow.to !== 'string') {
+                const results = await this.executeMultiTarget(flow, context);
+                // Aggregate results
+                return this.aggregateResults(results, startTime);
+            }
+            // Validate flow
+            const validation = this.validateFlow(flow);
+            if (!validation.valid) {
+                return {
+                    source: this.normalizeFromPattern(flow.from),
+                    target: flow.to,
+                    success: false,
+                    transformed: false,
+                    error: new Error(`Invalid flow: ${validation.errors.map(e => e.message).join(', ')}`),
+                    executionTime: Date.now() - startTime,
+                };
+            }
+            // Evaluate conditions
+            if (flow.when && !this.evaluateCondition(flow.when, context)) {
+                const normalized = this.normalizeFromPattern(flow.from);
+                logger.debug(`Flow skipped due to condition: ${normalized} -> ${flow.to}`);
+                return {
+                    source: normalized,
+                    target: flow.to,
+                    success: true,
+                    transformed: false,
+                    warnings: ['Flow skipped due to condition'],
+                    executionTime: Date.now() - startTime,
+                };
+            }
+            // Resolve source paths (may return multiple files for glob patterns)
+            const resolution = await this.resolveSourcePattern(flow.from, context);
+            const sourcePaths = resolution.paths;
+            const resolutionWarnings = resolution.warnings;
+            // If no files matched, return success with no files processed
+            if (sourcePaths.length === 0) {
+                return {
+                    source: this.normalizeFromPattern(flow.from),
+                    target: flow.to,
+                    success: true,
+                    transformed: false,
+                    warnings: resolutionWarnings.length > 0 ? resolutionWarnings : ['No files matched pattern'],
+                    executionTime: Date.now() - startTime,
+                };
+            }
+            // Execute pipeline for each matched file
+            const results = [];
+            const firstFromPattern = this.getFirstPattern(flow.from);
+            for (const sourcePath of sourcePaths) {
+                const targetPath = this.resolveTargetFromGlob(sourcePath, firstFromPattern, flow.to, context);
+                const result = await this.executePipeline(flow, sourcePath, targetPath, context);
+                results.push({
+                    ...result,
+                    executionTime: Date.now() - startTime,
+                });
+            }
+            // Add resolution warnings to first result if any
+            if (resolutionWarnings.length > 0 && results.length > 0) {
+                results[0].warnings = [
+                    ...(results[0].warnings || []),
+                    ...resolutionWarnings,
+                ];
+            }
+            // If single file, return single result
+            if (results.length === 1) {
+                return results[0];
+            }
+            // Aggregate multiple results
+            return this.aggregateResults(results, startTime);
+        }
+        catch (error) {
+            return {
+                source: this.normalizeFromPattern(flow.from),
+                target: typeof flow.to === 'string' ? flow.to : Object.keys(flow.to).join(', '),
+                success: false,
+                transformed: false,
+                error: error instanceof Error ? error : new Error(String(error)),
+                executionTime: Date.now() - startTime,
+            };
+        }
+    }
+    /**
+     * Execute multiple flows
+     */
+    async executeFlows(flows, context) {
+        const results = [];
+        for (const flow of flows) {
+            const result = await this.executeFlow(flow, context);
+            results.push(result);
+        }
+        return results;
+    }
+    /**
+     * Execute a multi-target flow
+     */
+    async executeMultiTarget(flow, context) {
+        if (typeof flow.to === 'string') {
+            throw new Error('Flow is not a multi-target flow');
+        }
+        const multiTarget = flow.to;
+        const normalizedFrom = this.normalizeFromPattern(flow.from);
+        // Resolve source paths (may be multiple files with glob)
+        const resolution = await this.resolveSourcePattern(flow.from, context);
+        const sourcePaths = resolution.paths;
+        // If no files matched
+        if (sourcePaths.length === 0) {
+            return Object.keys(multiTarget).map(target => ({
+                source: normalizedFrom,
+                target,
+                success: true,
+                transformed: false,
+                warnings: resolution.warnings.length > 0 ? resolution.warnings : ['No files matched pattern'],
+            }));
+        }
+        // Execute each source file
+        const allResults = [];
+        const firstFromPattern = this.getFirstPattern(flow.from);
+        for (const sourcePath of sourcePaths) {
+            // Load and parse source once
+            const sourceContent = await this.loadSourceFile(sourcePath, context);
+            // Execute each target
+            for (const [targetPath, targetFlow] of Object.entries(multiTarget)) {
+                const startTime = Date.now();
+                try {
+                    // Merge target flow with base flow
+                    const mergedFlow = {
+                        ...flow,
+                        ...targetFlow,
+                        from: flow.from,
+                        to: targetPath,
+                    };
+                    // Evaluate conditions
+                    if (mergedFlow.when && !this.evaluateCondition(mergedFlow.when, context)) {
+                        logger.debug(`Multi-target flow skipped due to condition: ${normalizedFrom} -> ${targetPath}`);
+                        allResults.push({
+                            source: normalizedFrom,
+                            target: targetPath,
+                            success: true,
+                            transformed: false,
+                            warnings: ['Flow skipped due to condition'],
+                            executionTime: Date.now() - startTime,
+                        });
+                        continue;
+                    }
+                    const resolvedTargetPath = this.resolveTargetFromGlob(sourcePath, firstFromPattern, targetPath, context);
+                    // Execute pipeline with pre-loaded content
+                    const result = await this.executePipelineWithContent(mergedFlow, sourceContent, sourcePath, resolvedTargetPath, context);
+                    allResults.push({
+                        ...result,
+                        executionTime: Date.now() - startTime,
+                    });
+                }
+                catch (error) {
+                    allResults.push({
+                        source: normalizedFrom,
+                        target: targetPath,
+                        success: false,
+                        transformed: false,
+                        error: error instanceof Error ? error : new Error(String(error)),
+                        executionTime: Date.now() - startTime,
+                    });
+                }
+            }
+        }
+        return allResults;
+    }
+    /**
+     * Validate a flow configuration
+     */
+    validateFlow(flow) {
+        const errors = [];
+        // Check required fields
+        if (!flow.from) {
+            errors.push({ message: 'Flow missing required field "from"', code: 'MISSING_FROM' });
+        }
+        if (!flow.to) {
+            errors.push({ message: 'Flow missing required field "to"', code: 'MISSING_TO' });
+        }
+        // Validate pick/omit
+        if (flow.pick && flow.omit) {
+            errors.push({ message: 'Flow cannot have both "pick" and "omit"', code: 'CONFLICTING_FILTERS' });
+        }
+        // Validate merge strategy
+        if (flow.merge && !['deep', 'shallow', 'replace', 'composite'].includes(flow.merge)) {
+            errors.push({
+                message: `Invalid merge strategy: ${flow.merge}. Must be one of: deep, shallow, replace, composite`,
+                code: 'INVALID_MERGE',
+            });
+        }
+        // Validate JSONPath expression
+        if (flow.path) {
+            try {
+                // Try to validate JSONPath syntax
+                JSONPath({ path: flow.path, json: {} });
+            }
+            catch (error) {
+                errors.push({
+                    message: `Invalid JSONPath expression: ${flow.path}`,
+                    code: 'INVALID_JSONPATH',
+                });
+            }
+        }
+        // Validate map pipeline
+        if (flow.map) {
+            const mapPipelineValidation = validateMapPipeline(flow.map);
+            if (!mapPipelineValidation.valid) {
+                for (const error of mapPipelineValidation.errors) {
+                    errors.push({
+                        message: error,
+                        code: 'INVALID_MAP_PIPELINE',
+                    });
+                }
+            }
+        }
+        return {
+            valid: errors.length === 0,
+            errors,
+            warnings: [],
+        };
+    }
+    /**
+     * Execute the transformation pipeline
+     */
+    async executePipeline(flow, sourcePath, targetPath, context) {
+        // Step 1: Load source file
+        const sourceContent = await this.loadSourceFile(sourcePath, context);
+        return this.executePipelineWithContent(flow, sourceContent, sourcePath, targetPath, context);
+    }
+    /**
+     * Execute pipeline with pre-loaded content (for multi-target flows)
+     */
+    async executePipelineWithContent(flow, sourceContent, sourcePath, targetPath, context) {
+        const warnings = [];
+        const conflicts = [];
+        let data = sourceContent.data;
+        let transformed = false;
+        try {
+            // Step 2: Extract JSONPath (if specified)
+            if (flow.path) {
+                data = this.extractJSONPath(data, flow.path);
+                transformed = true;
+            }
+            // Step 3: Pick/omit keys
+            if (flow.pick) {
+                data = this.pickKeys(data, flow.pick);
+                transformed = true;
+            }
+            else if (flow.omit) {
+                data = this.omitKeys(data, flow.omit);
+                transformed = true;
+            }
+            // Step 4: Apply map pipeline
+            // Split into schema operations and pipe operations
+            // Schema ops are applied BEFORE merge, pipe ops are applied AFTER merge
+            let contributedKeys;
+            let schemaOps = [];
+            let pipeOps = [];
+            if (flow.map) {
+                // Separate schema operations from pipe operations
+                for (const op of flow.map) {
+                    if ('$pipe' in op) {
+                        pipeOps.push(op);
+                    }
+                    else {
+                        schemaOps.push(op);
+                    }
+                }
+                // Apply schema operations first (before merge)
+                if (schemaOps.length > 0) {
+                    const mapContext = createMapContext({
+                        filename: path.basename(sourcePath, path.extname(sourcePath)),
+                        dirname: path.basename(path.dirname(sourcePath)),
+                        path: path.relative(context.packageRoot, sourcePath),
+                        ext: path.extname(sourcePath),
+                    });
+                    // For markdown files, apply to frontmatter
+                    if (data && typeof data === 'object' && 'frontmatter' in data) {
+                        data.frontmatter = applyMapPipeline(data.frontmatter, schemaOps, mapContext, this.transformRegistry);
+                    }
+                    else {
+                        // Apply to entire document
+                        data = applyMapPipeline(data, schemaOps, mapContext, this.transformRegistry);
+                    }
+                    transformed = true;
+                }
+            }
+            // Track keys AFTER schema transforms but BEFORE merge and pipe transforms
+            // This represents the structured data this package contributes
+            const shouldTrackKeys = Boolean(flow.merge) &&
+                flow.merge !== 'replace' &&
+                flow.merge !== 'composite';
+            if (shouldTrackKeys && typeof data === 'object' && data !== null) {
+                // Extract from frontmatter if it's a markdown file
+                const dataToExtract = (data && 'frontmatter' in data) ? data.frontmatter : data;
+                if (typeof dataToExtract === 'object' && dataToExtract !== null) {
+                    contributedKeys = extractAllKeys(dataToExtract);
+                }
+            }
+            const targetExists = await fsUtils.exists(targetPath);
+            // Step 6: Embed in target structure
+            if (flow.embed) {
+                data = this.embedContent(data, flow.embed);
+                transformed = true;
+            }
+            // Step 7: Merge with existing target (if needed)
+            if (targetExists) {
+                const targetContent = await this.loadSourceFile(targetPath, context);
+                // Special handling for composite merge - works with raw text
+                if (flow.merge === 'composite') {
+                    // Use raw content for composite merge
+                    const sourceRaw = sourceContent.raw;
+                    const targetRaw = targetContent.raw;
+                    data = mergePackageContentIntoRootFile(targetRaw, context.packageName, sourceRaw);
+                    transformed = true;
+                }
+                else {
+                    // Normal merge for other strategies
+                    const mergeResult = this.mergeContent(data, targetContent.data, flow.merge || 'replace', context);
+                    data = mergeResult.data;
+                    conflicts.push(...mergeResult.conflicts);
+                    if (mergeResult.conflicts.length > 0) {
+                        transformed = true;
+                    }
+                }
+            }
+            // Step 7.5: Apply pipe operations AFTER merge (format conversions)
+            // These operations may convert the data to a string format (e.g., json-to-toml)
+            if (pipeOps.length > 0) {
+                const mapContext = createMapContext({
+                    filename: path.basename(sourcePath, path.extname(sourcePath)),
+                    dirname: path.basename(path.dirname(sourcePath)),
+                    path: path.relative(context.packageRoot, sourcePath),
+                    ext: path.extname(sourcePath),
+                });
+                // For markdown files, apply to frontmatter
+                if (data && typeof data === 'object' && 'frontmatter' in data) {
+                    data.frontmatter = applyMapPipeline(data.frontmatter, pipeOps, mapContext, this.transformRegistry);
+                }
+                else {
+                    // Apply to entire document
+                    data = applyMapPipeline(data, pipeOps, mapContext, this.transformRegistry);
+                }
+                transformed = true;
+            }
+            // Step 8: Write to target file
+            if (!context.dryRun) {
+                await this.writeTargetFile(targetPath, data, sourceContent.format);
+            }
+            return {
+                source: sourcePath,
+                target: targetPath,
+                success: true,
+                transformed,
+                keys: contributedKeys,
+                merge: flow.merge,
+                warnings: warnings.length > 0 ? warnings : undefined,
+                conflicts: conflicts.length > 0 ? conflicts : undefined,
+                pipeline: this.getPipeline(flow),
+            };
+        }
+        catch (error) {
+            return {
+                source: sourcePath,
+                target: targetPath,
+                success: false,
+                transformed,
+                error: error instanceof Error ? error : new Error(String(error)),
+                warnings: warnings.length > 0 ? warnings : undefined,
+            };
+        }
+    }
+    /**
+     * Load and parse source file
+     */
+    async loadSourceFile(filePath, context) {
+        let raw = await fsUtils.readTextFile(filePath);
+        const format = this.detectFormat(filePath, raw);
+        // Apply platform-specific frontmatter overrides for markdown files during install
+        if ((format === 'markdown' || format === 'md') && context?.platform && context?.direction === 'install') {
+            raw = mergeInlinePlatformOverride(raw, context.platform, context.workspaceRoot);
+        }
+        const data = this.parseSourceContent(raw, format);
+        return { data, format, raw };
+    }
+    /**
+     * Write transformed content to target file
+     */
+    async writeTargetFile(filePath, content, sourceFormat) {
+        await fsUtils.ensureDir(path.dirname(filePath));
+        // Detect target format from file extension
+        const targetFormat = this.detectFormat(filePath, '');
+        const serialized = this.serializeTargetContent(content, targetFormat);
+        await fsUtils.writeTextFile(filePath, serialized);
+    }
+    /**
+     * Parse source content based on format
+     */
+    parseSourceContent(content, format) {
+        try {
+            switch (format) {
+                case 'json':
+                case 'jsonc':
+                    // Remove comments for JSONC
+                    const cleaned = content;
+                    if (format === 'jsonc') {
+                        const parsed = parseJsonc(cleaned);
+                        if (parsed === undefined) {
+                            throw new Error('jsonc-parser returned undefined');
+                        }
+                        return parsed;
+                    }
+                    return JSON.parse(cleaned);
+                case 'yaml':
+                case 'yml':
+                    return yaml.load(content);
+                case 'toml':
+                    try {
+                        return TOML.parse(content);
+                    }
+                    catch (error) {
+                        throw new Error(`TOML parse error: ${error instanceof Error ? error.message : String(error)}`);
+                    }
+                case 'markdown':
+                case 'md':
+                    return this.parseMarkdown(content);
+                case 'text':
+                case 'txt':
+                default:
+                    return content;
+            }
+        }
+        catch (error) {
+            throw new Error(`Failed to parse ${format} content: ${error instanceof Error ? error.message : String(error)}`);
+        }
+    }
+    /**
+     * Serialize content to target format
+     */
+    serializeTargetContent(content, format) {
+        try {
+            switch (format) {
+                case 'json':
+                case 'jsonc':
+                    return JSON.stringify(content, null, 2);
+                case 'yaml':
+                case 'yml':
+                    return yaml.dump(content, { indent: 2, lineWidth: -1 });
+                case 'toml':
+                    // If a pipeline already produced TOML text (e.g. via domain transforms),
+                    // don't stringify again.
+                    if (typeof content === 'string') {
+                        return content;
+                    }
+                    try {
+                        // Serialize to TOML
+                        let toml = TOML.stringify(content);
+                        // Apply inline table formatting for Codex MCP configs
+                        if (content && typeof content === 'object' && content.mcp_servers) {
+                            toml = this.applyCodexTomlFormatting(toml);
+                        }
+                        return toml;
+                    }
+                    catch (error) {
+                        throw new Error(`TOML stringify error: ${error instanceof Error ? error.message : String(error)}`);
+                    }
+                case 'markdown':
+                case 'md':
+                    return this.serializeMarkdown(content);
+                case 'text':
+                case 'txt':
+                default:
+                    return typeof content === 'string' ? content : JSON.stringify(content, null, 2);
+            }
+        }
+        catch (error) {
+            throw new Error(`Failed to serialize ${format} content: ${error instanceof Error ? error.message : String(error)}`);
+        }
+    }
+    /**
+     * Detect format from file extension or content
+     */
+    detectFormat(filePath, content) {
+        const ext = path.extname(filePath).toLowerCase();
+        const extMap = {
+            '.json': 'json',
+            '.jsonc': 'jsonc',
+            '.yaml': 'yaml',
+            '.yml': 'yml',
+            '.toml': 'toml',
+            '.md': 'markdown',
+            '.markdown': 'markdown',
+            '.txt': 'text',
+        };
+        if (extMap[ext]) {
+            return extMap[ext];
+        }
+        // Try to detect from content
+        if (content.trim().startsWith('{') || content.trim().startsWith('[')) {
+            return content.includes('//') || content.includes('/*') ? 'jsonc' : 'json';
+        }
+        if (content.includes('---\n') || content.includes('\n---\n')) {
+            return 'markdown';
+        }
+        return 'text';
+    }
+    /**
+     * Extract data using JSONPath
+     */
+    extractJSONPath(data, jsonPath) {
+        try {
+            const result = JSONPath({ path: jsonPath, json: data });
+            return result.length === 1 ? result[0] : result;
+        }
+        catch (error) {
+            throw new Error(`JSONPath extraction failed: ${error instanceof Error ? error.message : String(error)}`);
+        }
+    }
+    /**
+     * Pick specified keys
+     */
+    pickKeys(data, keys) {
+        if (typeof data !== 'object' || data === null) {
+            return data;
+        }
+        const result = Array.isArray(data) ? [] : {};
+        for (const key of keys) {
+            if (key.includes('.')) {
+                // Handle nested keys
+                this.setNestedValue(result, key, this.getNestedValue(data, key));
+            }
+            else if (key in data) {
+                result[key] = data[key];
+            }
+        }
+        return result;
+    }
+    /**
+     * Omit specified keys
+     */
+    omitKeys(data, keys) {
+        if (typeof data !== 'object' || data === null) {
+            return data;
+        }
+        const result = Array.isArray(data) ? [...data] : { ...data };
+        for (const key of keys) {
+            if (key.includes('.')) {
+                // Handle nested keys
+                this.deleteNestedValue(result, key);
+            }
+            else {
+                delete result[key];
+            }
+        }
+        return result;
+    }
+    /**
+     * Apply pipe transforms
+     */
+    async applyPipeTransforms(data, transforms, context) {
+        let result = data;
+        for (const transformSpec of transforms) {
+            try {
+                // Parse transform specification
+                // Format: "transform-name" or "transform-name(option1=value1,option2=value2)"
+                const { name, options } = this.parseTransformSpec(transformSpec);
+                logger.debug(`Applying transform: ${name}`, options);
+                // Execute transform
+                result = this.transformRegistry.execute(name, result, options);
+            }
+            catch (error) {
+                throw new Error(`Transform '${transformSpec}' failed: ${error instanceof Error ? error.message : String(error)}`);
+            }
+        }
+        return result;
+    }
+    /**
+     * Parse transform specification
+     * Examples: "trim", "number", "pick-keys(keys=[a,b,c])"
+     */
+    parseTransformSpec(spec) {
+        const match = spec.match(/^([a-z-]+)(?:\((.+)\))?$/);
+        if (!match) {
+            throw new Error(`Invalid transform specification: ${spec}`);
+        }
+        const [, name, optionsStr] = match;
+        if (!optionsStr) {
+            return { name };
+        }
+        // Parse options (simple key=value format)
+        const options = {};
+        const pairs = optionsStr.split(',').map(s => s.trim());
+        for (const pair of pairs) {
+            const [key, value] = pair.split('=').map(s => s.trim());
+            // Parse value type
+            if (value.startsWith('[') && value.endsWith(']')) {
+                // Array
+                options[key] = value.slice(1, -1).split(',').map(s => s.trim());
+            }
+            else if (value === 'true' || value === 'false') {
+                // Boolean
+                options[key] = value === 'true';
+            }
+            else if (!isNaN(Number(value))) {
+                // Number
+                options[key] = Number(value);
+            }
+            else {
+                // String
+                options[key] = value;
+            }
+        }
+        return { name, options };
+    }
+    /**
+     * Embed content under a key
+     */
+    embedContent(data, key) {
+        return { [key]: data };
+    }
+    /**
+     * Merge content with priority-based conflict resolution
+     */
+    mergeContent(source, target, strategy, context) {
+        const conflicts = [];
+        let merged;
+        switch (strategy) {
+            case 'replace':
+                merged = source;
+                break;
+            case 'shallow':
+                merged = { ...target, ...source };
+                break;
+            case 'deep':
+                merged = this.deepMerge(target, source, conflicts, context);
+                break;
+            case 'composite':
+                // Composite merge is handled earlier in the pipeline (Step 7)
+                // This case should not be reached
+                merged = source;
+                break;
+            default:
+                merged = source;
+        }
+        return { data: merged, conflicts };
+    }
+    /**
+     * Deep merge two objects
+     */
+    deepMerge(target, source, conflicts, context, keyPath = '') {
+        if (typeof source !== 'object' || source === null) {
+            return source;
+        }
+        if (typeof target !== 'object' || target === null) {
+            return source;
+        }
+        if (Array.isArray(source) && Array.isArray(target)) {
+            // Merge arrays
+            return [...target, ...source];
+        }
+        const result = { ...target };
+        for (const key of Object.keys(source)) {
+            const currentPath = keyPath ? `${keyPath}.${key}` : key;
+            if (!(key in target)) {
+                result[key] = source[key];
+            }
+            else if (typeof source[key] === 'object' && typeof target[key] === 'object') {
+                result[key] = this.deepMerge(target[key], source[key], conflicts, context, currentPath);
+            }
+            else if (source[key] !== target[key]) {
+                // Conflict detected
+                conflicts.push({
+                    path: currentPath,
+                    winner: context.packageName,
+                    losers: ['existing'],
+                    type: 'value',
+                    resolution: 'last-writer-wins',
+                });
+                result[key] = source[key];
+            }
+        }
+        return result;
+    }
+    /**
+     * Evaluate condition
+     */
+    /**
+     * Composite merge using comment delimiters to preserve multiple package contributions
+     * Each package's content is wrapped in HTML comment markers with package name
+     */
+    evaluateCondition(condition, context) {
+        if (condition.and) {
+            return condition.and.every((c) => this.evaluateCondition(c, context));
+        }
+        if (condition.or) {
+            return condition.or.some((c) => this.evaluateCondition(c, context));
+        }
+        if (condition.not) {
+            return !this.evaluateCondition(condition.not, context);
+        }
+        if (condition.exists) {
+            const testPath = path.join(context.workspaceRoot, condition.exists);
+            // Use existsSync for synchronous condition evaluation
+            return fsSync.existsSync(testPath);
+        }
+        if (condition.platform) {
+            return context.platform === condition.platform;
+        }
+        return true;
+    }
+    /**
+     * Resolve pattern with glob support and priority-based array matching
+     * Returns resolved file paths (glob patterns return multiple files)
+     */
+    async resolveSourcePattern(pattern, context) {
+        const result = await this.sourceResolver.resolve(pattern, {
+            baseDir: context.packageRoot,
+            logWarnings: true,
+        });
+        return {
+            paths: result.paths,
+            warnings: result.warnings,
+        };
+    }
+    /**
+     * Resolve target path from source path and patterns
+     * Handles single-level (*) and recursive (**) globs
+     */
+    resolveTargetFromGlob(sourcePath, fromPattern, toPattern, context) {
+        // Get relative path from package root
+        const relativePath = path.relative(context.packageRoot, sourcePath);
+        // If 'to' pattern has glob, map the structure
+        if (toPattern.includes('*')) {
+            // Handle ** recursive patterns
+            if (fromPattern.includes('**') && toPattern.includes('**')) {
+                // Extract the base directories before **
+                const fromParts = fromPattern.split('**');
+                const toParts = toPattern.split('**');
+                const fromBase = fromParts[0].replace(/\/$/, '');
+                const toBase = toParts[0].replace(/\/$/, '');
+                // Get the file pattern after **
+                const fromSuffix = fromParts[1] || '';
+                const toSuffix = toParts[1] || '';
+                // Extract the relative path after the base directory
+                let relativeSubpath = relativePath;
+                if (fromBase) {
+                    relativeSubpath = relativePath.startsWith(fromBase + '/')
+                        ? relativePath.slice(fromBase.length + 1)
+                        : relativePath;
+                }
+                // Handle extension mapping if suffixes specify extensions
+                // e.g., /**/*.md -> /**/*.mdc
+                if (fromSuffix && toSuffix) {
+                    const fromExt = fromSuffix.replace(/^\/?\*+/, '');
+                    const toExt = toSuffix.replace(/^\/?\*+/, '');
+                    if (fromExt && toExt && fromExt !== toExt) {
+                        relativeSubpath = relativeSubpath.replace(new RegExp(fromExt.replace('.', '\\.') + '$'), toExt);
+                    }
+                }
+                // Build target path
+                const targetPath = toBase ? path.join(toBase, relativeSubpath) : relativeSubpath;
+                return path.join(context.workspaceRoot, targetPath);
+            }
+            // Handle single-level * patterns
+            const sourceFileName = path.basename(sourcePath);
+            const sourceExt = path.extname(sourcePath);
+            const sourceBase = path.basename(sourcePath, sourceExt);
+            const toParts = toPattern.split('*');
+            const toPrefix = toParts[0];
+            const toSuffix = toParts[1] || '';
+            const targetExt = toSuffix.startsWith('.') ? toSuffix : (sourceExt + toSuffix);
+            const targetFileName = sourceBase + targetExt;
+            const resolvedTo = toPrefix + targetFileName;
+            return path.join(context.workspaceRoot, resolvedTo);
+        }
+        // No glob in target - use as-is
+        return path.join(context.workspaceRoot, toPattern);
+    }
+    /**
+     * Get pipeline description
+     */
+    getPipeline(flow) {
+        const pipeline = ['load'];
+        if (flow.path)
+            pipeline.push('extract');
+        if (flow.pick)
+            pipeline.push('pick');
+        if (flow.omit)
+            pipeline.push('omit');
+        if (flow.map)
+            pipeline.push('map');
+        if (flow.embed)
+            pipeline.push('embed');
+        if (flow.merge)
+            pipeline.push(`merge:${flow.merge}`);
+        pipeline.push('write');
+        return pipeline;
+    }
+    /**
+     * Normalize from pattern for display in results
+     * Converts array to comma-separated string
+     */
+    normalizeFromPattern(pattern) {
+        return Array.isArray(pattern) ? pattern.join(', ') : pattern;
+    }
+    /**
+     * Get the first pattern from a pattern or array of patterns
+     * Used for path resolution when multiple sources are specified
+     */
+    getFirstPattern(pattern) {
+        return Array.isArray(pattern) ? pattern[0] : pattern;
+    }
+    /**
+     * Aggregate multi-target results
+     */
+    aggregateResults(results, startTime) {
+        const successful = results.filter(r => r.success);
+        const failed = results.filter(r => !r.success);
+        if (failed.length > 0) {
+            return {
+                source: results[0]?.source || '',
+                target: results.map(r => r.target).join(', '),
+                success: false,
+                transformed: results.some(r => r.transformed),
+                error: failed[0]?.error,
+                warnings: results.flatMap(r => r.warnings || []),
+                executionTime: Date.now() - startTime,
+            };
+        }
+        return {
+            source: results[0]?.source || '',
+            target: results.flatMap(r => typeof r.target === 'string' ? [r.target] : r.target),
+            success: true,
+            transformed: results.some(r => r.transformed),
+            warnings: results.flatMap(r => r.warnings || []),
+            conflicts: results.flatMap(r => r.conflicts || []),
+            executionTime: Date.now() - startTime,
+        };
+    }
+    /**
+     * Parse markdown with frontmatter
+     */
+    parseMarkdown(content) {
+        const frontmatterRegex = /^---\n([\s\S]*?)\n---\n([\s\S]*)$/;
+        const match = content.match(frontmatterRegex);
+        if (match) {
+            const frontmatter = yaml.load(match[1]);
+            const body = match[2];
+            return { frontmatter, body };
+        }
+        return { body: content };
+    }
+    /**
+     * Serialize markdown with frontmatter
+     */
+    serializeMarkdown(content) {
+        if (typeof content === 'string') {
+            return content;
+        }
+        if (content.frontmatter) {
+            const frontmatterStr = yaml.dump(content.frontmatter, { indent: 2 });
+            return `---\n${frontmatterStr}---\n${content.body || ''}`;
+        }
+        return content.body || '';
+    }
+    /**
+     * Strip JSON comments
+     */
+    stripJSONComments(content) {
+        // Remove single-line comments
+        let result = content.replace(/\/\/.*$/gm, '');
+        // Remove multi-line comments
+        result = result.replace(/\/\*[\s\S]*?\*\//g, '');
+        return result;
+    }
+    /**
+     * Get nested value using dot notation (delegates to key mapper)
+     */
+    getNestedValue(obj, path) {
+        return getNestedValue(obj, path);
+    }
+    /**
+     * Set nested value using dot notation (delegates to key mapper)
+     */
+    setNestedValue(obj, path, value) {
+        setNestedValue(obj, path, value);
+    }
+    /**
+     * Delete nested value using dot notation (delegates to key mapper)
+     */
+    deleteNestedValue(obj, path) {
+        deleteNestedValue(obj, path);
+    }
+    /**
+     * Apply Codex-specific TOML formatting
+     * Converts nested table sections to inline format for http_headers and env_http_headers
+     */
+    applyCodexTomlFormatting(toml) {
+        const inlineKeys = ['http_headers', 'env_http_headers'];
+        let result = toml;
+        for (const key of inlineKeys) {
+            // Pattern to match nested table sections for the key
+            const pattern = new RegExp(`\\[([\\w-]+(?:\\.[\\w-]+|\\."[^"]+")*)?\\.${key}\\]\\s*\\n([\\s\\S]*?)(?=\\n\\[|\\n*$)`, 'g');
+            result = result.replace(pattern, (match, parentPath, content) => {
+                const pairs = [];
+                const lines = content.trim().split('\n');
+                for (const line of lines) {
+                    const trimmed = line.trim();
+                    if (!trimmed || trimmed.startsWith('#'))
+                        continue;
+                    const kvMatch = trimmed.match(/^([\w-]+)\s*=\s*(.+)$/);
+                    if (kvMatch) {
+                        const [, k, v] = kvMatch;
+                        pairs.push(`"${k}" = ${v}`);
+                    }
+                }
+                if (pairs.length === 0)
+                    return match;
+                const inlineTable = `{ ${pairs.join(', ')} }`;
+                return `${key} = ${inlineTable}`;
+            });
+        }
+        return result;
+    }
+}
+/**
+ * Create a flow executor instance
+ */
+export function createFlowExecutor() {
+    return new DefaultFlowExecutor();
+}
+//# sourceMappingURL=flow-executor.js.map