npm - @j0hanz/fetch-url-mcp - Versions diffs - 1.9.3 → 1.9.4 - Mend

@j0hanz/fetch-url-mcp 1.9.3 → 1.9.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (19) hide show

package/dist/lib/code-lang.d.ts +4 -0
package/dist/lib/code-lang.d.ts.map +1 -0
package/dist/lib/code-lang.js +315 -0
package/dist/lib/dom-prep.d.ts +4 -0
package/dist/lib/dom-prep.d.ts.map +1 -0
package/dist/lib/dom-prep.js +606 -0
package/dist/lib/md-cleanup.d.ts +13 -0
package/dist/lib/md-cleanup.d.ts.map +1 -0
package/dist/lib/md-cleanup.js +391 -0
package/dist/lib/md-metadata.d.ts +6 -0
package/dist/lib/md-metadata.d.ts.map +1 -0
package/dist/lib/md-metadata.js +186 -0
package/dist/transform/html-translators.js +1 -1
package/dist/transform/transform.d.ts.map +1 -1
package/dist/transform/transform.js +160 -176
package/package.json +1 -1
package/dist/lib/content.d.ts +0 -18
package/dist/lib/content.d.ts.map +0 -1
package/dist/lib/content.js +0 -1433

package/dist/transform/transform.js CHANGED Viewed

@@ -3,11 +3,14 @@ import diagnosticsChannel from 'node:diagnostics_channel';
 import { performance } from 'node:perf_hooks';
 import { isProbablyReaderable, Readability } from '@mozilla/readability';
 import { parseHTML } from 'linkedom';
-import { addSourceToMarkdown, buildMetadataFooter, cleanupMarkdownArtifacts, detectLanguageFromCode, extractLanguageFromClassName, extractTitleFromRawMarkdown, isRawTextContent, prepareDocumentForMarkdown, removeNoiseFromHtml, serializeDocumentForMarkdown, } from '../lib/content.js';
+import { detectLanguageFromCode, extractLanguageFromClassName, } from '../lib/code-lang.js';
 import { config } from '../lib/core.js';
 import { getOperationId, getRequestId, logDebug, logError, logInfo, logWarn, redactUrl, } from '../lib/core.js';
+import { prepareDocumentForMarkdown, removeNoiseFromHtml, serializeDocumentForMarkdown, } from '../lib/dom-prep.js';
 import { isRawTextContentUrl } from '../lib/http.js';
-import { createAbortError, throwIfAborted } from '../lib/utils.js';
+import { cleanupMarkdownArtifacts, processFencedContent, } from '../lib/md-cleanup.js';
+import { addSourceToMarkdown, buildMetadataFooter, extractTitleFromRawMarkdown, isRawTextContent, } from '../lib/md-metadata.js';
+import { throwIfAborted } from '../lib/utils.js';
 import { FetchError, getErrorMessage, toError } from '../lib/utils.js';
 import { isObject } from '../lib/utils.js';
 import { translateHtmlFragmentToMarkdown } from './html-translators.js';
@@ -34,7 +37,6 @@ function decodeInput(input, encoding) {
 function asError(value) {
     return value instanceof Error ? value : undefined;
 }
-const abortPolicy = { throwIfAborted, createAbortError };
 function isWhitespaceChar(code) {
     return code === 9 || code === 10 || code === 12 || code === 13 || code === 32;
 }
@@ -306,7 +308,7 @@ function extractArticle(document, url, signal) {
         return null;
     }
     const checkAbort = (stage) => {
-        abortPolicy.throwIfAborted(signal, url, stage);
+        throwIfAborted(signal, url, stage);
     };
     try {
         const doc = document;
@@ -391,29 +393,43 @@ function applyBaseUri(document, url) {
         });
     }
 }
+function createEmptyExtractionContext() {
+    const { document } = parseHTML('<html></html>');
+    return { article: null, metadata: {}, document };
+}
+function extractEarlyMetadataIfNeeded(html, url) {
+    if (!willTruncate(html))
+        return null;
+    return stageTracker.run(url, 'extract:early-metadata', () => extractMetadataFromHead(html, url));
+}
+function parseExtractionDocument(html, url, inputTruncated) {
+    const { html: limitedHtml, truncated } = truncateHtml(html, inputTruncated);
+    const { document } = stageTracker.run(url, 'extract:parse', () => parseHTML(limitedHtml));
+    return { document, truncated };
+}
+function extractMergedMetadata(html, url, document) {
+    const earlyMetadata = extractEarlyMetadataIfNeeded(html, url);
+    const lateMetadata = stageTracker.run(url, 'extract:metadata', () => extractMetadata(document, url));
+    return mergeMetadata(earlyMetadata, lateMetadata);
+}
+function extractArticleIfRequested(document, url, options) {
+    if (!options.extractArticle)
+        return null;
+    return stageTracker.run(url, 'extract:article', () => extractArticle(document, url, options.signal));
+}
 function extractContentContext(html, url, options) {
     if (!isValidInput(html, url)) {
-        const { document } = parseHTML('<html></html>');
-        return { article: null, metadata: {}, document };
+        return createEmptyExtractionContext();
     }
     try {
-        abortPolicy.throwIfAborted(options.signal, url, 'extract:begin');
-        // F2: Extract metadata from <head> BEFORE truncation to preserve it
-        const earlyMetadata = willTruncate(html)
-            ? stageTracker.run(url, 'extract:early-metadata', () => extractMetadataFromHead(html, url))
-            : null;
-        const { html: limitedHtml, truncated } = truncateHtml(html, options.inputTruncated);
-        const { document } = stageTracker.run(url, 'extract:parse', () => parseHTML(limitedHtml));
-        abortPolicy.throwIfAborted(options.signal, url, 'extract:parsed');
+        throwIfAborted(options.signal, url, 'extract:begin');
+        const { document, truncated } = parseExtractionDocument(html, url, options.inputTruncated);
+        throwIfAborted(options.signal, url, 'extract:parsed');
         applyBaseUri(document, url);
-        const lateMetadata = stageTracker.run(url, 'extract:metadata', () => extractMetadata(document, url));
-        abortPolicy.throwIfAborted(options.signal, url, 'extract:metadata');
-        // Merge early (pre-truncation) with late (post-truncation) metadata
-        const metadata = mergeMetadata(earlyMetadata, lateMetadata);
-        const article = options.extractArticle
-            ? stageTracker.run(url, 'extract:article', () => extractArticle(document, url, options.signal))
-            : null;
-        abortPolicy.throwIfAborted(options.signal, url, 'extract:article');
+        const metadata = extractMergedMetadata(html, url, document);
+        throwIfAborted(options.signal, url, 'extract:metadata');
+        const article = extractArticleIfRequested(document, url, options);
+        throwIfAborted(options.signal, url, 'extract:article');
         return {
             article,
             metadata,
@@ -424,10 +440,9 @@ function extractContentContext(html, url, options) {
     catch (error) {
         if (error instanceof FetchError)
             throw error;
-        abortPolicy.throwIfAborted(options.signal, url, 'extract:error');
+        throwIfAborted(options.signal, url, 'extract:error');
         logError('Failed to extract content', asError(error));
-        const { document } = parseHTML('<html></html>');
-        return { article: null, metadata: {}, document };
+        return createEmptyExtractionContext();
     }
 }
 export function extractContent(html, url, options = {
@@ -436,8 +451,6 @@ export function extractContent(html, url, options = {
     const result = extractContentContext(html, url, options);
     return { article: result.article, metadata: result.metadata };
 }
-const ABORT_CHECK_LINE_INTERVAL = 500;
-const CR_CHAR_CODE = 13;
 function resolveRelativeHref(href, baseUrl, origin) {
     const trimmedHref = href.trim();
     if (!trimmedHref || containsWhitespace(trimmedHref))
@@ -501,7 +514,6 @@ function isAbsoluteOrSpecialUrl(href) {
         return true;
     return URL.canParse(trimmedHref);
 }
-const FENCE_LINE_PATTERN = /^\s*(`{3,}|~{3,})/;
 function resolveRelativeUrlsInSegment(markdown, baseUrl, origin) {
     let cursor = 0;
     let output = '';
@@ -527,71 +539,20 @@ function resolveRelativeUrls(markdown, baseUrl, signal) {
     }
     if (!markdown)
         return markdown;
-    let output = '';
-    let buffer = '';
-    let fenceMarker = null;
-    const len = markdown.length;
-    let lastIndex = 0;
-    let lineCount = 0;
-    while (lastIndex < len) {
-        if (++lineCount % ABORT_CHECK_LINE_INTERVAL === 0) {
-            abortPolicy.throwIfAborted(signal, baseUrl, 'markdown:resolve-urls');
-        }
-        // Extract next line (handling CR+LF)
-        let nextIndex = markdown.indexOf('\n', lastIndex);
-        const isLastLine = nextIndex === -1;
-        if (isLastLine)
-            nextIndex = len;
-        const lineWithNewline = isLastLine
-            ? markdown.slice(lastIndex)
-            : markdown.slice(lastIndex, nextIndex + 1);
-        const lineEnd = !isLastLine &&
-            nextIndex > lastIndex &&
-            markdown.charCodeAt(nextIndex - 1) === CR_CHAR_CODE
-            ? nextIndex - 1
-            : isLastLine
-                ? len
-                : nextIndex;
-        const trimmed = markdown.slice(lastIndex, lineEnd).trimStart();
-        if (fenceMarker) {
-            // Inside a code fence — pass through without URL resolution
-            output += lineWithNewline;
-            if (trimmed.startsWith(fenceMarker) &&
-                trimmed.slice(fenceMarker.length).trim() === '') {
-                fenceMarker = null;
-            }
-        }
-        else {
-            const fenceMatch = FENCE_LINE_PATTERN.exec(markdown.slice(lastIndex, lineEnd));
-            if (fenceMatch?.[1]) {
-                // Entering a code fence — flush buffered content first
-                if (buffer) {
-                    output += resolveRelativeUrlsInSegment(buffer, baseUrl, origin);
-                    buffer = '';
-                }
-                output += lineWithNewline;
-                fenceMarker = fenceMatch[1];
-            }
-            else {
-                buffer += lineWithNewline;
-            }
-        }
-        lastIndex = isLastLine ? len : nextIndex + 1;
-    }
-    if (buffer) {
-        output += resolveRelativeUrlsInSegment(buffer, baseUrl, origin);
-    }
-    return output;
+    return processFencedContent(markdown, (text) => {
+        throwIfAborted(signal, baseUrl, 'markdown:resolve-urls');
+        return resolveRelativeUrlsInSegment(text, baseUrl, origin);
+    });
 }
 function translateHtmlToMarkdown(params) {
     const { html, url, signal, document, skipNoiseRemoval } = params;
-    abortPolicy.throwIfAborted(signal, url, 'markdown:begin');
+    throwIfAborted(signal, url, 'markdown:begin');
     const cleanedHtml = skipNoiseRemoval
         ? html
         : stageTracker.run(url, 'markdown:noise', () => removeNoiseFromHtml(html, document, url, signal));
-    abortPolicy.throwIfAborted(signal, url, 'markdown:cleaned');
+    throwIfAborted(signal, url, 'markdown:cleaned');
     const content = stageTracker.run(url, 'markdown:translate', () => translateHtmlFragmentToMarkdown(cleanedHtml));
-    abortPolicy.throwIfAborted(signal, url, 'markdown:translated');
+    throwIfAborted(signal, url, 'markdown:translated');
     const cleaned = cleanupMarkdownArtifacts(content, signal ? { signal, url } : { url });
     return url ? resolveRelativeUrls(cleaned, url, signal) : cleaned;
 }
@@ -937,57 +898,74 @@ export const TransformHeuristics = {
     findPrimaryHeading,
     isGithubRepositoryRootUrl,
 };
-function shouldUseArticleContent(article, document) {
-    const articleLength = article.textContent.length;
+const ARTICLE_INTERACTIVE_SELECTOR = 'button,[role="tab"],[role="tabpanel"],[aria-controls]';
+function buildArticleDocument(article) {
+    return parseHTML(`<!DOCTYPE html><html><body>${article.content}</body></html>`).document;
+}
+function hasSufficientArticleContentRatio(article, document) {
     const originalLength = getVisibleTextLength(document);
-    if (originalLength >= MIN_HTML_LENGTH_FOR_GATE) {
-        const ratio = articleLength / originalLength;
-        if (ratio < MIN_CONTENT_RATIO)
-            return false;
-    }
-    const { document: articleDoc } = parseHTML(`<!DOCTYPE html><html><body>${article.content}</body></html>`);
+    if (originalLength < MIN_HTML_LENGTH_FOR_GATE)
+        return true;
+    return article.textContent.length / originalLength >= MIN_CONTENT_RATIO;
+}
+function retainsEnoughHeadings(articleDoc, document) {
     const originalHeadings = countMatchingElements(document, 'h1,h2,h3,h4,h5,h6');
+    if (originalHeadings === 0)
+        return true;
     const articleHeadings = countMatchingElements(articleDoc, 'h1,h2,h3,h4,h5,h6');
-    if (originalHeadings > 0) {
-        const retentionRatio = articleHeadings / originalHeadings;
-        if (retentionRatio < MIN_HEADING_RETENTION_RATIO)
-            return false;
-    }
+    return articleHeadings / originalHeadings >= MIN_HEADING_RETENTION_RATIO;
+}
+function retainsEnoughCodeBlocks(articleDoc, document) {
     const originalCodeBlocks = countMatchingElements(document, 'pre');
-    if (originalCodeBlocks > 0) {
-        const articleCodeBlocks = countMatchingElements(articleDoc, 'pre');
-        const codeRetentionRatio = articleCodeBlocks / originalCodeBlocks;
-        if (codeRetentionRatio < MIN_CODE_BLOCK_RETENTION_RATIO)
-            return false;
-    }
+    if (originalCodeBlocks === 0)
+        return true;
+    const articleCodeBlocks = countMatchingElements(articleDoc, 'pre');
+    return (articleCodeBlocks / originalCodeBlocks >= MIN_CODE_BLOCK_RETENTION_RATIO);
+}
+function retainsEnoughTables(articleDoc, document) {
     const originalTables = countMatchingElements(document, 'table');
-    if (originalTables > 0) {
-        const articleTables = countMatchingElements(articleDoc, 'table');
-        const tableRetentionRatio = articleTables / originalTables;
-        if (tableRetentionRatio < MIN_TABLE_RETENTION_RATIO)
-            return false;
-    }
+    if (originalTables === 0)
+        return true;
+    const articleTables = countMatchingElements(articleDoc, 'table');
+    return articleTables / originalTables >= MIN_TABLE_RETENTION_RATIO;
+}
+function retainsEnoughImages(articleDoc, document) {
     const originalImages = countMatchingElements(document, 'img');
-    if (originalImages >= MIN_IMAGE_ELEMENTS_FOR_GATE) {
-        const articleImages = countMatchingElements(articleDoc, 'img');
-        const imageRetentionRatio = articleImages / originalImages;
-        if (imageRetentionRatio < MIN_IMAGE_RETENTION_RATIO)
-            return false;
-    }
-    const interactiveSelector = 'button,[role="tab"],[role="tabpanel"],[aria-controls]';
-    const originalInteractive = countMatchingElements(document, interactiveSelector);
-    if (originalInteractive >= MIN_INTERACTIVE_ELEMENTS_FOR_GATE) {
-        const articleInteractive = countMatchingElements(articleDoc, interactiveSelector);
-        const interactiveRetentionRatio = articleInteractive / originalInteractive;
-        if (interactiveRetentionRatio < MIN_INTERACTIVE_RETENTION_RATIO) {
-            return false;
-        }
-    }
-    if (articleHeadings >= MIN_HEADINGS_FOR_EMPTY_SECTION_GATE) {
-        const emptySectionRatio = countEmptyHeadingSections(articleDoc) / articleHeadings;
-        if (emptySectionRatio > MAX_EMPTY_SECTION_RATIO)
-            return false;
-    }
+    if (originalImages < MIN_IMAGE_ELEMENTS_FOR_GATE)
+        return true;
+    const articleImages = countMatchingElements(articleDoc, 'img');
+    return articleImages / originalImages >= MIN_IMAGE_RETENTION_RATIO;
+}
+function retainsEnoughInteractiveElements(articleDoc, document) {
+    const originalInteractive = countMatchingElements(document, ARTICLE_INTERACTIVE_SELECTOR);
+    if (originalInteractive < MIN_INTERACTIVE_ELEMENTS_FOR_GATE)
+        return true;
+    const articleInteractive = countMatchingElements(articleDoc, ARTICLE_INTERACTIVE_SELECTOR);
+    return (articleInteractive / originalInteractive >= MIN_INTERACTIVE_RETENTION_RATIO);
+}
+function hasAcceptableEmptySectionRatio(articleDoc) {
+    const articleHeadings = countMatchingElements(articleDoc, 'h1,h2,h3,h4,h5,h6');
+    if (articleHeadings < MIN_HEADINGS_FOR_EMPTY_SECTION_GATE)
+        return true;
+    const emptySectionRatio = countEmptyHeadingSections(articleDoc) / articleHeadings;
+    return emptySectionRatio <= MAX_EMPTY_SECTION_RATIO;
+}
+function shouldUseArticleContent(article, document) {
+    if (!hasSufficientArticleContentRatio(article, document))
+        return false;
+    const articleDoc = buildArticleDocument(article);
+    if (!retainsEnoughHeadings(articleDoc, document))
+        return false;
+    if (!retainsEnoughCodeBlocks(articleDoc, document))
+        return false;
+    if (!retainsEnoughTables(articleDoc, document))
+        return false;
+    if (!retainsEnoughImages(articleDoc, document))
+        return false;
+    if (!retainsEnoughInteractiveElements(articleDoc, document))
+        return false;
+    if (!hasAcceptableEmptySectionRatio(articleDoc))
+        return false;
     return !hasTruncatedSentences(article.textContent);
 }
 function buildContentSource(params) {
@@ -1339,6 +1317,33 @@ function resolveContentSource(params) {
         ...(params.signal ? { signal: params.signal } : {}),
     });
 }
+function shouldStripGithubPrimaryHeading(context, url) {
+    return (context.primaryHeading !== undefined &&
+        TransformHeuristics.isGithubRepositoryRootUrl(url));
+}
+function maybeStripGithubPrimaryHeading(markdown, context, url) {
+    if (!shouldStripGithubPrimaryHeading(context, url))
+        return markdown;
+    return stripLeadingHeading(markdown, context.primaryHeading ?? '');
+}
+function buildSyntheticTitlePrefix(url, favicon) {
+    if (!favicon)
+        return ' ';
+    let alt = '';
+    try {
+        alt = new URL(url).hostname;
+    }
+    catch {
+        /* skip */
+    }
+    return ` ![${alt}](${favicon}) `;
+}
+function maybePrependSyntheticTitle(markdown, context, url) {
+    if (!context.title || /^(#{1,6})\s/.test(markdown.trimStart())) {
+        return markdown;
+    }
+    return `#${buildSyntheticTitlePrefix(url, context.favicon)}${context.title}\n\n${markdown}`;
+}
 function buildMarkdownFromContext(context, url, signal) {
     let content = stageTracker.run(url, 'transform:markdown', () => htmlToMarkdown(context.sourceHtml, context.metadata, {
         url,
@@ -1346,25 +1351,8 @@ function buildMarkdownFromContext(context, url, signal) {
         ...(context.document ? { document: context.document } : {}),
         ...(context.skipNoiseRemoval ? { skipNoiseRemoval: true } : {}),
     }));
-    if (context.primaryHeading &&
-        TransformHeuristics.isGithubRepositoryRootUrl(url)) {
-        content = stripLeadingHeading(content, context.primaryHeading);
-    }
-    if (context.title && !/^(#{1,6})\s/.test(content.trimStart())) {
-        const icon = context.favicon;
-        let prefix = ' ';
-        if (icon) {
-            let alt = '';
-            try {
-                alt = new URL(url).hostname;
-            }
-            catch {
-                /* skip */
-            }
-            prefix = ` ![${alt}](${icon}) `;
-        }
-        content = `#${prefix}${context.title}\n\n${content}`;
-    }
+    content = maybeStripGithubPrimaryHeading(content, context, url);
+    content = maybePrependSyntheticTitle(content, context, url);
     content = supplementMarkdownFromNextFlight(content, context.originalHtml);
     content = cleanupMarkdownArtifacts(content, signal ? { signal, url } : { url });
     return {
@@ -1422,10 +1410,24 @@ export function transformHtmlToMarkdownInProcess(html, url, options) {
     const signal = buildTransformSignal(options.signal);
     const totalStage = stageTracker.start(url, 'transform:total');
     try {
-        abortPolicy.throwIfAborted(signal, url, 'transform:begin');
+        throwIfAborted(signal, url, 'transform:begin');
         validateBinaryContent(html, url);
-        const result = tryRawContentPipeline(html, url, options) ??
-            tryHtmlContentPipeline(html, url, options, signal);
+        const result = stageTracker.run(url, 'transform:raw', () => tryTransformRawContent({
+            html,
+            url,
+            includeMetadata: options.includeMetadata,
+            ...(options.inputTruncated ? { inputTruncated: true } : {}),
+        })) ??
+            (() => {
+                const context = stageTracker.run(url, 'transform:extract', () => resolveContentSource({
+                    html,
+                    url,
+                    includeMetadata: options.includeMetadata,
+                    ...(signal ? { signal } : {}),
+                    ...(options.inputTruncated ? { inputTruncated: true } : {}),
+                }));
+                return buildMarkdownFromContext(context, url, signal);
+            })();
         stageTracker.end(totalStage, { truncated: result.truncated });
         return result;
     }
@@ -1439,24 +1441,6 @@ function validateBinaryContent(html, url) {
         throw new FetchError('Content appears to be binary data (high replacement character ratio or null bytes)', url, 415, { reason: 'binary_content_detected', stage: 'transform:validate' });
     }
 }
-function tryRawContentPipeline(html, url, options) {
-    return stageTracker.run(url, 'transform:raw', () => tryTransformRawContent({
-        html,
-        url,
-        includeMetadata: options.includeMetadata,
-        ...(options.inputTruncated ? { inputTruncated: true } : {}),
-    }));
-}
-function tryHtmlContentPipeline(html, url, options, signal) {
-    const context = stageTracker.run(url, 'transform:extract', () => resolveContentSource({
-        html,
-        url,
-        includeMetadata: options.includeMetadata,
-        ...(signal ? { signal } : {}),
-        ...(options.inputTruncated ? { inputTruncated: true } : {}),
-    }));
-    return buildMarkdownFromContext(context, url, signal);
-}
 export function getTransformPoolStats() {
     return getWorkerPoolStats();
 }
@@ -1466,7 +1450,7 @@ export async function shutdownTransformWorkerPool() {
 function transformInputInProcess(htmlOrBuffer, url, options) {
     return transformHtmlToMarkdownInProcess(decodeInput(htmlOrBuffer, options.encoding), url, options);
 }
-function buildWorkerTransformOptions(options) {
+function workerTransformOptions(options) {
     return {
         includeMetadata: options.includeMetadata,
         ...(options.signal ? { signal: options.signal } : {}),
@@ -1479,10 +1463,10 @@ async function transformWithWorkerPool(htmlOrBuffer, url, options) {
         return transformInputInProcess(htmlOrBuffer, url, options);
     }
     if (typeof htmlOrBuffer === 'string') {
-        return pool.transform(htmlOrBuffer, url, buildWorkerTransformOptions(options));
+        return pool.transform(htmlOrBuffer, url, workerTransformOptions(options));
     }
     return pool.transform(htmlOrBuffer, url, {
-        ...buildWorkerTransformOptions(options),
+        ...workerTransformOptions(options),
         ...(options.encoding ? { encoding: options.encoding } : {}),
     });
 }
@@ -1494,7 +1478,7 @@ function resolveWorkerFallback(error, htmlOrBuffer, url, options) {
         });
         return transformInputInProcess(htmlOrBuffer, url, options);
     }
-    abortPolicy.throwIfAborted(options.signal, url, 'transform:worker-fallback');
+    throwIfAborted(options.signal, url, 'transform:worker-fallback');
     if (error instanceof FetchError)
         throw error;
     if (!(error instanceof Error))
@@ -1521,7 +1505,7 @@ async function runWorkerTransformWithFallback(htmlOrBuffer, url, options) {
 async function transformInputToMarkdown(htmlOrBuffer, url, options) {
     const totalStage = stageTracker.start(url, 'transform:total');
     try {
-        abortPolicy.throwIfAborted(options.signal, url, 'transform:begin');
+        throwIfAborted(options.signal, url, 'transform:begin');
         const result = await runWorkerTransformWithFallback(htmlOrBuffer, url, options);
         stageTracker.end(totalStage, { truncated: result.truncated });
         return result;

package/package.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "name": "@j0hanz/fetch-url-mcp",
-  "version": "1.9.3",
+  "version": "1.9.4",
   "mcpName": "io.github.j0hanz/fetch-url-mcp",
   "description": "A web content fetcher MCP server that converts HTML to clean, AI and human readable markdown.",
   "type": "module",

package/dist/lib/content.d.ts DELETED Viewed

@@ -1,18 +0,0 @@
-import { type MetadataBlock } from '../transform/types.js';
-export declare function serializeDocumentForMarkdown(document: Document, fallback: string): string;
-export declare function prepareDocumentForMarkdown(document: Document, baseUrl?: string, signal?: AbortSignal): void;
-export declare function removeNoiseFromHtml(html: string, document?: Document, baseUrl?: string, signal?: AbortSignal): string;
-export declare function extractLanguageFromClassName(className: string): string | undefined;
-export declare function resolveLanguageFromAttributes(className: string, dataLang: string): string | undefined;
-export declare function detectLanguageFromCode(code: string): string | undefined;
-interface CleanupOptions {
-    signal?: AbortSignal;
-    url?: string;
-}
-export declare function cleanupMarkdownArtifacts(content: string, options?: CleanupOptions): string;
-export declare function extractTitleFromRawMarkdown(content: string): string | undefined;
-export declare function addSourceToMarkdown(content: string, url: string): string;
-export declare function isRawTextContent(content: string): boolean;
-export declare function buildMetadataFooter(metadata?: MetadataBlock, fallbackUrl?: string): string;
-export {};
-//# sourceMappingURL=content.d.ts.map

package/dist/lib/content.d.ts.map DELETED Viewed

@@ -1 +0,0 @@

- {"version":3,"file":"content.d.ts","sourceRoot":"","sources":["../../src/lib/content.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,KAAK,aAAa,EAAE,MAAM,uBAAuB,CAAC;AAwjB3D,wBAAgB,4BAA4B,CAC1C,QAAQ,EAAE,QAAQ,EAClB,QAAQ,EAAE,MAAM,GACf,MAAM,CAQR;AA2DD,wBAAgB,0BAA0B,CACxC,QAAQ,EAAE,QAAQ,EAClB,OAAO,CAAC,EAAE,MAAM,EAChB,MAAM,CAAC,EAAE,WAAW,GACnB,IAAI,CAiBN;AA0BD,wBAAgB,mBAAmB,CACjC,IAAI,EAAE,MAAM,EACZ,QAAQ,CAAC,EAAE,QAAQ,EACnB,OAAO,CAAC,EAAE,MAAM,EAChB,MAAM,CAAC,EAAE,WAAW,GACnB,MAAM,CAcR;AAwQD,wBAAgB,4BAA4B,CAC1C,SAAS,EAAE,MAAM,GAChB,MAAM,GAAG,SAAS,CAuBpB;AAqBD,wBAAgB,6BAA6B,CAC3C,SAAS,EAAE,MAAM,EACjB,QAAQ,EAAE,MAAM,GACf,MAAM,GAAG,SAAS,CAKpB;AACD,wBAAgB,sBAAsB,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,GAAG,SAAS,CAqBvE;AAsDD,UAAU,cAAc;IACtB,MAAM,CAAC,EAAE,WAAW,CAAC;IACrB,GAAG,CAAC,EAAE,MAAM,CAAC;CACd;AAqTD,wBAAgB,wBAAwB,CACtC,OAAO,EAAE,MAAM,EACf,OAAO,CAAC,EAAE,cAAc,GACvB,MAAM,CA6CR;AAgGD,wBAAgB,2BAA2B,CACzC,OAAO,EAAE,MAAM,GACd,MAAM,GAAG,SAAS,CAOpB;AACD,wBAAgB,mBAAmB,CAAC,OAAO,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,GAAG,MAAM,CAuCxE;AAmBD,wBAAgB,gBAAgB,CAAC,OAAO,EAAE,MAAM,GAAG,OAAO,CAczD;AAaD,wBAAgB,mBAAmB,CACjC,QAAQ,CAAC,EAAE,aAAa,EACxB,WAAW,CAAC,EAAE,MAAM,GACnB,MAAM,CAmBR"}