nfunc-mcp 0.3.0 → 0.5.0

package/dist/utils/urlInput.js

@@ -0,0 +1,130 @@
+/**
+ * One input, three shapes.
+ *
+ * People arrive with a URL, a list they pasted from a spreadsheet, or a CSV
+ * export. Making them say which is which is a tax on the common case, and an
+ * agent relaying a user's request should not have to guess a parameter name.
+ * So the `url` input accepts any of the three and this module works out what it
+ * was given.
+ *
+ * Detection is by shape, not by a flag, and deliberately conservative: anything
+ * ambiguous is reported rather than guessed at, because silently auditing the
+ * wrong set of pages is worse than an error.
+ */
+import { existsSync } from "node:fs";
+import { readUrlsFromCsv } from "./csvReader.js";
+const LOOKS_LIKE_URL = /^https?:\/\//i;
+function normalise(raw) {
+    const trimmed = raw.trim().replace(/^['"]|['"]$/g, "");
+    if (!trimmed)
+        return null;
+    // A bare domain is what people actually paste; assume https rather than
+    // rejecting it, and say so in a warning.
+    const candidate = LOOKS_LIKE_URL.test(trimmed) ? trimmed : `https://${trimmed}`;
+    try {
+        return new URL(candidate).toString();
+    }
+    catch {
+        return null;
+    }
+}
+function looksLikeCsvPath(value) {
+    const trimmed = value.trim();
+    if (LOOKS_LIKE_URL.test(trimmed))
+        return false;
+    if (trimmed.includes(","))
+        return false; // a list, not a path
+    return /\.(csv|tsv|txt)$/i.test(trimmed) || existsSync(trimmed);
+}
+/**
+ * Resolve whatever was handed in into a URL list.
+ *
+ * Order matters: the CSV check runs first because a path can contain no commas
+ * and no scheme, which would otherwise be coerced into `https://./file.csv`.
+ */
+export function resolveUrlInput(input) {
+    const warnings = [];
+    const raw = input.trim();
+    if (!raw)
+        throw new Error("No URL, list, or CSV path was provided.");
+    if (looksLikeCsvPath(raw)) {
+        if (!existsSync(raw)) {
+            throw new Error(`"${raw}" looks like a file path but does not exist. Pass a URL, a ` +
+                `comma-separated list of URLs, or the path to a readable CSV.`);
+        }
+        const csv = readUrlsFromCsv(raw);
+        warnings.push(...csv.warnings);
+        if (csv.urls.length === 0) {
+            throw new Error(`No usable URLs found in ${raw}.`);
+        }
+        return {
+            urls: csv.urls,
+            kind: "csv",
+            source: `${raw} (column "${csv.column}")`,
+            warnings,
+        };
+    }
+    // Split on commas and newlines — pasted lists arrive both ways, and a
+    // multi-line paste is the same intent as a comma-separated one.
+    const parts = raw
+        .split(/[,\n]/)
+        .map((p) => p.trim())
+        .filter(Boolean);
+    const urls = [];
+    const rejected = [];
+    const seen = new Set();
+    let coerced = 0;
+    for (const part of parts) {
+        const url = normalise(part);
+        if (!url) {
+            rejected.push(part);
+            continue;
+        }
+        if (!LOOKS_LIKE_URL.test(part.trim()))
+            coerced++;
+        if (seen.has(url))
+            continue;
+        seen.add(url);
+        urls.push(url);
+    }
+    if (urls.length === 0) {
+        throw new Error(`Could not read any URL from "${raw.slice(0, 120)}". Pass a URL, a ` +
+            `comma-separated list, or a path to a CSV.`);
+    }
+    if (rejected.length > 0) {
+        warnings.push(`Ignored ${rejected.length} unparseable entr${rejected.length === 1 ? "y" : "ies"}: ${rejected.slice(0, 3).join(", ")}`);
+    }
+    if (coerced > 0) {
+        warnings.push(`Assumed https:// for ${coerced} entr${coerced === 1 ? "y" : "ies"} that had no scheme.`);
+    }
+    if (parts.length > urls.length + rejected.length) {
+        warnings.push(`Removed ${parts.length - urls.length - rejected.length} duplicate URL(s).`);
+    }
+    return { urls, kind: urls.length === 1 ? "single" : "list", warnings };
+}
+/** Accept an explicit array too, for callers that already have one. */
+export function resolveUrlInputs(input, urls) {
+    if (urls && urls.length > 0) {
+        const resolved = [];
+        const seen = new Set();
+        const warnings = [];
+        for (const raw of urls) {
+            const url = normalise(raw);
+            if (!url) {
+                warnings.push(`Ignored unparseable URL "${raw}".`);
+                continue;
+            }
+            if (seen.has(url))
+                continue;
+            seen.add(url);
+            resolved.push(url);
+        }
+        if (resolved.length === 0)
+            throw new Error("None of the supplied urls were parseable.");
+        return { urls: resolved, kind: resolved.length === 1 ? "single" : "list", warnings };
+    }
+    if (!input)
+        throw new Error("Provide `url` (a URL, comma-separated list, or CSV path) or `urls`.");
+    return resolveUrlInput(input);
+}
+//# sourceMappingURL=urlInput.js.map

package/dist/utils/urlInput.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"urlInput.js","sourceRoot":"","sources":["../../src/utils/urlInput.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;GAYG;AAEH,OAAO,EAAE,UAAU,EAAE,MAAM,SAAS,CAAC;AACrC,OAAO,EAAE,eAAe,EAAE,MAAM,gBAAgB,CAAC;AAYjD,MAAM,cAAc,GAAG,eAAe,CAAC;AAEvC,SAAS,SAAS,CAAC,GAAW;IAC5B,MAAM,OAAO,GAAG,GAAG,CAAC,IAAI,EAAE,CAAC,OAAO,CAAC,cAAc,EAAE,EAAE,CAAC,CAAC;IACvD,IAAI,CAAC,OAAO;QAAE,OAAO,IAAI,CAAC;IAC1B,wEAAwE;IACxE,yCAAyC;IACzC,MAAM,SAAS,GAAG,cAAc,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,WAAW,OAAO,EAAE,CAAC;IAChF,IAAI,CAAC;QACH,OAAO,IAAI,GAAG,CAAC,SAAS,CAAC,CAAC,QAAQ,EAAE,CAAC;IACvC,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,IAAI,CAAC;IACd,CAAC;AACH,CAAC;AAED,SAAS,gBAAgB,CAAC,KAAa;IACrC,MAAM,OAAO,GAAG,KAAK,CAAC,IAAI,EAAE,CAAC;IAC7B,IAAI,cAAc,CAAC,IAAI,CAAC,OAAO,CAAC;QAAE,OAAO,KAAK,CAAC;IAC/C,IAAI,OAAO,CAAC,QAAQ,CAAC,GAAG,CAAC;QAAE,OAAO,KAAK,CAAC,CAAC,qBAAqB;IAC9D,OAAO,mBAAmB,CAAC,IAAI,CAAC,OAAO,CAAC,IAAI,UAAU,CAAC,OAAO,CAAC,CAAC;AAClE,CAAC;AAED;;;;;GAKG;AACH,MAAM,UAAU,eAAe,CAAC,KAAa;IAC3C,MAAM,QAAQ,GAAa,EAAE,CAAC;IAC9B,MAAM,GAAG,GAAG,KAAK,CAAC,IAAI,EAAE,CAAC;IAEzB,IAAI,CAAC,GAAG;QAAE,MAAM,IAAI,KAAK,CAAC,yCAAyC,CAAC,CAAC;IAErE,IAAI,gBAAgB,CAAC,GAAG,CAAC,EAAE,CAAC;QAC1B,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,EAAE,CAAC;YACrB,MAAM,IAAI,KAAK,CACb,IAAI,GAAG,6DAA6D;gBAClE,8DAA8D,CACjE,CAAC;QACJ,CAAC;QACD,MAAM,GAAG,GAAG,eAAe,CAAC,GAAG,CAAC,CAAC;QACjC,QAAQ,CAAC,IAAI,CAAC,GAAG,GAAG,CAAC,QAAQ,CAAC,CAAC;QAC/B,IAAI,GAAG,CAAC,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC1B,MAAM,IAAI,KAAK,CAAC,2BAA2B,GAAG,GAAG,CAAC,CAAC;QACrD,CAAC;QACD,OAAO;YACL,IAAI,EAAE,GAAG,CAAC,IAAI;YACd,IAAI,EAAE,KAAK;YACX,MAAM,EAAE,GAAG,GAAG,aAAa,GAAG,CAAC,MAAM,IAAI;YACzC,QAAQ;SACT,CAAC;IACJ,CAAC;IAED,sEAAsE;IACtE,gEAAgE;IAChE,MAAM,KAAK,GAAG,GAAG;SACd,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;SACpB,MAAM,CAAC,OAAO,CAAC,CAAC;IAEnB,MAAM,IAAI,GAAa,EAAE,CAAC;IAC1B,MAAM,QAAQ,GAAa,EAAE,CAAC;IAC9B,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,IAAI,OAAO,GAAG,CAAC,CAAC;IAEhB,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,MAAM,GAAG,GAAG,SAAS,CAAC,IAAI,CAAC,CAAC;QAC5B,IAAI,CAAC,GAAG,EAAE,CAAC;YACT,QAAQ,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;YACpB,SAAS;QACX,CAAC;QACD,IAAI,CAAC,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC;YAAE,OAAO,EAAE,CAAC;QACjD,IAAI,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC;YAAE,SAAS;QAC5B,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;QACd,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;IACjB,CAAC;IAED,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACtB,MAAM,IAAI,KAAK,CACb,gCAAgC,GAAG,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,mBAAmB;YAClE,2CAA2C,CAC9C,CAAC;IACJ,CAAC;IACD,IAAI,QAAQ,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACxB,QAAQ,CAAC,IAAI,CAAC,WAAW,QAAQ,CAAC,MAAM,oBAAoB,QAAQ,CAAC,MAAM,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,KAAK,QAAQ,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;IACzI,CAAC;IACD,IAAI,OAAO,GAAG,CAAC,EAAE,CAAC;QAChB,QAAQ,CAAC,IAAI,CAAC,wBAAwB,OAAO,QAAQ,OAAO,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,sBAAsB,CAAC,CAAC;IAC1G,CAAC;IACD,IAAI,KAAK,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC;QACjD,QAAQ,CAAC,IAAI,CAAC,WAAW,KAAK,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,GAAG,QAAQ,CAAC,MAAM,oBAAoB,CAAC,CAAC;IAC7F,CAAC;IAED,OAAO,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,CAAC,MAAM,KAAK,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,MAAM,EAAE,QAAQ,EAAE,CAAC;AACzE,CAAC;AAED,uEAAuE;AACvE,MAAM,UAAU,gBAAgB,CAC9B,KAAyB,EACzB,IAA0B;IAE1B,IAAI,IAAI,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QAC5B,MAAM,QAAQ,GAAa,EAAE,CAAC;QAC9B,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;QAC/B,MAAM,QAAQ,GAAa,EAAE,CAAC;QAC9B,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;YACvB,MAAM,GAAG,GAAG,SAAS,CAAC,GAAG,CAAC,CAAC;YAC3B,IAAI,CAAC,GAAG,EAAE,CAAC;gBACT,QAAQ,CAAC,IAAI,CAAC,4BAA4B,GAAG,IAAI,CAAC,CAAC;gBACnD,SAAS;YACX,CAAC;YACD,IAAI,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC;gBAAE,SAAS;YAC5B,IAAI,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;YACd,QAAQ,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;QACrB,CAAC;QACD,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC;YAAE,MAAM,IAAI,KAAK,CAAC,2CAA2C,CAAC,CAAC;QACxF,OAAO,EAAE,IAAI,EAAE,QAAQ,EAAE,IAAI,EAAE,QAAQ,CAAC,MAAM,KAAK,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,MAAM,EAAE,QAAQ,EAAE,CAAC;IACvF,CAAC;IACD,IAAI,CAAC,KAAK;QAAE,MAAM,IAAI,KAAK,CAAC,qEAAqE,CAAC,CAAC;IACnG,OAAO,eAAe,CAAC,KAAK,CAAC,CAAC;AAChC,CAAC"}