@pipeworx/mcp-guam-code 0.1.2 → 0.1.4

package/README.md

@@ -3,7 +3,7 @@
 Guam Code Annotated (GCA) — Guam territorial statutes, by citation and by
 topic/keyword search. Keyless.
 
-Part of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1815+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.
+Part of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1879+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.
 
 ## Tools
 
@@ -224,7 +224,7 @@ directly, instead of just this one's:
 }
 ```
 
-Both URLs reach the same gateway and the same 1815+ data sources. The
+Both URLs reach the same gateway and the same 1879+ data sources. The
 only difference is which pack's tools are listed **directly**; `ask_pipeworx`
 reaches all of them from either one.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pipeworx/mcp-guam-code",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "description": "Guam Code Annotated (GCA) — section lookup and search. Fleet #2888.",
   "type": "module",
   "main": "src/index.ts",
@@ -26,7 +26,7 @@
     "@cloudflare/workers-types": "^4.20260405.1"
   },
   "pipeworx": {
-    "sourceHash": "v1-0e8b06ad75d31b79367ea7bccec933bfae1665a234e87b2fd47a9220c214f878",
-    "sourceCommit": "fe38188882d0fe91e5a45d82b7b989b4aae1e8b6"
+    "sourceHash": "v1-def998606f47935be067505462c08c2cb618014e238faf74206f49ad1d914e33",
+    "sourceCommit": "445bacc5596a845c12d6945fbfcdf330ef60dcc4"
   }
 }

package/server.json

@@ -3,7 +3,7 @@
   "name": "io.github.pipeworx-io/guam-code",
   "title": "Guam Code",
   "description": "Guam Code Annotated (GCA) — section lookup and search. Fleet #2888.",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "websiteUrl": "https://pipeworx.io/packs/guam-code",
   "repository": {
     "url": "https://github.com/pipeworx-io/mcp-guam-code",

package/src/index.ts

@@ -295,12 +295,28 @@ async function httpErrorMessage(res: Response, name: string): Promise<string> {
   // internal-host-class.ts; no-op for every third-party upstream, which is why
   // this touches 481 packs' error text and changes none of it.
   return markInternalOrigin(
-    `${name}: ${res.status}${detailSuffix(await readDetail(res))}`,
+    `${name}: ${res.status}${detailSuffix(await readDetail(res))}${responseHints(res)}`,
     res.url,
     res.status,
   );
 }
 
+/**
+ * Facts about a FAILED response that live in its headers, not its body, appended
+ * so the gateway's classifier (which only ever sees this string) can tell a limit
+ * from a wall from a refusal (fleet #3477). Both are standard, both are absent on
+ * almost every response, and neither is invented: `Retry-After` is the server
+ * saying "come back later" (the one thing that makes a 403 a throttle), and
+ * Cloudflare marks a bot challenge with `cf-mitigated: challenge`.
+ */
+function responseHints(res: Response): string {
+  const hints: string[] = [];
+  const retryAfter = res.headers.get('retry-after');
+  if (retryAfter) hints.push(`retry-after: ${retryAfter.slice(0, 40)}`);
+  if ((res.headers.get('cf-mitigated') ?? '').toLowerCase() === 'challenge') hints.push('Cloudflare bot challenge (cf-mitigated: challenge)');
+  return hints.length ? ` (${hints.join('; ')})` : '';
+}
+
 /**
  * Just the upstream's own explanation — no name, no status.
  *
@@ -671,6 +687,254 @@ function collapse(s: string): string {
   return s.replace(/\s+/g, ' ').trim();
 }
 
+// ── Bounded readers — peak memory, not just wall time (fleet #3322) ──────
+//
+// `parseJson` above does `await res.text()` then `JSON.parse(raw)`, so a
+// multi-megabyte upstream body is held TWICE at peak: once as the raw string
+// and once as the object graph (which for an array of small objects runs
+// several times the byte size). Nothing in `workers/gateway` can see that
+// peak — it happens inside the pack, before any gateway code holds a payload —
+// which is why the gateway-side payload caps of #3293 and #3320 did not move
+// the 503/1102 rate at all.
+//
+// Measured on the live gateway, 2026-10-19, sha fa5de5c34, direct tools/call:
+//   statcan_list_cubes  8,049,566 B   <- one upstream array, no limit argument
+//   gnomad_gene BRCA1   2,732,706 B
+// Five of those in flight at the quality runner's width is ~40 MB of parse
+// peak against a 128 MB isolate; Cloudflare kills the isolate and every
+// request in flight returns 503/1102, which is why the victims look random.
+//
+// `readJsonArrayPrefix` is the fix for the dominant shape: an upstream that
+// returns one enormous top-level JSON array with no server-side paging. It
+// parses elements off the stream and DISCARDS the consumed bytes, so peak
+// retained memory is one element plus one chunk rather than the whole body,
+// and it cancels the body as soon as it has what it was asked for.
+
+/** Default ceiling for a single upstream JSON body read into an isolate.
+ *  Chosen so the known tail (statcan 8.0 MB, gnomad 2.7 MB) is bounded well
+ *  below the point where a handful of concurrent reads exhausts 128 MB. */
+const DEFAULT_MAX_JSON_BYTES = 1_500_000;
+
+interface JsonArrayPrefix<T> {
+  /** The decoded elements, at most `maxItems` of them. */
+  items: T[];
+  /** True when the upstream array had more elements than we read. */
+  truncated: boolean;
+  /** Bytes actually pulled off the socket — the number that bounds memory. */
+  bytesRead: number;
+  /** `Content-Length` if the upstream declared one, else null. */
+  declaredBytes: number | null;
+}
+
+/**
+ * Read the FIRST `maxItems` elements of a top-level JSON array off a response
+ * stream without ever materialising the whole body.
+ *
+ * Use this instead of `parseJson` when the upstream returns one big array and
+ * offers no `limit`/`offset` of its own. Peak retained memory is one element
+ * plus the current chunk; the body is cancelled as soon as `maxItems` is
+ * reached, so the remaining megabytes are never transferred at all.
+ *
+ *     const { items, truncated } = await readJsonArrayPrefix<Cube>(
+ *       res, 'StatCan WDS', { maxItems: limit + offset });
+ *
+ * `maxBytes` is a second, independent backstop for a pathological element (a
+ * single 50 MB object): hitting it stops the read and sets `truncated`, rather
+ * than throwing, because a partial answer plus an explicit truncated flag is
+ * more useful to a caller than an error.
+ */
+async function readJsonArrayPrefix<T>(
+  res: Response,
+  name: string,
+  opts: { maxItems: number; maxBytes?: number },
+): Promise<JsonArrayPrefix<T>> {
+  const maxBytes = opts.maxBytes ?? DEFAULT_MAX_JSON_BYTES;
+  const declaredRaw = Number(res.headers.get('content-length'));
+  const declaredBytes = Number.isFinite(declaredRaw) && declaredRaw > 0 ? declaredRaw : null;
+
+  // No streaming body available (a mock, or an already-buffered response):
+  // fall back to the whole-text path so behaviour stays correct, just unbounded.
+  if (!res.body) {
+    const all = await parseJson<T[]>(res, name);
+    if (!Array.isArray(all)) throw new Error(`upstream_down: ${name} returned ${typeof all} where a JSON array was expected.`);
+    return {
+      items: all.slice(0, opts.maxItems),
+      truncated: all.length > opts.maxItems,
+      bytesRead: declaredBytes ?? 0,
+      declaredBytes,
+    };
+  }
+
+  const reader = res.body.getReader();
+  const decoder = new TextDecoder();
+  const items: T[] = [];
+
+  let buf = '';
+  let scan = 0;
+  let bytesRead = 0;
+  let started = false; // consumed the opening '['
+  let arrayEnded = false;
+  let depth = 0;
+  let inStr = false;
+  let esc = false;
+  let elemStart = -1;
+  let truncated = false;
+
+  const push = (text: string) => {
+    const t = text.trim();
+    if (!t) return;
+    items.push(JSON.parse(t) as T);
+  };
+
+  try {
+    outer: while (items.length < opts.maxItems && !arrayEnded) {
+      const { value, done } = await reader.read();
+      if (done) break;
+      bytesRead += value.byteLength;
+      buf += decoder.decode(value, { stream: true });
+      if (bytesRead > maxBytes) { truncated = true; break; }
+
+      for (; scan < buf.length; scan++) {
+        const ch = buf[scan];
+
+        if (inStr) {
+          if (esc) { esc = false; continue; }
+          if (ch === '\\') { esc = true; continue; }
+          if (ch === '"') inStr = false;
+          continue;
+        }
+
+        if (!started) {
+          if (ch === '[') { started = true; continue; }
+          if (ch.trim() === '') continue;
+          throw new Error(
+            `upstream_down: ${name} returned a body that does not begin with a JSON array (saw ${JSON.stringify(ch)}). ` +
+              'This reader is only for upstreams that answer with one top-level array.',
+          );
+        }
+
+        if (ch === '"') { if (depth === 0 && elemStart < 0) elemStart = scan; inStr = true; continue; }
+
+        if (ch === '{' || ch === '[') {
+          if (depth === 0 && elemStart < 0) elemStart = scan;
+          depth++;
+          continue;
+        }
+
+        if (ch === '}' || ch === ']') {
+          if (depth > 0) {
+            depth--;
+            if (depth === 0 && elemStart >= 0) {
+              push(buf.slice(elemStart, scan + 1));
+              elemStart = -1;
+              if (items.length >= opts.maxItems) { truncated = true; scan++; break outer; }
+            }
+            continue;
+          }
+          // depth 0 and a closing bracket: the end of the top-level array.
+          if (elemStart >= 0) { push(buf.slice(elemStart, scan)); elemStart = -1; }
+          arrayEnded = true;
+          break outer;
+        }
+
+        if (ch === ',' && depth === 0) {
+          if (elemStart >= 0) {
+            push(buf.slice(elemStart, scan));
+            elemStart = -1;
+            if (items.length >= opts.maxItems) { truncated = true; scan++; break outer; }
+          }
+          continue;
+        }
+
+        // A bare scalar element (number / true / false / null).
+        if (depth === 0 && elemStart < 0 && ch.trim() !== '') elemStart = scan;
+      }
+
+      // Drop everything already consumed. THIS is what bounds memory: without
+      // it `buf` grows to the full body and the reader is streaming in name only.
+      if (elemStart >= 0) {
+        buf = buf.slice(elemStart);
+        scan -= elemStart;
+        elemStart = 0;
+      } else {
+        buf = buf.slice(scan);
+        scan = 0;
+      }
+    }
+  } finally {
+    // Cancel rather than drain: the point is that the remaining megabytes are
+    // never transferred. A cancel on an already-finished body is a no-op.
+    await reader.cancel().catch(() => {});
+  }
+
+  return { items, truncated, bytesRead, declaredBytes };
+}
+
+/**
+ * `parseJson`, with a hard ceiling on how many bytes are read.
+ *
+ * For an upstream whose body is not a single array (so `readJsonArrayPrefix`
+ * does not apply) but which can still answer with megabytes. Unlike the array
+ * reader this one THROWS when the cap is hit, because a truncated prefix of an
+ * arbitrary JSON document is not parseable and there is no partial answer to
+ * hand back. A declared `Content-Length` over the cap is refused before a
+ * single chunk is read.
+ */
+async function parseJsonBounded<T>(
+  res: Response,
+  name: string,
+  maxBytes: number = DEFAULT_MAX_JSON_BYTES,
+): Promise<T> {
+  const declaredRaw = Number(res.headers.get('content-length'));
+  const declared = Number.isFinite(declaredRaw) && declaredRaw > 0 ? declaredRaw : null;
+
+  const tooBig = (n: number) =>
+    new Error(
+      `upstream_down: ${name} answered with ${n.toLocaleString()} bytes, over this pack's ${maxBytes.toLocaleString()}-byte ceiling. ` +
+        'The body is not read past the ceiling on purpose — a few concurrent reads of a body that size exhaust the ' +
+        'Worker isolate and take every other in-flight request down with them. Narrow the request (a smaller range, ' +
+        'a specific id, or a lower limit) and retry.',
+    );
+
+  if (declared !== null && declared > maxBytes) {
+    await res.body?.cancel().catch(() => {});
+    throw tooBig(declared);
+  }
+  if (!res.body) return parseJson<T>(res, name);
+
+  const reader = res.body.getReader();
+  const decoder = new TextDecoder();
+  let raw = '';
+  let bytesRead = 0;
+  try {
+    for (;;) {
+      const { value, done } = await reader.read();
+      if (done) break;
+      bytesRead += value.byteLength;
+      if (bytesRead > maxBytes) throw tooBig(bytesRead);
+      raw += decoder.decode(value, { stream: true });
+    }
+    raw += decoder.decode();
+  } finally {
+    await reader.cancel().catch(() => {});
+  }
+
+  if (!raw.trim()) {
+    throw new Error(
+      `upstream_down: ${name} answered HTTP ${res.status} with an EMPTY body where JSON was expected. ` +
+        'Nothing about the request can cause this — it is an upstream fault, and the same call may well work on retry.',
+    );
+  }
+  try {
+    return JSON.parse(raw) as T;
+  } catch {
+    throw new Error(
+      `upstream_down: ${name} answered HTTP ${res.status} with a body that is not valid JSON. ` +
+        `It begins: ${stripMarkup(raw).slice(0, 120) || '(unreadable)'}`,
+    );
+  }
+}
+
 
 /**
  * Minimal text extractor for simple, text-based PDFs — the kind a
@@ -858,7 +1122,10 @@ interface PdfOutlineEntry {
 }
 
 type FontDecoder =
-  | { kind: 'winansi' }
+  // `simple`: the font is a declared simple (1-byte-code) font with
+  // /Encoding /WinAnsiEncoding, so a HEX string shown through it is a run of
+  // 1-byte windows-1252 codes, not 2-byte CIDs (fleet #3478).
+  | { kind: 'winansi'; simple?: boolean }
   | { kind: 'cid'; map: Map<number, number> }
   // Simple (1-byte-code) font decoded via its own /ToUnicode CMap rather than
   // WinAnsi — see fontDecoderFor's comment on North Dakota's subsetted fonts.
@@ -1319,13 +1586,25 @@ async function extractPdfText(buf: ArrayBuffer, options?: PdfExtractOptions): Pr
         if (map.size) return { kind: 'mapped1', map };
       }
       warnings.push(`font object ${fontObjNum} has a non-WinAnsi simple encoding and no usable /ToUnicode; decoded as WinAnsi anyway`);
+      return { kind: 'winansi' };
     }
-    return { kind: 'winansi' };
+    return { kind: 'winansi', simple: /\/Subtype\s*\/(?:TrueType|Type1)\b/.test(obj.dict) };
   }
 
   const TOKEN_RE =
     /\/(\S+)\s+[\d.]+\s+Tf|\[((?:\\.|[^\]])*)\]\s*TJ|<([0-9A-Fa-f\s]*)>\s*Tj|\(((?:\\.|[^()])*)\)\s*Tj|1\s+0\s+0\s+1\s+[\d.-]+\s+([\d.-]+)\s+Tm/g;
-  const TJ_PIECE_RE = /\(((?:\\.|[^()])*)\)|<([0-9A-Fa-f\s]*)>/g;
+  // Third alternative (bare number) added for Indonesia's JDIH statute PDFs
+  // (jdih.setneg.go.id, fleet #3121): that producer never writes a literal
+  // space byte between words — "REPUBLIK INDONESIA" is encoded as one TJ run
+  // `[(…K)-269.0(I)…]`, where the -269 IS the word gap, not kerning. Every
+  // sampled font's intra-word kerning tops out around +-30 (observed: 2.0,
+  // -2.9, -14.0, -24.0); the real word gaps in the same document measured
+  // -269, -286, -290, -692, -1396, -1400 — two clusters with nothing between
+  // them, so a flat threshold cleanly separates "kerning" from "space" without
+  // needing the current font size (the TJ number is already in thousandths of
+  // text space, before the Tfs/Th scale-up ISO 32000 §9.4.3 applies).
+  const TJ_PIECE_RE = /\(((?:\\.|[^()])*)\)|<([0-9A-Fa-f\s]*)>|(-?\d+(?:\.\d+)?)/g;
+  const TJ_WORD_GAP_THRESHOLD = -100;
 
   function decodeRun(dec: FontDecoder, literal: string | undefined, hex: string | undefined): string {
     if (literal !== undefined) {
@@ -1378,6 +1657,16 @@ async function extractPdfText(buf: ArrayBuffer, options?: PdfExtractOptions): Pr
     }
     const cleaned = (hex ?? '').replace(/\s+/g, '');
     let out = '';
+    // A hex string shown through a declared simple WinAnsi font is 1-byte
+    // codes. Reading it as 2-byte CIDs (the fall-through below) kept only
+    // every other byte's low half and silently dropped Danish ø/å/æ from
+    // Folketinget minister answers: `<F86265>` is "øbe", not one glyph 'b'
+    // (fleet #3478). Odd trailing nibble is ignored.
+    if (dec.kind === 'winansi' && dec.simple) {
+      const bytes = new Uint8Array(Math.floor(cleaned.length / 2));
+      for (let i = 0; i < bytes.length; i++) bytes[i] = parseInt(cleaned.slice(i * 2, i * 2 + 2), 16);
+      return new TextDecoder('windows-1252').decode(bytes);
+    }
     if (dec.kind === 'mapped1') {
       // 1-byte code per glyph (a simple font's code space), unlike Type0's
       // 2-byte CIDs below — see fontDecoderFor.
@@ -1450,7 +1739,17 @@ async function extractPdfText(buf: ArrayBuffer, options?: PdfExtractOptions): Pr
       }
       const dec = (currentFont && fontMap.get(currentFont)) || { kind: 'winansi' as const };
       if (m[2] !== undefined) {
-        for (const piece of m[2].matchAll(TJ_PIECE_RE)) out += decodeRun(dec, piece[1], piece[2]);
+        for (const piece of m[2].matchAll(TJ_PIECE_RE)) {
+          if (piece[3] !== undefined) {
+            // A big negative adjustment between two TJ pieces with no
+            // literal space byte at all — see TJ_PIECE_RE's comment above.
+            // Skip if `out` already ends in whitespace so a producer that
+            // DOES write literal spaces never gets a doubled one.
+            if (parseFloat(piece[3]) <= TJ_WORD_GAP_THRESHOLD && !/\s$/.test(out)) out += ' ';
+            continue;
+          }
+          out += decodeRun(dec, piece[1], piece[2]);
+        }
       } else if (m[3] !== undefined) {
         out += decodeRun(dec, undefined, m[3]);
       } else if (m[4] !== undefined) {

package/src/server.ts

@@ -11,7 +11,7 @@ import * as bakedIndexModule from './gu-baked-data.js';
 const bakedIndex = Object.fromEntries(Object.entries(bakedIndexModule));
 
 const server = new Server(
-  { name: '@pipeworx/mcp-guam-code', version: '0.1.2' },
+  { name: '@pipeworx/mcp-guam-code', version: '0.1.4' },
   { capabilities: { tools: {} } },
 );