captchakraken 3.1.0 → 3.2.0

package/python/examples/_harness.py

@@ -1,177 +0,0 @@
-import argparse
-import json
-import os
-import sys
-import time
-from dataclasses import dataclass, replace
-
-from camoufox.sync_api import Camoufox
-
-from captchakraken.page_solver import CaptchaSolveError, PageSolver
-
-
-@dataclass
-class DemoSpec:
-    name: str
-    url: str
-    vendor: str
-
-
-def spec_from_argv(default: DemoSpec, argv=None) -> DemoSpec:
-    ap = argparse.ArgumentParser(
-        description=f"CaptchaKraken demo — {default.name}",
-        epilog="With no URL, runs against the built-in demo page.")
-    ap.add_argument("url", nargs="?", default=default.url,
-                    help=f"page to solve (default: {default.url})")
-    ap.add_argument("--vendor", default=default.vendor,
-                    choices=("hcaptcha", "recaptcha"),
-                    help="only affects failure wording; the solver auto-detects")
-    ap.add_argument("--name", default=None, help="label for the report")
-    ap.add_argument("--headed", action="store_true",
-                    help="show the browser (same as HEADLESS=0)")
-    args = ap.parse_args(argv)
-
-    if args.headed:
-        os.environ["HEADLESS"] = "0"
-    name = args.name or (default.name if args.url == default.url else args.url)
-    return replace(default, url=args.url, vendor=args.vendor, name=name)
-
-
-def _launch_kwargs() -> dict:
-    kw = dict(headless=os.getenv("HEADLESS", "1") != "0",
-              humanize=os.getenv("HUMANIZE", "0") != "0", geoip=False)
-    binary = os.getenv("CAMOUFOX_BINARY") or os.getenv("CAMOUFOX_EXECUTABLE_PATH")
-    if binary:
-        kw["executable_path"] = binary
-    return kw
-
-
-def _page_kwargs() -> dict:
-    out = os.getenv("CAPTCHA_DEMO_VIDEO_DIR")
-    if not out:
-        return {}
-    w, _, h = os.getenv("CAPTCHA_DEMO_VIDEO_SIZE", "1280x800").partition("x")
-    return {"record_video_dir": out,
-            "record_video_size": {"width": int(w), "height": int(h)}}
-
-
-def _finish_video(page):
-    video = getattr(page, "video", None)
-    if video is None:
-        return None
-    try:
-        page.context.close()
-        return video.path()
-    except Exception:
-        return None
-
-
-def _tokens(usage) -> tuple[int, int]:
-    inp = out = 0
-    for u in usage or []:
-        inp += int(u.get("prompt_tokens", u.get("inputTokens", 0)) or 0)
-        out += int(u.get("completion_tokens", u.get("outputTokens", 0)) or 0)
-    return inp, out
-
-
-def _explain(vendor: str, solved: bool, err) -> str:
-    msg = str(err or "").lower()
-    if "unsupported" in msg or "cannot solve" in msg:
-        if vendor == "hcaptcha":
-            return ("hCaptcha served a challenge type the solver does not handle "
-                    "yet (drag / video / choose-the-card) — re-run for a grid.")
-        return "Not a supported challenge type — re-run to try again."
-    if "no captcha" in msg or "nocaptchafound" in msg:
-        return ("No captcha widget was found on that page. Check the URL, or "
-                "the widget may load only after an interaction.")
-    if "vllm" in msg or "connection" in msg or "refused" in msg or "max retries" in msg:
-        return ("Could not reach the vLLM server — is it up and is VLLM_BASE_URL "
-                "correct? (A local server auto-starts only with captchakraken[serve].)")
-    if "timeout" in msg:
-        return "Timed out — the page or challenge never became interactable."
-    if err:
-        return str(err)
-    if not solved:
-        return ("The answer was submitted but the vendor did not accept it. Most "
-                "often IP reputation or fingerprint flagging rather than a wrong "
-                "answer — try a cleaner IP or a residential proxy.")
-    return "Unknown failure."
-
-
-def run_demo(spec: DemoSpec) -> None:
-    spec = spec_from_argv(spec)
-
-    t0 = time.time()
-    ok = False
-    reason = None
-    inp = out = 0
-    solve_s = 0.0
-    video_path = None
-    err = None
-
-    try:
-        with Camoufox(**_launch_kwargs()) as browser:
-            page = browser.new_page(**_page_kwargs())
-            try:
-                page.goto(spec.url, wait_until="domcontentloaded", timeout=60_000)
-                try:
-                    page.wait_for_load_state("networkidle", timeout=15_000)
-                except Exception:
-                    pass
-                page.wait_for_timeout(3000)
-
-                s0 = time.time()
-                try:
-                    result = PageSolver().solve(page)
-                finally:
-                    solve_s = time.time() - s0
-                ok = bool(result.is_solved)
-                inp, out = _tokens(result.token_usage)
-                if not ok:
-                    reason = _explain(spec.vendor, ok, None)
-            except CaptchaSolveError as e:
-                err = e
-                reason = _explain(spec.vendor, False, e)
-            finally:
-                video_path = _finish_video(page)
-    except Exception as e:
-        err = e
-        reason = _explain(spec.vendor, False, e)
-
-    tps = out / solve_s if out and solve_s > 0 else 0.0
-    _report(spec, ok=ok, total_s=time.time() - t0, solve_s=solve_s,
-            inp=inp, out=out, tps=tps, reason=reason, video=video_path)
-
-
-def _fmt(s: float) -> str:
-    return f"{s:.1f}s" if s >= 1 else f"{int(s * 1000)}ms"
-
-
-def _report(spec, *, ok, total_s, solve_s, inp, out, tps, reason, video=None):
-    line = "─" * 52
-    print(f"\n{line}")
-    print(f"  CaptchaKraken demo — {spec.name}")
-    print(f"  {spec.url}")
-    print(line)
-    print(f"  result        : {'✓ SOLVED' if ok else '✗ not solved'}")
-    print(f"  total time    : {_fmt(total_s)}   (solve: {_fmt(solve_s)})")
-    print(f"  tokens        : {inp} in / {out} out")
-    print(f"  gen speed     : {f'~{tps:.1f} tok/s' if tps > 0 else 'n/a'}")
-    if video:
-        print(f"  video         : {video}")
-    if reason:
-        print(f"  reason        : {reason}")
-    print(f"{line}\n")
-
-    record = os.getenv("CAPTCHA_DEMO_RECORD")
-    if record:
-        with open(record, "a", encoding="utf-8") as fh:
-            fh.write(json.dumps({
-                "demo": spec.name, "url": spec.url, "vendor": spec.vendor,
-                "ok": ok, "reason": reason,
-                "total_s": round(total_s, 1), "solve_s": round(solve_s, 1),
-                "input_tokens": inp, "output_tokens": out,
-                "tokens_per_sec": round(tps, 1), "video": video,
-            }) + "\n")
-
-    sys.exit(0 if ok else 1)

package/python/examples/demoHcaptcha.py

@@ -1,8 +0,0 @@
-from _harness import DemoSpec, run_demo
-
-if __name__ == "__main__":
-    run_demo(DemoSpec(
-        name="hCaptcha",
-        url="https://accounts.hcaptcha.com/demo",
-        vendor="hcaptcha",
-    ))

package/python/examples/demoRecaptcha.py

@@ -1,8 +0,0 @@
-from _harness import DemoSpec, run_demo
-
-if __name__ == "__main__":
-    run_demo(DemoSpec(
-        name="reCAPTCHA v2",
-        url="https://www.google.com/recaptcha/api2/demo",
-        vendor="recaptcha",
-    ))

package/python/examples/watch_playwright.py

@@ -1,31 +0,0 @@
-import sys
-
-from playwright.sync_api import sync_playwright
-
-from captchakraken import PageSolver
-
-URL = sys.argv[1] if len(sys.argv) > 1 else "https://www.google.com/recaptcha/api2/demo"
-SECONDS = int(sys.argv[2]) if len(sys.argv) > 2 else 30
-
-
-def main() -> int:
-    with sync_playwright() as p:
-        browser = p.chromium.launch(headless=True)
-        page = browser.new_page()
-
-        watcher = PageSolver().watch(
-            page,
-            on_solved=lambda r: print("✅ solved one:", r.is_solved),
-            on_error=lambda e: print("solve failed:", e),
-        )
-
-        page.goto(URL)
-        watcher.run(timeout_ms=SECONDS * 1000)
-
-        print(f"stopped after {watcher.solves} solve(s)")
-        browser.close()
-        return 0 if watcher.solves else 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

package/python/examples/with_local_gguf.py

@@ -1,28 +0,0 @@
-import os
-import sys
-
-from captchakraken import solve_captcha
-
-TEXT_CAPTCHAS = {"--text"}
-
-
-def main() -> int:
-    args = [a for a in sys.argv[1:] if a not in TEXT_CAPTCHAS]
-    if not args:
-        print(__doc__)
-        return 2
-    image = args[0]
-    text_mode = bool(TEXT_CAPTCHAS & set(sys.argv[1:]))
-
-    print(f"endpoint : {os.getenv('VLLM_BASE_URL', '(default)')}")
-    print(f"model    : {os.getenv('CAPTCHA_LORA_NAME', '(registry default)')}")
-    print(f"prompts  : {os.getenv('CAPTCHA_LORA_ADAPTER', '(latest)')}")
-
-    result = solve_captcha(image, text_mode=text_mode)
-
-    print(f"answer   : {result}")
-    return 0 if result else 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

package/python/examples/with_playwright.py

@@ -1,24 +0,0 @@
-import sys
-
-from playwright.sync_api import sync_playwright
-
-from captchakraken import PageSolver
-
-URL = sys.argv[1] if len(sys.argv) > 1 else "https://www.google.com/recaptcha/api2/demo"
-
-
-def main() -> int:
-    with sync_playwright() as p:
-        browser = p.chromium.launch(headless=True)
-        page = browser.new_page()
-        page.goto(URL)
-
-        result = PageSolver().solve(page)
-
-        print("✅ solved" if result.is_solved else "❌ not solved")
-        browser.close()
-        return 0 if result.is_solved else 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())

package/scripts/copy-python.mjs

@@ -1,29 +0,0 @@
-import { cpSync, rmSync, existsSync } from 'node:fs'
-import { dirname, resolve } from 'node:path'
-import { fileURLToPath } from 'node:url'
-
-const here = dirname(fileURLToPath(import.meta.url))
-const src = resolve(here, '../../python')
-const dest = resolve(here, '../python')
-
-if (!existsSync(src)) {
-  console.error(`[copy-python] source engine not found at ${src}`)
-  process.exit(1)
-}
-
-const SKIP = new Set([
-  '.venv', '__pycache__', 'dist', 'build', '.pytest_cache', '.ruff_cache', 'tests',
-  'coverage', 'htmlcov', '.nyc_output', 'coverage.xml', 'lcov.info',
-
-  'latestDebugRun',
-])
-
-rmSync(dest, { recursive: true, force: true })
-cpSync(src, dest, {
-  recursive: true,
-  filter: (p) =>
-    !p
-      .split(/[\\/]/)
-      .some((seg) => SKIP.has(seg) || seg.endsWith('.egg-info') || seg.startsWith('.coverage')),
-})
-console.log('[copy-python] bundled python/ engine -> js/python/')