patchright-cli 0.1.0__py3-none-any.whl

patchright_cli/snapshot.py

@@ -0,0 +1,242 @@
+"""Generate YAML accessibility tree snapshots from Patchright pages.
+
+Uses a two-pass approach:
+1. TreeWalker in strict document order to assign data-patchright-ref attributes
+2. Recursive tree builder to produce a nested YAML structure
+
+This guarantees ref order matches DOM order (unlike recursive walk which
+can assign refs depth-first, causing mismatches).
+"""
+
+from __future__ import annotations
+
+import time
+from pathlib import Path
+
+_SNAPSHOT_JS = r"""
+(() => {
+    const SKIP = new Set(["SCRIPT","STYLE","NOSCRIPT","SVG","LINK","META","BR","HR"]);
+    const INTERACTIVE = new Set(["A","BUTTON","INPUT","TEXTAREA","SELECT","DETAILS","SUMMARY","LABEL"]);
+    const SEMANTIC = new Set(["H1","H2","H3","H4","H5","H6","NAV","MAIN","HEADER","FOOTER","ARTICLE","SECTION","ASIDE","FORM","TABLE","UL","OL","LI","IMG","VIDEO","AUDIO","IFRAME"]);
+
+    document.querySelectorAll("[data-patchright-ref]").forEach(el => el.removeAttribute("data-patchright-ref"));
+
+    function getRole(el) {
+        const ar = el.getAttribute && el.getAttribute("role");
+        if (ar) return ar;
+        const t = el.tagName;
+        const m = {A:"link",BUTTON:"button",TEXTAREA:"textbox",SELECT:"combobox",IMG:"img",TABLE:"table",FORM:"form",NAV:"navigation",MAIN:"main",HEADER:"banner",FOOTER:"contentinfo"};
+        if (m[t]) return m[t];
+        if (t === "INPUT") { const tp = (el.type||"text").toLowerCase(); return tp==="checkbox"?"checkbox":tp==="radio"?"radio":(tp==="submit"||tp==="button")?"button":"textbox"; }
+        if (t === "UL" || t === "OL") return "list";
+        if (t === "LI") return "listitem";
+        if (/^H[1-6]$/.test(t)) return "heading";
+        return t.toLowerCase();
+    }
+
+    function getName(el) {
+        const t = el.tagName;
+        const ar = el.getAttribute && el.getAttribute("aria-label");
+        if (ar) return ar.substring(0, 120);
+        if (t === "IMG") return (el.getAttribute("alt") || "").substring(0, 120);
+        if (t === "INPUT" || t === "TEXTAREA") return (el.getAttribute("placeholder") || "").substring(0, 80);
+        if (t === "A" || t === "BUTTON" || /^H[1-6]$/.test(t) || t === "LABEL" || t === "LI" || t === "SUMMARY")
+            return (el.innerText || "").replace(/\s+/g, " ").trim().substring(0, 120);
+        const ti = el.getAttribute && el.getAttribute("title");
+        if (ti) return ti.substring(0, 80);
+        return "";
+    }
+
+    function isVis(el) {
+        if (!el.offsetParent && el.tagName !== "BODY" && el.tagName !== "HTML") return false;
+        try { const s = getComputedStyle(el); return s.display !== "none" && s.visibility !== "hidden" && s.opacity !== "0"; }
+        catch(e) { return false; }
+    }
+
+    function shouldTag(el) {
+        return INTERACTIVE.has(el.tagName) || SEMANTIC.has(el.tagName) || !!el.getAttribute("role") || !!el.getAttribute("data-testid") || !!getName(el);
+    }
+
+    // Pass 1: TreeWalker in strict document order — assign refs
+    const tw = document.createTreeWalker(document.body, NodeFilter.SHOW_ELEMENT, {
+        acceptNode(n) {
+            if (SKIP.has(n.tagName)) return NodeFilter.FILTER_REJECT;
+            if (!isVis(n)) return NodeFilter.FILTER_REJECT;
+            return NodeFilter.FILTER_ACCEPT;
+        }
+    });
+
+    let counter = 0;
+    const flatList = [];
+    let node;
+    while (node = tw.nextNode()) {
+        if (shouldTag(node)) {
+            counter++;
+            const ref = "e" + counter;
+            node.setAttribute("data-patchright-ref", ref);
+            const role = getRole(node);
+            const name = getName(node);
+            const entry = { ref, role, name, tag: node.tagName };
+            if (node.tagName === "INPUT" || node.tagName === "TEXTAREA" || node.tagName === "SELECT") {
+                if (node.value) entry.value = node.value.substring(0, 200);
+                if (node.type === "checkbox" || node.type === "radio") entry.checked = node.checked;
+                if (node.disabled) entry.disabled = true;
+            }
+            if (node.tagName === "A" && node.href) entry.url = node.href.substring(0, 200);
+            if (/^H[1-6]$/.test(node.tagName)) entry.level = parseInt(node.tagName[1]);
+            flatList.push(entry);
+        }
+    }
+
+    // Pass 2: Build nested tree reading refs from DOM attributes
+    function buildTree(el, depth) {
+        if (depth > 12 || !el || !el.tagName) return null;
+        if (SKIP.has(el.tagName)) return null;
+        const ref = el.getAttribute("data-patchright-ref");
+        const children = [];
+        for (const ch of el.children) {
+            if (!isVis(ch)) continue;
+            const c = buildTree(ch, depth + 1);
+            if (c) children.push(c);
+        }
+        if (!ref && !children.length) return null;
+        const out = {};
+        if (ref) {
+            const f = flatList.find(x => x.ref === ref);
+            if (f) Object.assign(out, f);
+            delete out.tag;
+        }
+        if (children.length) out.children = children;
+        return out;
+    }
+
+    const tree = buildTree(document.body, 0) || { role: "document", name: document.title, children: [] };
+    return { tree, flatList };
+})()
+"""
+
+
+def _walk_tree(node: dict, depth: int = 0) -> list[str]:
+    """Recursively walk the parsed tree and produce YAML lines."""
+    lines: list[str] = []
+    ref = node.get("ref", "")
+    role = node.get("role", "")
+    name = node.get("name", "")
+
+    if ref:
+        indent = "  " * depth
+        lines.append(f"{indent}- ref: {ref}")
+        lines.append(f"{indent}  role: {role}")
+        if name:
+            lines.append(f"{indent}  name: {_yaml_escape(name)}")
+        for key in ("value", "url"):
+            val = node.get(key, "")
+            if val:
+                lines.append(f"{indent}  {key}: {_yaml_escape(str(val))}")
+        if node.get("checked") is not None:
+            lines.append(f"{indent}  checked: {str(node['checked']).lower()}")
+        if node.get("disabled"):
+            lines.append(f"{indent}  disabled: true")
+        if node.get("level") is not None:
+            lines.append(f"{indent}  level: {node['level']}")
+
+        children = node.get("children", [])
+        if children:
+            lines.append(f"{indent}  children:")
+            for child in children:
+                lines.extend(_walk_tree(child, depth + 1))
+    else:
+        for child in node.get("children", []):
+            lines.extend(_walk_tree(child, depth))
+
+    return lines
+
+
+def _yaml_escape(s: str) -> str:
+    if not s:
+        return '""'
+    if any(
+        c in s
+        for c in (
+            ":",
+            "#",
+            "'",
+            '"',
+            "\n",
+            "[",
+            "]",
+            "{",
+            "}",
+            ",",
+            "&",
+            "*",
+            "?",
+            "|",
+            "-",
+            "<",
+            ">",
+            "=",
+            "!",
+            "%",
+            "@",
+            "`",
+        )
+    ):
+        escaped = s.replace("\\", "\\\\").replace('"', '\\"').replace("\n", "\\n")
+        return f'"{escaped}"'
+    return s
+
+
+async def take_snapshot(page) -> tuple[str, dict[str, dict]]:
+    """Take a DOM snapshot. Returns (yaml_text, ref_map)."""
+    result = None
+    try:
+        result = await page.evaluate(_SNAPSHOT_JS, isolated_context=False)
+    except TypeError:
+        try:
+            result = await page.evaluate(_SNAPSHOT_JS)
+        except Exception:
+            pass
+    except Exception:
+        pass
+
+    if not result or not result.get("tree"):
+        return "# Empty page - no accessible elements found\n", {}
+
+    flat_list = result.get("flatList", [])
+
+    refs: dict[str, dict] = {}
+    lines: list[str] = []
+    for item in flat_list:
+        refs[item["ref"]] = item
+        lines.append(f"- ref: {item['ref']}")
+        lines.append(f"  role: {item.get('role', '')}")
+        name = item.get("name", "")
+        if name:
+            lines.append(f"  name: {_yaml_escape(name)}")
+        for key in ("value", "url"):
+            val = item.get(key, "")
+            if val:
+                lines.append(f"  {key}: {_yaml_escape(str(val))}")
+        if item.get("checked") is not None:
+            lines.append(f"  checked: {str(item['checked']).lower()}")
+        if item.get("disabled"):
+            lines.append("  disabled: true")
+        if item.get("level") is not None:
+            lines.append(f"  level: {item['level']}")
+
+    yaml_text = "\n".join(lines) + "\n"
+    return yaml_text, refs
+
+
+def save_snapshot(yaml_text: str, cwd: str | None = None) -> str:
+    """Save snapshot YAML to .patchright-cli/ directory."""
+    base = Path(cwd) if cwd else Path.cwd()
+    snap_dir = base / ".patchright-cli"
+    snap_dir.mkdir(parents=True, exist_ok=True)
+
+    timestamp = int(time.time() * 1000)
+    filename = f"page-{timestamp}.yml"
+    filepath = snap_dir / filename
+    filepath.write_text(yaml_text, encoding="utf-8")
+    return str(filepath)

patchright_cli-0.1.0.dist-info/METADATA

@@ -0,0 +1,311 @@
+Metadata-Version: 2.4
+Name: patchright-cli
+Version: 0.1.0
+Summary: Anti-detect browser automation CLI using Patchright (undetected Playwright fork)
+Project-URL: Homepage, https://github.com/AhaiMk01/patchright-cli
+Project-URL: Repository, https://github.com/AhaiMk01/patchright-cli
+Project-URL: Issues, https://github.com/AhaiMk01/patchright-cli/issues
+Author: zpahai
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: anti-detect,automation,browser,cli,patchright,playwright,scraping
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Internet :: WWW/HTTP :: Browsers
+Classifier: Topic :: Software Development :: Testing
+Requires-Python: >=3.10
+Requires-Dist: click
+Requires-Dist: patchright
+Requires-Dist: pyyaml
+Description-Content-Type: text/markdown
+
+# patchright-cli
+
+<!-- badges (hidden until PyPI publish)
+[![PyPI version](https://img.shields.io/pypi/v/patchright-cli?color=blue&label=PyPI)](https://pypi.org/project/patchright-cli/)
+[![Python](https://img.shields.io/pypi/pyversions/patchright-cli?label=Python)](https://pypi.org/project/patchright-cli/)
+[![License](https://img.shields.io/github/license/AhaiMk01/patchright-cli?color=green)](LICENSE)
+[![Downloads](https://img.shields.io/pypi/dm/patchright-cli?color=orange&label=Downloads)](https://pypi.org/project/patchright-cli/)
+-->
+
+Anti-detect browser automation CLI. Same command interface as [Microsoft's playwright-cli](https://github.com/microsoft/playwright-cli) but using [Patchright](https://github.com/kaliiiiiiiiii/patchright-python) (undetected Playwright fork) to bypass bot detection.
+
+## Why
+
+Regular Playwright and Chrome DevTools get blocked by Akamai, Cloudflare, and other anti-bot systems. Patchright patches Chromium's detection vectors (`navigator.webdriver`, WebGL fingerprints, etc.) so automated browsers look like real users.
+
+This CLI wraps Patchright in the same command interface as playwright-cli, so AI agents (Claude Code, etc.) can automate anti-bot-protected sites with the same workflow they'd use for regular Playwright.
+
+## Install
+
+### For Humans
+
+Just tell your AI agent:
+
+> Install and set up patchright-cli by following the instructions here:
+> https://raw.githubusercontent.com/AhaiMk01/patchright-cli/main/docs/installation.md
+
+Seriously — let the agent do it. It handles configs better than you.
+
+### For LLM Agents
+
+```bash
+curl -s https://raw.githubusercontent.com/AhaiMk01/patchright-cli/main/docs/installation.md
+```
+
+### Manual Install
+
+```bash
+# With pip
+pip install patchright-cli
+
+# With uv (recommended)
+uv tool install patchright-cli
+
+# Run without installing (like npx)
+uvx patchright-cli open https://example.com
+```
+
+### From Source
+
+```bash
+git clone https://github.com/AhaiMk01/patchright-cli.git
+cd patchright-cli
+uv venv && uv pip install -e .
+```
+
+## Quick Start
+
+```bash
+# Launch undetected Chrome and navigate
+patchright-cli open https://example.com
+
+# Take a snapshot to see interactive elements with refs
+patchright-cli snapshot
+
+# Interact using refs from the snapshot
+patchright-cli click e2
+patchright-cli fill e5 "search query"
+patchright-cli press Enter
+
+# Take a screenshot
+patchright-cli screenshot
+
+# Close the browser
+patchright-cli close
+```
+
+## Architecture
+
+```mermaid
+graph LR
+    A[CLI Client<br/>cli.py] -->|TCP/JSON<br/>localhost:9321| B[Daemon<br/>daemon.py]
+    B -->|Patchright<br/>CDP| C[Chrome<br/>stealth]
+```
+
+- **Daemon** (`daemon.py`): Long-running Python process managing browser sessions via Patchright. Listens on `localhost:9321`. Auto-starts on first `open` command.
+- **CLI** (`cli.py`): Thin client. Parses args, sends JSON to daemon, prints result. Each invocation connects and disconnects — the browser stays open between commands.
+- **Snapshot** (`snapshot.py`): Walks the DOM with `TreeWalker` in document order, assigns `data-patchright-ref` attributes, outputs a flat YAML list of interactive elements.
+
+## Commands
+
+### Core
+```bash
+patchright-cli open [url]              # Launch browser
+patchright-cli open --persistent       # With persistent profile
+patchright-cli goto <url>              # Navigate
+patchright-cli click <ref>             # Click element
+patchright-cli dblclick <ref>          # Double-click
+patchright-cli fill <ref> <value>      # Fill text input
+patchright-cli type <text>             # Type via keyboard
+patchright-cli hover <ref>             # Hover over element
+patchright-cli select <ref> <value>    # Select dropdown option
+patchright-cli check <ref>             # Check checkbox
+patchright-cli uncheck <ref>           # Uncheck checkbox
+patchright-cli drag <from> <to>        # Drag and drop
+patchright-cli snapshot                # Accessibility snapshot
+patchright-cli snapshot --filename=f   # Save to custom path
+patchright-cli eval <expr>             # Run JavaScript
+patchright-cli screenshot              # Full page screenshot
+patchright-cli screenshot <ref>        # Element screenshot
+patchright-cli screenshot --filename=f # Save to custom path
+patchright-cli close                   # Close session
+```
+
+### Navigation
+```bash
+patchright-cli go-back
+patchright-cli go-forward
+patchright-cli reload
+```
+
+### Keyboard / Mouse
+```bash
+patchright-cli press Enter
+patchright-cli keydown Shift
+patchright-cli keyup Shift
+patchright-cli mousemove 150 300
+patchright-cli mousedown [button]
+patchright-cli mouseup [button]
+patchright-cli mousewheel 0 100
+```
+
+### Dialog
+```bash
+patchright-cli dialog-accept [text]    # Accept next alert/confirm/prompt
+patchright-cli dialog-dismiss          # Dismiss next dialog
+```
+
+### Upload / Resize
+```bash
+patchright-cli upload ./file.pdf       # Upload to first file input
+patchright-cli upload ./file.pdf e5    # Upload to specific input
+patchright-cli resize 1920 1080        # Resize viewport
+```
+
+### Tabs
+```bash
+patchright-cli tab-list
+patchright-cli tab-new [url]
+patchright-cli tab-select <index>
+patchright-cli tab-close [index]
+```
+
+### State Persistence
+```bash
+patchright-cli state-save [file]       # Save cookies + localStorage
+patchright-cli state-load <file>       # Restore saved state
+```
+
+### Storage
+```bash
+# Cookies
+patchright-cli cookie-list
+patchright-cli cookie-list --domain=example.com
+patchright-cli cookie-get <name>
+patchright-cli cookie-set <name> <value>
+patchright-cli cookie-set <name> <value> --domain=example.com --httpOnly --secure
+patchright-cli cookie-delete <name>
+patchright-cli cookie-clear
+
+# localStorage
+patchright-cli localstorage-list
+patchright-cli localstorage-get <key>
+patchright-cli localstorage-set <key> <value>
+patchright-cli localstorage-delete <key>
+patchright-cli localstorage-clear
+
+# sessionStorage
+patchright-cli sessionstorage-list
+patchright-cli sessionstorage-get <key>
+patchright-cli sessionstorage-set <key> <value>
+patchright-cli sessionstorage-delete <key>
+patchright-cli sessionstorage-clear
+```
+
+### Request Mocking
+```bash
+patchright-cli route "**/*.jpg" --status=404
+patchright-cli route "https://api.example.com/**" --body='{"mock":true}'
+patchright-cli route-list
+patchright-cli unroute "**/*.jpg"
+patchright-cli unroute                 # Remove all routes
+```
+
+### Tracing / PDF
+```bash
+patchright-cli tracing-start
+patchright-cli tracing-stop            # Saves .zip trace file
+patchright-cli pdf --filename=page.pdf
+```
+
+### DevTools
+```bash
+patchright-cli console                 # All console messages
+patchright-cli console warning         # Filter by level
+patchright-cli network                 # Network request log
+```
+
+### Sessions
+```bash
+patchright-cli -s=mysession open https://example.com --persistent
+patchright-cli -s=mysession click e6
+patchright-cli -s=mysession close
+patchright-cli list                    # List all sessions
+patchright-cli close-all
+patchright-cli kill-all
+patchright-cli delete-data             # Delete persistent profile
+```
+
+## Snapshots
+
+After each state-changing command, the CLI outputs page info and a YAML snapshot:
+
+```
+### Page
+- Page URL: https://example.com/
+- Page Title: Example Domain
+### Snapshot
+[Snapshot](.patchright-cli/page-1774376207818.yml)
+```
+
+The snapshot lists interactive elements with refs:
+
+```yaml
+- ref: e1
+  role: heading
+  name: Example Domain
+  level: 1
+- ref: e2
+  role: link
+  name: Learn more
+  url: "https://iana.org/domains/example"
+```
+
+Use refs in commands: `patchright-cli click e2`, `patchright-cli fill e5 "text"`.
+
+## Anti-Detect Features
+
+- Real Chrome browser (not Chromium)
+- Patchright patches `navigator.webdriver` and other detection vectors
+- Persistent profiles maintain cookies/sessions across runs
+- No custom user-agent or headers (natural fingerprint)
+- Headed by default (headless is more detectable)
+
+## Claude Code Integration
+
+Add the skill to your project:
+
+```bash
+cp -r skills/patchright-cli ~/.claude/skills/
+```
+
+Then use in Claude Code:
+```
+patchright-cli open https://protected-site.com
+patchright-cli snapshot
+patchright-cli fill e3 "username"
+patchright-cli fill e4 "password"
+patchright-cli click e5
+```
+
+## Star History
+
+<a href="https://star-history.com/#AhaiMk01/patchright-cli&Date">
+ <picture>
+   <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=AhaiMk01/patchright-cli&type=Date&theme=dark" />
+   <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=AhaiMk01/patchright-cli&type=Date" />
+   <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=AhaiMk01/patchright-cli&type=Date" />
+ </picture>
+</a>
+
+## License
+
+Apache 2.0 — same as [playwright-cli](https://github.com/microsoft/playwright-cli)

patchright_cli-0.1.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+patchright_cli/__init__.py,sha256=l3g0CuQ8etdd49K2XOpNKHxK0LTJMU-3YrmBH5LfGeQ,100
+patchright_cli/__main__.py,sha256=iVuBuI_rPOENLhffOEHoS7JWCDBQciqT1kPXjRuhbZ4,127
+patchright_cli/cli.py,sha256=2GDRDwrkzGo3sW0pTwGErRQr_qRHy6SOGTLuHdLOgDM,12241
+patchright_cli/daemon.py,sha256=w7L7s6mVjEs9DbHHsLqRpJPjijpqXrL6W49KFBQPqvg,35427
+patchright_cli/snapshot.py,sha256=KTIcRyUl2kUFs3yfpPi3YVvz17zRjr4rcfqFgiyt1PM,8906
+patchright_cli-0.1.0.dist-info/METADATA,sha256=ia2kLfALkYX_eccpgtJVnr6Su77mN1Xend_PntOP3ls,9961
+patchright_cli-0.1.0.dist-info/WHEEL,sha256=QccIxa26bgl1E6uMy58deGWi-0aeIkkangHcxk2kWfw,87
+patchright_cli-0.1.0.dist-info/entry_points.txt,sha256=Syh904eOQA8ylakyqIvG4LmIHx3LMjgeNcnFap3O0NQ,59
+patchright_cli-0.1.0.dist-info/licenses/LICENSE,sha256=adszqL4YR4pQUrc7LeAFxeg5Mc2cvljXp7QPlzKVWm8,10760
+patchright_cli-0.1.0.dist-info/RECORD,,

patchright_cli-0.1.0.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: hatchling 1.29.0
+Root-Is-Purelib: true
+Tag: py3-none-any

patchright_cli-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+patchright-cli = patchright_cli.cli:main