zrb 1.14.5__tar.gz → 1.14.6__tar.gz

{zrb-1.14.5 → zrb-1.14.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: zrb
-Version: 1.14.5
+Version: 1.14.6
 Summary: Your Automation Powerhouse
 Home-page: https://github.com/state-alchemists/zrb
 License: AGPL-3.0-or-later
@@ -22,17 +22,15 @@ Requires-Dist: chromadb (>=0.6.3,<0.7.0) ; extra == "rag" or extra == "all"
 Requires-Dist: fastapi[standard] (>=0.116.1,<0.117.0)
 Requires-Dist: isort (>=6.0.1,<6.1.0)
 Requires-Dist: libcst (>=1.7.0,<2.0.0)
-Requires-Dist: openai (>=1.86.0,<2.0.0) ; extra == "rag" or extra == "all"
-Requires-Dist: pdfplumber (>=0.11.6,<0.12.0) ; extra == "rag" or extra == "all"
+Requires-Dist: openai (>=1.92.0,<=1.99.1)
+Requires-Dist: pdfplumber (>=0.11.6,<0.12.0)
 Requires-Dist: playwright (>=1.53.0,<2.0.0) ; extra == "playwright" or extra == "all"
-Requires-Dist: prompt-toolkit (>=3.0.51,<4.0.0)
 Requires-Dist: psutil (>=7.0.0,<8.0.0)
-Requires-Dist: pydantic-ai (>=0.4.5,<0.5.0)
+Requires-Dist: pydantic-ai-slim[anthropic,bedrock,cli,cohere,google,groq,huggingface,mcp,mistral,openai,vertexai] (>=0.6.2,<0.7.0)
 Requires-Dist: pyjwt (>=2.10.1,<3.0.0)
 Requires-Dist: python-dotenv (>=1.1.1,<2.0.0)
 Requires-Dist: python-jose[cryptography] (>=3.4.0,<4.0.0)
 Requires-Dist: requests (>=2.32.4,<3.0.0)
-Requires-Dist: rich (>=14.0.0,<15.0.0)
 Requires-Dist: tiktoken (>=0.8.0,<0.9.0)
 Requires-Dist: ulid-py (>=1.1.0,<2.0.0)
 Project-URL: Documentation, https://github.com/state-alchemists/zrb

{zrb-1.14.5 → zrb-1.14.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "zrb"
-version = "1.14.5"
+version = "1.14.6"
 description = "Your Automation Powerhouse"
 authors = ["Go Frendi Gunawan <gofrendiasgard@gmail.com>"]
 license = "AGPL-3.0-or-later"
@@ -43,26 +43,31 @@ black = "~25.1.0"
 isort = "~6.0.1"
 requests = "^2.32.4"
 libcst = "^1.7.0"
-chromadb = {version = "^0.6.3", optional = true}
-pdfplumber = {version = "^0.11.6", optional = true}
 beautifulsoup4 = "^4.13.3"
+pdfplumber = "^0.11.6"
 fastapi = {extras = ["standard"], version = "^0.116.1"}
 python-jose = {extras = ["cryptography"], version = "^3.4.0"}
 ulid-py = "^1.1.0"
-pydantic-ai = "^0.4.5"
-tiktoken = "^0.8.0"
-openai = {version = "^1.86.0", optional = true}
-playwright = {version = "^1.53.0", optional = true}
 psutil = "^7.0.0"
 pyjwt = "^2.10.1"
-prompt-toolkit = "^3.0.51"
-rich = "^14.0.0"
+
+# NOTE: Look at https://github.com/pydantic/pydantic-ai/blob/main/pydantic_ai_slim/pyproject.toml
+# - We want to include all models, as well as mcp capability.
+# - We also need prompt-toolkit and rich which are included in pydantic ai's cli optional dependency
+pydantic-ai-slim = {extras = ["openai", "cohere", "vertexai", "google", "anthropic", "groq", "mistral", "bedrock", "huggingface", "cli", "mcp"], version = "^0.6.2"} 
+
+# NOTE: On Pydantic AI 0.6.2, The last working version of OpenAI is 0.99.1
+openai = ">=1.92.0, <=1.99.1"
+
+tiktoken = "^0.8.0"
+playwright = {version = "^1.53.0", optional = true}
+chromadb = {version = "^0.6.3", optional = true}
 
 [tool.poetry.extras]
 # poetry install -E rag
-rag = ["chromadb", "pdfplumber", "openai"]
+rag = ["chromadb"]
 playwright = ["playwright"]
-all = ["chromadb", "pdfplumber", "openai", "playwright"]
+all = ["chromadb", "playwright"]
 
 [tool.poetry.group.dev.dependencies]
 alembic = "^1.16.2" # FastApp dependencies

{zrb-1.14.5 → zrb-1.14.6}/src/zrb/builtin/llm/tool/sub_agent.py

@@ -15,8 +15,6 @@ if TYPE_CHECKING:
     from pydantic_ai.toolsets import AbstractToolset
 
     ToolOrCallable = Tool | Callable
-else:
-    ToolOrCallable = Any
 
 
 def create_sub_agent_tool(
@@ -25,7 +23,7 @@ def create_sub_agent_tool(
     system_prompt: str | None = None,
     model: "str | Model | None" = None,
     model_settings: "ModelSettings | None" = None,
-    tools: list[ToolOrCallable] = [],
+    tools: "list[ToolOrCallable]" = [],
     toolsets: list["AbstractToolset[Agent]"] = [],
     is_yolo_mode: bool | None = None,
 ) -> Callable[[AnyContext, str], Coroutine[Any, Any, str]]:
@@ -97,7 +95,8 @@ def create_sub_agent_tool(
             ctx=ctx,
             agent=sub_agent_agent,
             user_prompt=query,
-            history_list=[],  # Start with empty history for the sub-agent
+            attachments=[],
+            history_list=[],
         )
 
         # Return the sub-agent's final message content

{zrb-1.14.5 → zrb-1.14.6}/src/zrb/builtin/llm/tool/web.py

@@ -1,64 +1,25 @@
 import json
 from collections.abc import Callable
+from urllib.parse import urljoin
+
+from bs4 import BeautifulSoup
 
 
 async def open_web_page(url: str) -> str:
     """
-    Fetches and parses the textual content of a given web page URL.
+    Fetches, parses, and converts the content of a web page to Markdown.
 
-    Use this tool to "read" a web page. It strips away HTML tags, scripts, and other non-textual elements to provide the clean text content. It also extracts any hyperlinks found on the page. This is useful when you need to understand the content of a specific URL that you have discovered through a search or from another source.
+    This tool "reads" a web page by fetching its content, stripping away non-essential elements like scripts and styles, and then converting the cleaned HTML into Markdown format. This preserves the semantic structure of the content (headings, lists, etc.) while removing clutter. It also extracts all hyperlinks and resolves them to absolute URLs.
 
     Args:
         url (str): The full URL of the web page to open (e.g., "https://example.com/article").
 
     Returns:
-        str: A JSON object containing the cleaned text `content` of the page and a list of `links_on_page`.
+        str: A JSON object containing the page's content in Markdown format and a list of all absolute links found on the page.
     """
-
-    async def get_page_content(page_url: str):
-        user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"  # noqa
-        try:
-            from playwright.async_api import async_playwright
-
-            async with async_playwright() as p:
-                browser = await p.chromium.launch(headless=True)
-                page = await browser.new_page()
-                await page.set_extra_http_headers({"User-Agent": user_agent})
-                try:
-                    # Navigate to the URL with a timeout of 30 seconds
-                    await page.goto(page_url, wait_until="networkidle", timeout=30000)
-                    # Wait for the content to load
-                    await page.wait_for_load_state("domcontentloaded")
-                    # Get the page content
-                    content = await page.content()
-                    # Extract all links from the page
-                    links = await page.eval_on_selector_all(
-                        "a[href]",
-                        """
-                        (elements) => elements.map(el => {
-                            const href = el.getAttribute('href');
-                            if (href && !href.startsWith('#') && !href.startsWith('/')) {
-                                return href;
-                            }
-                            return null;
-                        }).filter(href => href !== null)
-                    """,
-                    )
-                    return {"content": content, "links_on_page": links}
-                finally:
-                    await browser.close()
-        except BaseException:
-            import requests
-
-            response = requests.get(url, headers={"User-Agent": user_agent})
-            if response.status_code != 200:
-                msg = f"Unable to retrieve search results. Status code: {response.status_code}"
-                raise Exception(msg)
-            return {"content": response.text, "links_on_page": []}
-
-    result = await get_page_content(url)
-    # Parse the HTML content
-    return json.dumps(parse_html_text(result["content"]))
+    html_content, links = await _fetch_page_content(url)
+    markdown_content = _convert_html_to_markdown(html_content)
+    return json.dumps({"content": markdown_content, "links_on_page": links})
 
 
 def create_search_internet_tool(serp_api_key: str) -> Callable[[str, int], str]:
@@ -85,14 +46,14 @@ def create_search_internet_tool(serp_api_key: str) -> Callable[[str, int], str]:
             num_results (int, optional): The desired number of search results. Defaults to 10.
 
         Returns:
-            str: A JSON object containing the parsed text content from the search results page.
+            str: A formatted string summarizing the search results, including titles, links, and snippets.
         """
         import requests
 
         response = requests.get(
             "https://serpapi.com/search",
             headers={
-                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"  # noqa
+                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
             },
             params={
                 "q": query,
@@ -104,9 +65,9 @@ def create_search_internet_tool(serp_api_key: str) -> Callable[[str, int], str]:
         )
         if response.status_code != 200:
             raise Exception(
-                f"Error: Unable to retrieve search results (status code: {response.status_code})"  # noqa
+                f"Error: Unable to retrieve search results (status code: {response.status_code})"
             )
-        return json.dumps(parse_html_text(response.text))
+        return response.json()
 
     return search_internet
 
@@ -150,30 +111,67 @@ def search_arxiv(query: str, num_results: int = 10) -> str:
     return response.content
 
 
-def parse_html_text(html_text: str) -> dict[str, str]:
-    from bs4 import BeautifulSoup
-
-    ignored_tags = [
-        "script",
-        "link",
-        "meta",
-        "style",
-        "code",
-        "footer",
-        "nav",
-        "header",
-        "aside",
-    ]
+async def _fetch_page_content(url: str) -> tuple[str, list[str]]:
+    """Fetches the HTML content and all absolute links from a URL."""
+    user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
+    try:
+        from playwright.async_api import async_playwright
+
+        async with async_playwright() as p:
+            browser = await p.chromium.launch(headless=True)
+            page = await browser.new_page()
+            await page.set_extra_http_headers({"User-Agent": user_agent})
+            try:
+                await page.goto(url, wait_until="networkidle", timeout=30000)
+                await page.wait_for_load_state("domcontentloaded")
+                content = await page.content()
+                links = await page.eval_on_selector_all(
+                    "a[href]",
+                    """
+                    (elements, baseUrl) => elements.map(el => {
+                        const href = el.getAttribute('href');
+                        if (!href || href.startsWith('#')) return null;
+                        try {
+                            return new URL(href, baseUrl).href;
+                        } catch (e) {
+                            return null;
+                        }
+                    }).filter(href => href !== null)
+                    """,
+                    url,
+                )
+                return content, links
+                # return json.dumps({"content": content, "links": links})
+            finally:
+                await browser.close()
+    except Exception:
+        import requests
+
+        response = requests.get(url, headers={"User-Agent": user_agent})
+        if response.status_code != 200:
+            raise Exception(
+                f"Unable to retrieve page content. Status code: {response.status_code}"
+            )
+        content = response.text
+        soup = BeautifulSoup(content, "html.parser")
+        links = [
+            urljoin(url, a["href"])
+            for a in soup.find_all("a", href=True)
+            if not a["href"].startswith("#")
+        ]
+        return content, links
+        # return json.dumps({"content": content, "links": links})
+
+
+def _convert_html_to_markdown(html_text: str) -> str:
+    """Converts HTML content to a clean Markdown representation."""
+    from markdownify import markdownify as md
+
     soup = BeautifulSoup(html_text, "html.parser")
-    links = []
-    for anchor in soup.find_all("a"):
-        if not anchor or "href" not in anchor.attrs:
-            continue
-        link: str = anchor["href"]
-        if link.startswith("#") or link.startswith("/"):
-            continue
-        links.append(link)
-    for tag in soup(ignored_tags):
+    # Remove non-content tags
+    for tag in soup(
+        ["script", "link", "meta", "style", "header", "footer", "nav", "aside"]
+    ):
         tag.decompose()
-    html_text = soup.get_text(separator=" ", strip=True)
-    return {"content": html_text, "links_on_page": links}
+    # Convert the cleaned HTML to Markdown
+    return md(str(soup))

{zrb-1.14.5 → zrb-1.14.6}/src/zrb/builtin/setup/latex/ubuntu.py

@@ -12,6 +12,7 @@ setup_latex_on_ubuntu = setup_latex_group.add_task(
             "texlive-fonts-extra texlive-latex-extra",
         ],
         render_cmd=False,
+        is_interactive=True,
     ),
     alias="ubuntu",
 )

{zrb-1.14.5 → zrb-1.14.6}/src/zrb/builtin/setup/ubuntu.py

@@ -22,7 +22,7 @@ setup_ubuntu = setup_group.add_task(
         description="🐧 Setup ubuntu",
         cmd=[
             "sudo apt install -y \\",
-            "build-essential python3-distutils libssl-dev zlib1g-dev \\"
+            "build-essential libssl-dev zlib1g-dev \\"
             "libbz2-dev libreadline-dev libsqlite3-dev libpq-dev python3-dev \\",
             "llvm libncurses5-dev libncursesw5-dev xz-utils tk-dev libffi-dev \\",
             "liblzma-dev python3-openssl libblas-dev liblapack-dev rustc \\",