| import os |
| import requests |
| from bs4 import BeautifulSoup |
| from duckduckgo_search import DDGS |
|
|
| class SimpleTextBrowser: |
| def __init__( |
| self, |
| timeout: int = 10, |
| user_agent: str | None = None, |
| viewport_size: int = 5000, |
| downloads_folder: str = "downloads", |
| ): |
| self.session = requests.Session() |
| self.session.headers.update({"User-Agent": user_agent or "Mozilla/5.0"}) |
| self.timeout = timeout |
| self.viewport_size = viewport_size |
| os.makedirs(downloads_folder, exist_ok=True) |
| self.pages: list[str] = [""] |
| self._idx = 0 |
|
|
| def _split(self, text: str): |
| self.pages = [ |
| text[i : i + self.viewport_size] |
| for i in range(0, len(text), self.viewport_size) |
| ] |
| self._idx = 0 |
|
|
| def visit_page(self, url: str) -> str: |
| """ |
| Internal: fetch and split a page into viewports. |
| """ |
| resp = self.session.get(url, timeout=self.timeout) |
| resp.raise_for_status() |
| text = resp.text |
| if "text/html" in resp.headers.get("content-type", ""): |
| text = BeautifulSoup(text, "lxml").get_text("\n") |
| self._split(text) |
| return self.pages[self._idx] |
|
|
| def search(self, query: str) -> str: |
| """ |
| Internal: DuckDuckGo search → snippets. |
| """ |
| results = list(DDGS().text(query, timelimit=None)) |
| hits = "\n".join( |
| f"{i+1}. {h['title']} — {h.get('snippet','')}" |
| for i, h in enumerate(results) |
| ) |
| full = f"Search results for '{query}':\n\n" + hits |
| self._split(full) |
| return self.pages[self._idx] |
|
|
| def page_down(self) -> str: |
| self._idx = min(self._idx + 1, len(self.pages) - 1) |
| return self.pages[self._idx] |
|
|
| def page_up(self) -> str: |
| self._idx = max(self._idx - 1, 0) |
| return self.pages[self._idx] |
|
|
| def find_on_page(self, search_string: str) -> str: |
| full = "\n".join(self.pages).lower() |
| pos = full.find(search_string.lower()) |
| if pos < 0: |
| return f"'{search_string}' not found." |
| start = max(0, pos - 100) |
| end = min(len(full), pos + 100) |
| return full[start:end] |
|
|
| def find_next(self) -> str: |
| return "Use `find_on_page_ctrl_f` again with a new string to find further occurrences." |
|
|
| |
| _BROWSER = SimpleTextBrowser() |
|
|
