Create Relay Bot MVP

app/attachments/extractors.py

@@ -0,0 +1,24 @@
+import csv
+import io
+from pathlib import Path
+
+from docx import Document
+from pypdf import PdfReader
+
+SUPPORTED_EXTENSIONS = {".txt", ".md", ".json", ".yaml", ".yml", ".csv", ".pdf", ".docx"}
+
+
+def extract_text(filename: str, content: bytes, max_chars: int = 60_000) -> str:
+    suffix = Path(filename).suffix.lower()
+    if suffix not in SUPPORTED_EXTENSIONS:
+        raise ValueError("Этот тип файла пока не поддерживается")
+    if suffix in {".txt", ".md", ".json", ".yaml", ".yml"}:
+        text = content.decode("utf-8", errors="replace")
+    elif suffix == ".csv":
+        rows = csv.reader(io.StringIO(content.decode("utf-8", errors="replace")))
+        text = "\n".join(" | ".join(row) for row in rows)
+    elif suffix == ".pdf":
+        text = "\n".join(page.extract_text() or "" for page in PdfReader(io.BytesIO(content)).pages)
+    else:
+        text = "\n".join(p.text for p in Document(io.BytesIO(content)).paragraphs)
+    return text[:max_chars]

app/attachments/service.py

@@ -0,0 +1,30 @@
+from pathlib import Path
+
+from app.attachments.extractors import SUPPORTED_EXTENSIONS, extract_text
+
+
+def explicitly_requested(
+    *,
+    is_private: bool,
+    caption: str | None,
+    replied_with_request: bool,
+    command: bool,
+    bot_username: str | None,
+) -> bool:
+    mention = bool(caption and bot_username and f"@{bot_username.lower()}" in caption.lower())
+    return is_private or mention or replied_with_request or command
+
+
+class AttachmentService:
+    def __init__(self, max_bytes: int):
+        self.max_bytes = max_bytes
+
+    def validate(self, filename: str, size: int) -> None:
+        if size > self.max_bytes:
+            raise ValueError("Файл слишком большой")
+        if Path(filename).suffix.lower() not in SUPPORTED_EXTENSIONS:
+            raise ValueError("Неподдерживаемый или небезопасный файл")
+
+    def extract(self, filename: str, content: bytes) -> str:
+        self.validate(filename, len(content))
+        return extract_text(filename, content)