hishel 0.1.4__py3-none-any.whl → 1.0.0b1__py3-none-any.whl

hishel/_utils.py

@@ -1,113 +1,17 @@
 from __future__ import annotations
 
 import calendar
-import hashlib
-import json
-import sqlite3
 import time
 import typing as tp
-from datetime import date
-from email.utils import parsedate_tz
-from typing import Any, AsyncIterator, Generator, Iterable, Iterator, TypeVar
-
-import anyio
-import anysqlite
-import httpcore
-import httpx
-from anyio import from_thread, to_thread
+from email.utils import formatdate, parsedate_tz
+from pathlib import Path
+from typing import AsyncIterator, Iterable, Iterator
 
 HEADERS_ENCODING = "iso-8859-1"
 
 T = tp.TypeVar("T")
 
 
-class BaseClock:
-    def now(self) -> int:
-        raise NotImplementedError()
-
-
-class Clock(BaseClock):
-    def now(self) -> int:
-        return int(time.time())
-
-
-def normalized_url(url: tp.Union[httpcore.URL, str, bytes]) -> str:
-    if isinstance(url, str):  # pragma: no cover
-        return url
-
-    if isinstance(url, bytes):  # pragma: no cover
-        return url.decode("ascii")
-
-    if isinstance(url, httpcore.URL):
-        port = f":{url.port}" if url.port is not None else ""
-        return f"{url.scheme.decode('ascii')}://{url.host.decode('ascii')}{port}{url.target.decode('ascii')}"
-    assert False, "Invalid type for `normalized_url`"  # pragma: no cover
-
-
-def get_safe_url(url: httpcore.URL) -> str:
-    httpx_url = httpx.URL(bytes(url).decode("ascii"))
-
-    schema = httpx_url.scheme
-    host = httpx_url.host
-    path = httpx_url.path
-
-    return f"{schema}://{host}{path}"
-
-
-def generate_key(request: httpcore.Request, body: bytes = b"") -> str:
-    encoded_url = normalized_url(request.url).encode("ascii")
-
-    key_parts = [request.method, encoded_url, body]
-
-    # FIPs mode disables blake2 algorithm, use sha256 instead when not found.
-    blake2b_hasher = None
-    sha256_hasher = hashlib.sha256(usedforsecurity=False)
-    try:
-        blake2b_hasher = hashlib.blake2b(digest_size=16, usedforsecurity=False)
-    except (ValueError, TypeError, AttributeError):
-        pass
-
-    hexdigest: str
-    if blake2b_hasher:
-        for part in key_parts:
-            blake2b_hasher.update(part)
-
-        hexdigest = blake2b_hasher.hexdigest()
-    else:
-        for part in key_parts:
-            sha256_hasher.update(part)
-
-        hexdigest = sha256_hasher.hexdigest()
-    return hexdigest
-
-
-def extract_header_values(
-    headers: tp.List[tp.Tuple[bytes, bytes]],
-    header_key: tp.Union[bytes, str],
-    single: bool = False,
-) -> tp.List[bytes]:
-    if isinstance(header_key, str):
-        header_key = header_key.encode(HEADERS_ENCODING)
-    extracted_headers = []
-    for key, value in headers:
-        if key.lower() == header_key.lower():
-            extracted_headers.append(value)
-            if single:
-                break
-    return extracted_headers
-
-
-def extract_header_values_decoded(
-    headers: tp.List[tp.Tuple[bytes, bytes]], header_key: bytes, single: bool = False
-) -> tp.List[str]:
-    values = extract_header_values(headers=headers, header_key=header_key, single=single)
-    return [value.decode(HEADERS_ENCODING) for value in values]
-
-
-def header_presents(headers: tp.List[tp.Tuple[bytes, bytes]], header_key: bytes) -> bool:
-    return bool(extract_header_values(headers, header_key, single=True))
-
-
 def parse_date(date: str) -> tp.Optional[int]:
     expires = parsedate_tz(date)
     if expires is None:
@@ -116,18 +20,10 @@ def parse_date(date: str) -> tp.Optional[int]:
     return timestamp
 
 
-async def asleep(seconds: tp.Union[int, float]) -> None:
-    await anyio.sleep(seconds)
-
-
 def sleep(seconds: tp.Union[int, float]) -> None:
     time.sleep(seconds)
 
 
-def float_seconds_to_int_milliseconds(seconds: float) -> int:
-    return int(seconds * 1000)
-
-
 def partition(iterable: tp.Iterable[T], predicate: tp.Callable[[T], bool]) -> tp.Tuple[tp.List[T], tp.List[T]]:
     """
     Partition an iterable into two lists: one for matching items and one for non-matching items.
@@ -155,69 +51,33 @@ def partition(iterable: tp.Iterable[T], predicate: tp.Callable[[T], bool]) -> tp
     return matching, non_matching
 
 
-def async_iterator_to_sync(iterator: AsyncIterator[bytes]) -> Iterator[bytes]:
-    """
-    Convert an asynchronous byte iterator to a synchronous one.
-    This function takes an asynchronous iterator that yields bytes and converts it into
-    a synchronous iterator.
-
-    Args:
-        iterator (AsyncIterator[bytes]): The asynchronous byte iterator to be converted.
-    Returns:
-        Iterator[bytes]: A synchronous iterator that yields the same byte chunks as the input iterator.
-    Example:
-        ```python
-        async_iter = some_async_byte_stream()
-        sync_iter = async_iterator_to_sync(async_iter)
-        for chunk in sync_iter:
-            process_bytes(chunk)
-        ```
-    """
-
-    while True:
-        try:
-            chunk = from_thread.run(iterator.__anext__)
-        except StopAsyncIteration:
-            break
-        yield chunk
-
-
-def _call_next(iterator: Iterator[bytes]) -> bytes:
-    try:
-        return iterator.__next__()
-    except StopIteration:
-        raise StopAsyncIteration
+async def make_async_iterator(
+    iterable: Iterable[bytes],
+) -> AsyncIterator[bytes]:
+    for item in iterable:
+        yield item
 
 
-async def sync_iterator_to_async(iterator: Iterator[bytes]) -> AsyncIterator[bytes]:
-    """
-    Converts a synchronous bytes iterator to an asynchronous one.
-    This function takes a synchronous iterator that yields bytes and converts it into an
-    asynchronous iterator, allowing it to be used in async contexts without blocking.
-    Args:
-        iterator (Iterator[bytes]): A synchronous iterator yielding bytes objects.
-    Returns:
-        AsyncIterator[bytes]: An asynchronous iterator yielding the same bytes objects.
-    Example:
-        ```
-        sync_iter = iter([b'data1', b'data2'])
-        async for chunk in sync_iterator_to_async(sync_iter):
-            await process_chunk(chunk)
-        ```
+def filter_mapping(mapping: tp.Mapping[str, T], keys_to_exclude: tp.Iterable[str]) -> tp.Dict[str, T]:
     """
+        Filter out specified keys from a string-keyed mapping using case-insensitive comparison.
 
-    while True:
-        try:
-            chunk = await to_thread.run_sync(_call_next, iterator)
-        except StopAsyncIteration:
-            break
+        Args:
+            mapping: The input mapping with string keys to filter.
+            keys_to_exclude: An iterable of string keys to exclude (case-insensitive).
 
-        yield chunk
+        Returns:
+            A new dictionary with the specified keys excluded.
 
-
-async def make_async_iterator(iterable: Iterable[bytes]) -> AsyncIterator[bytes]:
-    for item in iterable:
-        yield item
+        Example:
+    ```python
+            original = {'a': 1, 'B': 2, 'c': 3}
+            filtered = filter_mapping(original, ['b'])
+            # filtered will be {'a': 1, 'c': 3}
+    ```
+    """
+    exclude_set = {k.lower() for k in keys_to_exclude}
+    return {k: v for k, v in mapping.items() if k.lower() not in exclude_set}
 
 
 def make_sync_iterator(iterable: Iterable[bytes]) -> Iterator[bytes]:
@@ -247,212 +107,23 @@ def snake_to_header(text: str) -> str:
     return "X-" + "-".join(word.capitalize() for word in text.split("_"))
 
 
-_T = TypeVar("_T")
-
-
-class GeneratorWithReturnValue:
-    def __init__(
-        self, gen: Generator[None, bytes | None, bytes], stream: AsyncIterator[bytes] | Iterator[bytes]
-    ) -> None:
-        self.gen = gen
-        self.stream = stream
-        self.value: bytes | None = None
-
-    def __iter__(self) -> Iterator[bytes]:
-        return self
+def ensure_cache_dict(base_path: str | None = None) -> Path:
+    _base_path = Path(base_path) if base_path is not None else Path(".cache/hishel")
+    _gitignore_file = _base_path / ".gitignore"
 
-    def __next__(self) -> bytes:
-        assert isinstance(self.stream, Iterator)
+    _base_path.mkdir(parents=True, exist_ok=True)
 
-        try:
-            chunk = next(self.stream)
-            self.gen.send(chunk)
-        except StopIteration as exc:
-            self.gen.send(None)
-            self.value = exc.value
-            raise
-        return chunk
+    if not _gitignore_file.is_file():
+        with open(_gitignore_file, "w", encoding="utf-8") as f:
+            f.write("# Automatically created by Hishel\n*")
+    return _base_path
 
-    def __aiter__(self) -> AsyncIterator[bytes]:
-        return self
 
-    async def __anext__(self) -> bytes:
-        assert isinstance(self.stream, AsyncIterator)
-        try:
-            chunk = await self.stream.__anext__()
-            self.gen.send(chunk)
-        except StopIteration as exc:
-            self.gen.send(None)
-            self.value = exc.value
-            raise
-        return chunk
-
-
-def print_sqlite_state(conn: sqlite3.Connection) -> str:
+def generate_http_date() -> str:
     """
-    Print all tables and their rows in a pretty format suitable for inline snapshots.
+    Generate a Date header value for HTTP responses.
+    Returns date in RFC 1123 format (required by HTTP/1.1).
 
-    Args:
-        conn: SQLite database connection
-
-    Returns:
-        Formatted string representation of the database state
-    """
-    cursor = conn.cursor()
-
-    # Get all table names
-    cursor.execute("SELECT name FROM sqlite_master WHERE type='table' ORDER BY name")
-    tables = [row[0] for row in cursor.fetchall()]
-
-    output_lines = []
-    output_lines.append("=" * 80)
-    output_lines.append("DATABASE SNAPSHOT")
-    output_lines.append("=" * 80)
-
-    for table_name in tables:
-        # Get column information
-        cursor.execute(f"PRAGMA table_info({table_name})")
-        columns = cursor.fetchall()
-        column_names = [col[1] for col in columns]
-        column_types = {col[1]: col[2] for col in columns}
-
-        # Get all rows
-        cursor.execute(f"SELECT * FROM {table_name}")
-        rows = cursor.fetchall()
-
-        output_lines.append("")
-        output_lines.append(f"TABLE: {table_name}")
-        output_lines.append("-" * 80)
-        output_lines.append(f"Rows: {len(rows)}")
-        output_lines.append("")
-
-        if not rows:
-            output_lines.append("  (empty)")
-            continue
-
-        # Format each row
-        for idx, row in enumerate(rows, 1):
-            output_lines.append(f"  Row {idx}:")
-
-            for col_name, value in zip(column_names, row):
-                col_type = column_types[col_name]
-                formatted_value = format_value(value, col_name, col_type)
-                output_lines.append(f"    {col_name:15} = {formatted_value}")
-
-            if idx < len(rows):
-                output_lines.append("")
-
-    output_lines.append("")
-    output_lines.append("=" * 80)
-
-    result = "\n".join(output_lines)
-    return result
-
-
-async def aprint_sqlite_state(conn: anysqlite.Connection) -> str:
-    """
-    Print all tables and their rows in a pretty format suitable for inline snapshots.
-
-    Args:
-        conn: SQLite database connection
-
-    Returns:
-        Formatted string representation of the database state
+    Example output: 'Sun, 26 Oct 2025 12:34:56 GMT'
     """
-    cursor = await conn.cursor()
-
-    # Get all table names
-    await cursor.execute("SELECT name FROM sqlite_master WHERE type='table' ORDER BY name")
-    tables = [row[0] for row in await cursor.fetchall()]
-
-    output_lines = []
-    output_lines.append("=" * 80)
-    output_lines.append("DATABASE SNAPSHOT")
-    output_lines.append("=" * 80)
-
-    for table_name in tables:
-        # Get column information
-        await cursor.execute(f"PRAGMA table_info({table_name})")
-        columns = await cursor.fetchall()
-        column_names = [col[1] for col in columns]
-        column_types = {col[1]: col[2] for col in columns}
-
-        # Get all rows
-        await cursor.execute(f"SELECT * FROM {table_name}")
-        rows = await cursor.fetchall()
-
-        output_lines.append("")
-        output_lines.append(f"TABLE: {table_name}")
-        output_lines.append("-" * 80)
-        output_lines.append(f"Rows: {len(rows)}")
-        output_lines.append("")
-
-        if not rows:
-            output_lines.append("  (empty)")
-            continue
-
-        # Format each row
-        for idx, row in enumerate(rows, 1):
-            output_lines.append(f"  Row {idx}:")
-
-            for col_name, value in zip(column_names, row):
-                col_type = column_types[col_name]
-                formatted_value = format_value(value, col_name, col_type)
-                output_lines.append(f"    {col_name:15} = {formatted_value}")
-
-            if idx < len(rows):
-                output_lines.append("")
-
-    output_lines.append("")
-    output_lines.append("=" * 80)
-
-    result = "\n".join(output_lines)
-    return result
-
-
-def format_value(value: Any, col_name: str, col_type: str) -> str:
-    """Format a value for display based on its type and column name."""
-
-    if value is None:
-        return "NULL"
-
-    # Handle BLOB columns
-    if col_type.upper() == "BLOB":
-        if isinstance(value, bytes):
-            # Try to decode as UTF-8 string first
-            try:
-                decoded = value.decode("utf-8")
-                # Check if it looks like JSON
-                if decoded.strip().startswith("{") or decoded.strip().startswith("["):
-                    try:
-                        parsed = json.loads(decoded)
-                        return f"(JSON) {json.dumps(parsed, indent=2)}"
-                    except json.JSONDecodeError:
-                        pass
-                # Show string if it's printable
-                if all(32 <= ord(c) <= 126 or c in "\n\r\t" for c in decoded):
-                    return f"(str) '{decoded}'"
-            except UnicodeDecodeError:
-                pass
-
-            # Show hex representation for binary data
-            hex_str = value.hex()
-            if len(hex_str) > 64:
-                return f"(bytes) 0x{hex_str[:60]}... ({len(value)} bytes)"
-            return f"(bytes) 0x{hex_str} ({len(value)} bytes)"
-        return repr(value)
-
-    # Handle timestamps - ONLY show date, not the raw timestamp
-    if col_name.endswith("_at") and isinstance(value, (int, float)):
-        try:
-            dt = date.fromtimestamp(value)
-            return dt.isoformat()  # Changed: removed the timestamp prefix
-        except (ValueError, OSError):
-            return str(value)
-
-    # Handle TEXT columns
-    if col_type.upper() == "TEXT":
-        return f"'{value}'"
-
-    # Handle other types
-    return str(value)
+    return formatdate(timeval=None, localtime=False, usegmt=True)