Observed memory leak in ssl library

[HenriBlacksmith]

Bug report

Bug description:

I already opened a bug on urllib3 for this issue, it looks that may application using MSAL for Python library (which relies on requests , which relies on urllib3, which relies on ssl) is leaking memory when being upgraded from Python 3.13 to Python 3.14 (I tested 3.14.{0, 1, 2})

The details are explained in the urllib3 ticket including outputs from memray and it points to ssl package (precisely ssl.SSLContext.load_verify_locations), see: https://github.com/urllib3/urllib3/issues/3738

CPython versions tested on:

3.14

Operating systems tested on:

Linux

[vstinner]

The details are explained in the urllib3 ticket including outputs from memray and it points to ssl package (precisely ssl.SSLContext.load_verify_locations), see: https://github.com/urllib3/urllib3/issues/3738

Please provide a reproducer, if possible using only the ssl module.

[zweizeichen]

No reproducer but if it helps: very similar situation here when connecting to valkey with a custom CA cert.

[zweizeichen]

I did some more testing to come up with a minimal repro: seems like rss just keeps growing while the heap stays somewhat constant. I was unable to reproduce it using a simple sync case locally and on the cluster and this started to smell an awful lot like https://github.com/python/cpython/issues/109534 which might be async-related. Like @HenriBlacksmith's case we're also seeing this in an ASGI app on k8s using the debian trixie base image since the 3.14 release.

I'll attempt to redeploy this glibc-less and report back.

Update: the alleged leak is still there even in an alpine container

[zweizeichen]

Potentially related: https://github.com/python/cpython/issues/84904

[HenriBlacksmith]

I applied a fix that improved the memory situation on my other FastAPI app by reducing the calls generated by MSAL.

I applied a few dependency upgrades on top including a bump of the Python image from 3.14.0-slim to 3.14.2-slim and it caused the memory leak to be worse.

Though I also need to collect detailed figures about this one as urllib3 version is different in this app because of some version constraints on a library.

I will try to get more info from memray on this one.

[HenriBlacksmith]

I was able to export new memray profiles from my other application.

This time the leak is coming from httpx which also relies on ssl package under the hood:

Sorry for the cuts on the capture, I am trying to "anonymize" things

[zweizeichen]

Thank you! I'm not 100% sure how to interpret these memray outputs to be honest, because if I understand this correctly this shows us where allocations do happen but not necessarily where the leak is (it may be correlated though). This is useful for finding out 'which objects eat up my memory?'. Technically speaking memray will only show us the allocations/leaks the python allocator knows about, but the leak kind of escapes the realm of the allocator by definition. Another layer which might be adding to this is that this might actually be fragmentation instead of leaks. I hope I ruled this out more or less by checking it's not glibc-related by also observing this behaviour on an alpine container, feel free to double-check.

I'm going to try to find some time to give the minimal (hopefully stdlib only) repro another go and explore analyzing a full heap dump.

[HenriBlacksmith]

I am trying to use valgrind instead of memray to get more details (but I am a bit out of my comfort zone to be honest), it does not report much until k8s restarts the container.

The command looks like:

valgrind --tool=memcheck --log-file=/tmp/valgrind.log --trace-children=yes --leak-check=yes --verbose
              python -m uvicorn app.main:shell --host 0.0.0.0 --log-config logging.yaml

But nothing really comes out of it

[HenriBlacksmith]

Another layer which might be adding to this is that this might actually be fragmentation instead of leaks. I hope I ruled this out more or less by checking it's not glibc-related by also observing this behaviour on an alpine container, feel free to double-check.

I just built an Alpine container and ran it and the behavior is similar (based on python:3.14.2-alpine3.23)

I am running out of ideas, if anyone has guidance to help with investigating this, I would be really happy to test those, if I have extra time, I will try to build a MRO but I am not sure how it can be done

[HenriBlacksmith]

Good news, I might have a MRE, it is using httpx but that's the best I was able to do, I will share the whole code in a repo, including commands to export the graphs.

Here is the code:

import httpx
import asyncio

async def _main() -> None:
    while(1):
        async with httpx.AsyncClient() as client:
            await client.request("GET", "https://example.com/")
        await asyncio.sleep(1)

if __name__ == "__main__":
    asyncio.run(_main())

FROM astral/uv:0.9.17-python3.14-bookworm AS builder
ENV UV_COMPILE_BYTECODE=1 UV_LINK_MODE=copy UV_PYTHON_DOWNLOADS=0

# Create and set the working directory
WORKDIR /mro-app

# Install runtime dependencies
RUN --mount=type=cache,target=/root/.cache/uv \
    --mount=type=bind,source=uv.lock,target=uv.lock \
    --mount=type=bind,source=pyproject.toml,target=pyproject.toml \
    uv sync --frozen --no-install-project --no-dev

# Copy the application code (excluding files in .dockerignore)
COPY . .
RUN --mount=type=cache,target=/root/.cache/uv \
        uv sync --frozen --no-dev

# Use an official lightweight Python image with slim variant
FROM python:3.14.2-slim

# Set environment variables
# See: https://docs.python.org/3/using/cmdline.html#envvar-PYTHONUNBUFFERED
ENV PYTHONUNBUFFERED=1

# Create and set the working directory
WORKDIR /mro-app

# Expose the port the app runs on
EXPOSE 8000

# Copy the application from the builder
COPY --from=builder /mro-app /mro-app

# Place executables in the environment at the front of the path
ENV PATH="/mro-app/.venv/bin:$PATH"

# Define the default command to run when the container starts
CMD ["memray", "run", "--native", "--output", "/tmp/capture.bin", "main.py"]

[project]
name = "mro"
version = "0.1.0"
description = "Add your description here"
readme = "README.md"
requires-python = ">=3.14"
dependencies = [
    "httpx>=0.28.1",
    "memray>=1.19.1",
    "urllib3>=2.6.2",
]

etc. will build a repo asap with all the other files

[HenriBlacksmith]

Here is the repo, I hope it is an actual valid MRE and not just "normal" usage: https://github.com/HenriBlacksmith/python-ssl-memory-usage-mro/tree/main

[HenriBlacksmith]

And I was finally able to get the native symbols:

I will try to compare with older Python versions to see if the sensitivity to Python version holds with the MRE.

I was able to reproduce on my Mac in a non-Docker environment.

[HenriBlacksmith]

I tested the same program under Python 3.13.11+debug and here is the memory graph:

The pattern shows something like GC is happening

I updated the MRE repo to reflect my tests, no need to use Docker, just running those two programs on my system shows differences

[vstinner]

Do you also see the leak on Python 3.14 if you call gc.collect() time to time?

[HenriBlacksmith]

Here is the same code with 3.14.2+debug running the same number of iterations (the other graph was on a shorter time frame):

I will see to run with a call to gc.collect() every iteration

import httpx
import asyncio
import gc


async def _main() -> None:
    for i in range(500):
        async with httpx.AsyncClient() as client:
            await client.request("GET", "https://example.com/")
        await asyncio.sleep(1)
        gc.collect()


if __name__ == "__main__":
    asyncio.run(_main())

[HenriBlacksmith]

Here are the results for the code above:

[vstinner]

So if you call explicitly gc.collect(), there is no more leak?

It seems like Python 3.14 GC behaves differently.

cc @nascheme @markshannon

[sergey-miryanov]

I would like to look into this if no one mind.

[vstinner]

Reproducer modified to use www.google.com and display the RSS memory usage at each iteration:

import httpx
import asyncio
import gc
import os


def get_rss():
    with open(f"/proc/{os.getpid()}/status") as fp:
        for line in fp:
            if line.startswith("VmRSS"):
                rss = line.split(":", 1)[1].rstrip()
                rss = rss.removesuffix(" kB")
                return int(rss)
    return None


async def _main() -> None:
    for i in range(500):
        async with httpx.AsyncClient() as client:
            await client.request("GET", "https://www.google.com/")
        await asyncio.sleep(0.2)
        #gc.collect()
        rss = get_rss()
        print(f"Loop #{i}: {rss:,} kB")


if __name__ == "__main__":
    asyncio.run(_main())

Output, the script leaks around 1 MiB per iteration:

Loop #0: 40,272 kB
Loop #1: 41,140 kB
Loop #2: 42,080 kB
Loop #3: 43,008 kB
Loop #4: 44,076 kB
Loop #5: 44,932 kB
(...)
Loop #48: 85,000 kB
Loop #49: 85,912 kB
Loop #50: 86,840 kB
(...)

If I uncomment gc.collect() line, the memory usage is basically stable:

Loop #0: 40,208 kB
Loop #1: 40,992 kB
Loop #2: 41,136 kB
Loop #3: 41,140 kB
Loop #4: 41,148 kB
Loop #5: 41,152 kB
(...)
Loop #48: 41,668 kB
Loop #49: 41,668 kB
Loop #50: 41,668 kB
(...)

[zweizeichen]

Thanks for putting in the time and effort! Unfortunately, I did not really have the time to look into this sooner.

I can confirm sprinkling gc.collect() in strategic locations (non-critical not that hot code paths that get called regularly in long-running processes, e.g. liveness probe endpoint) in the codebase alleviates the issue. It's not pretty but at least a workaround until a potential fix makes it into a 3.14.x release - still better than downgrading to 3.13 :D

[nascheme]

Some additional info: 3.13 memory use stable, 3.14.1 grows, 3.14.2 grows, 3.14.2t grows. I initially suspected this would be fixed by https://github.com/python/cpython/pull/142001 but since it leaks for both the 3.14.1 release and for 3.14.2t, that's not it. It's also should not be the "incremental GC" added to 3.14. Needs additional investigation to find why adding gc.collect() fixes the leak.

[sergey-miryanov]

@nascheme On main (1eddef81930ad9f7e2f411f153c35af16a6edf14) I see that rss only grows, on my gc-fix branch it looks like a saw https://gist.github.com/sergey-miryanov/8e8d11311d154ab2c7656b8e546e64aa

[sergey-miryanov]

Following doesn't leak on main (92243dc62ce10715ab0d9074b23dea5a1bfa9dcc):

import httpx
import psutil
import asyncio
import gc
import os
import urllib.parse


def get_rss():
    p = psutil.Process()
    mem_info = p.memory_info()
    vms = mem_info.vms
    rss = mem_info.rss
    return rss / 1024.0


async def _main() -> None:
    for i in range(120):
        url = urllib.parse.urlsplit('https://www.google.com/')
        if url.scheme == 'https':
            reader, writer = await asyncio.open_connection(
                url.hostname, 443, ssl=True)
        else:
            reader, writer = await asyncio.open_connection(
                url.hostname, 80)

        query = (
            f"HEAD {url.path or '/'} HTTP/1.0\r\n"
            f"Host: {url.hostname}\r\n"
            f"\r\n"
        )

        writer.write(query.encode('latin-1'))
        while True:
            line = await reader.readline()
            if not line:
                break

        # Ignore the body, close the socket
        writer.close()
        await writer.wait_closed()

        await asyncio.sleep(0.1)
        #gc.collect()
        rss = get_rss()
        print(f"Loop #{i}: {rss:,} kB")


if __name__ == "__main__":
    asyncio.run(_main())


Loop #100: 40,100.0 kB
Loop #101: 40,592.0 kB
Loop #102: 40,604.0 kB
Loop #103: 40,612.0 kB
Loop #104: 40,616.0 kB
Loop #105: 40,616.0 kB
Loop #106: 40,088.0 kB
Loop #107: 40,520.0 kB
Loop #108: 40,532.0 kB
Loop #109: 40,596.0 kB
Loop #110: 40,068.0 kB
Loop #111: 40,544.0 kB
Loop #112: 40,128.0 kB
Loop #113: 40,608.0 kB
Loop #114: 40,644.0 kB
Loop #115: 40,088.0 kB
Loop #116: 40,536.0 kB
Loop #117: 40,564.0 kB
Loop #118: 40,072.0 kB
Loop #119: 40,528.0 kB

[sergey-miryanov]

@nascheme Memory grows on 3.13.1 and 3.13.9, but it stabilizes at some point:

Details

(.venv) ➜  3.13.1 python ../../issues/b142516.py  
Loop #0: 34,360.0 kB
Loop #1: 35,152.0 kB
Loop #2: 35,944.0 kB
Loop #3: 36,736.0 kB
Loop #4: 37,792.0 kB
Loop #5: 38,584.0 kB
Loop #6: 39,376.0 kB
Loop #7: 40,168.0 kB
Loop #8: 40,960.0 kB
Loop #9: 41,752.0 kB
Loop #10: 42,808.0 kB
Loop #11: 43,600.0 kB
Loop #12: 44,392.0 kB
Loop #13: 45,184.0 kB
Loop #14: 45,976.0 kB
Loop #15: 46,768.0 kB
Loop #16: 47,824.0 kB
Loop #17: 47,824.0 kB
Loop #18: 47,824.0 kB
Loop #19: 47,824.0 kB
Loop #20: 47,824.0 kB
Loop #21: 47,824.0 kB
Loop #22: 47,824.0 kB
Loop #23: 47,824.0 kB
Loop #24: 47,824.0 kB
Loop #25: 47,824.0 kB
Loop #26: 47,824.0 kB
Loop #27: 47,824.0 kB
Loop #28: 47,824.0 kB
Loop #29: 47,824.0 kB
Loop #30: 47,824.0 kB
Loop #31: 47,824.0 kB
Loop #32: 47,824.0 kB
Loop #33: 48,676.0 kB
Loop #34: 49,468.0 kB
Loop #35: 50,260.0 kB
Loop #36: 51,052.0 kB
Loop #37: 51,844.0 kB
Loop #38: 52,636.0 kB
Loop #39: 53,692.0 kB
Loop #40: 54,484.0 kB
Loop #41: 55,276.0 kB
Loop #42: 56,068.0 kB
Loop #43: 56,860.0 kB
Loop #44: 56,860.0 kB
Loop #45: 56,860.0 kB
Loop #46: 56,860.0 kB
Loop #47: 56,860.0 kB
Loop #48: 56,860.0 kB
Loop #49: 56,860.0 kB
Loop #50: 56,860.0 kB
Loop #51: 56,860.0 kB
Loop #52: 56,860.0 kB
Loop #53: 56,860.0 kB
Loop #54: 56,860.0 kB
Loop #55: 56,860.0 kB
Loop #56: 56,860.0 kB
Loop #57: 56,860.0 kB
Loop #58: 56,860.0 kB
Loop #59: 56,860.0 kB
Loop #60: 56,860.0 kB
Loop #61: 56,860.0 kB
Loop #62: 56,860.0 kB
Loop #63: 56,860.0 kB
Loop #64: 56,860.0 kB
Loop #65: 56,860.0 kB
Loop #66: 56,860.0 kB
Loop #67: 56,860.0 kB
Loop #68: 56,860.0 kB
Loop #69: 56,860.0 kB
Loop #70: 56,860.0 kB
Loop #71: 57,652.0 kB
Loop #72: 58,444.0 kB
Loop #73: 59,236.0 kB
Loop #74: 59,236.0 kB
Loop #75: 59,236.0 kB
Loop #76: 59,236.0 kB
Loop #77: 59,236.0 kB
Loop #78: 59,236.0 kB
Loop #79: 59,236.0 kB
Loop #80: 59,236.0 kB
Loop #81: 59,236.0 kB
Loop #82: 59,236.0 kB
Loop #83: 59,236.0 kB
Loop #84: 59,236.0 kB
Loop #85: 59,236.0 kB
Loop #86: 59,236.0 kB
Loop #87: 59,500.0 kB
Loop #88: 59,500.0 kB
Loop #89: 59,500.0 kB
Loop #90: 59,500.0 kB
Loop #91: 59,500.0 kB
Loop #92: 59,500.0 kB
Loop #93: 59,500.0 kB
Loop #94: 59,500.0 kB
Loop #95: 59,500.0 kB
Loop #96: 59,500.0 kB
Loop #97: 59,500.0 kB
Loop #98: 59,500.0 kB
Loop #99: 59,500.0 kB
Loop #100: 59,500.0 kB
Loop #101: 59,500.0 kB
Loop #102: 59,500.0 kB
Loop #103: 59,500.0 kB
Loop #104: 59,500.0 kB
Loop #105: 59,500.0 kB
Loop #106: 59,500.0 kB
Loop #107: 59,500.0 kB
Loop #108: 59,500.0 kB
Loop #109: 59,500.0 kB
Loop #110: 59,500.0 kB
Loop #111: 59,500.0 kB
Loop #112: 59,500.0 kB
Loop #113: 59,500.0 kB
Loop #114: 59,500.0 kB
Loop #115: 59,500.0 kB
Loop #116: 59,500.0 kB
Loop #117: 59,500.0 kB
Loop #118: 59,500.0 kB
Loop #119: 59,500.0 kB

[nascheme]

I tried to narrow down the issue a bit better. It seems SSL is not required. I'm running the script below. It loads from a local web server (port 8001 without SSL) and the memory still goes up quite high. With 3.14.2, it stabilizes at about 420 MB. Version 3.14.0 stabilizes at about 80 MB (sometimes 60 MB). Bisecting between those two versions shows this commit as the cause for the increased memory use: d1a434f7b211b7061883b8cf4c8687cf00e0c2c7 (backport of GH-140262).

I'm still not sure what's going on. With the free-threaded GC (which is quite different), the memory usage is also very high. I think "httpx" is creating reference cycles and the GC is not going a good job to free them. If you run full GC collections frequently enough then memory usage stays low. But if you defer collection too long, memory usage goes much higher. That behavior makes me suspect weak references are involved or maybe finalizers that are resurrecting objects.

# /// script
# requires-python = ">=3.13"
# dependencies = [
#   "httpx",
# ]
# ///
import httpx
import asyncio
import gc
import os
import sys


def get_rss():
    with open(f"/proc/{os.getpid()}/status") as fp:
        for line in fp:
            if line.startswith("VmRSS"):
                rss = line.split(":", 1)[1].rstrip()
                rss = rss.removesuffix(" kB")
                return int(rss)
    return None


URL = 'http://localhost:8001/'


async def _main() -> None:
    for i in range(3000):
        async with httpx.AsyncClient() as client:
            await client.get(URL)
        rss = get_rss()
        print(f"Loop #{i}: {rss:,} kB")


if __name__ == "__main__":

    def gc_callback(phase, info):
        if phase == 'stop':
            print(
                f'gen={info.get("generation")} collected={info.get("collected")}',
                file=sys.stderr,
            )

    if False:
        gc.callbacks.append(gc_callback)
    if False:
        gc.set_threshold(200, 2)
    asyncio.run(_main())

[sergey-miryanov]

Yeah, if we set first gc threshold quite low then memory usage is stable (I tried set_threshold(100)). Yes, httpx (httpx -> httpcore -> h11) creates cycles for example for sentinel values CLIENT, SERVER and others. Also, if I call gc_collect_full instead of gc_collect_increment then memory usage grows but not so fast.

For 120 loops gc called about 12 times. First six gc runs do nothing (because of work_to_do is low). Maybe we can't break all cycles with one pass?

[sergey-miryanov]

One example of the cycle:

Response -> BoundAsyncStream -> Response -> ...

gc: collectable <Response 0x7fd8e3c4cc20>
  --> <bytes 0x561a014e9218 rc=3221225472> <b''>
  --> <IdentityDecoder 0x7fd8e3c4d010 rc=1> <<httpx._decoders.IdentityDecoder object at 0x7fd8e3c4d010>>
    --> <type 0x561a238f7ad0 rc=17> <<class 'httpx._decoders.IdentityDecoder'>>
  --> <datetime.timedelta 0x7fd8e551a490 rc=1> <datetime.timedelta(microseconds=329171)>
  --> <int 0x561a014e71f8 rc=3221225472> <0>
  --> <Request 0x7fd8e3cf4ec0 rc=1> <<Request('HEAD', 'http://www.google.com/')>>
    --> <bytes 0x561a014e9218 rc=3221225472> <b''>
    --> <dict 0x7fd8e3bd1200 rc=2> <{'timeout': {'connect': 5.0, 'read': 5.0, 'write': 5.0, 'pool': 5.0}}>
    --> <Headers 0x7fd8e3d68d10 rc=1> <Headers({'host': 'www.google.com', 'accept': '*/*', 'accept-encoding': 'gzip, deflate', 'connection': 'keep-alive', 'user-agent': 'python-httpx/0.28.1'})>
      --> <str 0x7fd8e54e26d0 rc=3221225472> <'ascii'>
      --> <list 0x7fd8e52edb80 rc=1> <[(b'Host', b'host', b'www.google.com'), (b'Accept', b'accept', b'*/*'), (b'Accept-Encoding', b'accept-encoding', b'gzip, deflate'), (b'Connection', b'connection', b'keep-alive'), (b'User-Agent', b'user-agent', b'python-httpx/0.28.1')]>
      --> <ABCMeta 0x561a23906090 rc=35> <<class 'httpx.Headers'>>
        --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
        --> <dict 0x7fd8e44e8280 rc=1> <{'__module__': 'httpx', '__firstlineno__': 139, '__doc__': '\nHTTP headers, as a case-insensitive multi-dict.\n', '__init__': <function Headers.__init__ at 0x7fd8e44ef320>, 'encoding': <property object at 0x7fd8e44b98f0>, 'raw': <property object at 0x7fd8e44fdc60>, 'keys': <function Headers.keys at 0x7fd8e44ef8a0>, 'values': <function Headers.values at 0x7fd8e44efa00>, 'items': <function Headers.items at 0x7fd8e44efb60>, 'multi_items': <function Headers.multi_items at 0x7fd8e44efcc0>, 'get': <function Headers.get at 0x7fd8e44efe20>, 'get_list': <function Headers.get_list at 0x7fd8e44f0040>, 'update': <function Headers.update at 0x7fd8e44f01a0>, 'copy': <function Headers.copy at 0x7fd8e44f0300>, '__getitem__': <function Headers.__getitem__ at 0x7fd8e44f0460>, '__setitem__': <function Headers.__setitem__ at 0x7fd8e44f05c0>, '__delitem__': <function Headers.__delitem__ at 0x7fd8e44f0720>, '__contains__': <function Headers.__contains__ at 0x7fd8e44f0880>, '__iter__': <function Headers.__iter__ at 0x7fd8e44f09e0>, '__len__': <function Headers.__len__ at 0x7fd8e44f0b40>, '__eq__': <function Headers.__eq__ at 0x7fd8e44f0ca0>, '__repr__': <function Headers.__repr__ at 0x7fd8e44f0e00>, '__static_attributes__': ('_encoding', '_list'), '__orig_bases__': (typing.MutableMapping[str, str],), '__dict__': <attribute '__dict__' of 'Headers' objects>, '__weakref__': <attribute '__weakref__' of 'Headers' objects>, '__hash__': None, '__parameters__': (), '__abstractmethods__': frozenset(), '_abc_impl': <_abc._abc_data object at 0x7fd8e44e9d80>}>
        --> <tuple 0x7fd8e448ca50 rc=1> <(<class 'httpx.Headers'>, <class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'typing.Generic'>, <class 'object'>)>
        --> <tuple 0x7fd8e44b7940 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'typing.Generic'>)>
        --> <ABCMeta 0x561a2365ddd0 rc=27> <<class 'collections.abc.MutableMapping'>>
          --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
          --> <dict 0x7fd8e535ee80 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 908, '__doc__': 'A MutableMapping is a generic container for associating\nkey/value pairs.\n\nThis class provides concrete generic implementations of all\nmethods except for __getitem__, __setitem__, __delitem__,\n__iter__, and __len__.\n', '__slots__': (), '__setitem__': <function MutableMapping.__setitem__ at 0x7fd8e5364670>, '__delitem__': <function MutableMapping.__delitem__ at 0x7fd8e5364720>, '_MutableMapping__marker': <object object at 0x7fd8e54c01b0>, 'pop': <function MutableMapping.pop at 0x7fd8e53647d0>, 'popitem': <function MutableMapping.popitem at 0x7fd8e5364880>, 'clear': <function MutableMapping.clear at 0x7fd8e5364930>, 'update': <function MutableMapping.update at 0x7fd8e53649e0>, 'setdefault': <function MutableMapping.setdefault at 0x7fd8e5364a90>, '__static_attributes__': (), '__abstractmethods__': frozenset({'__setitem__', '__delitem__', '__getitem__', '__len__', '__iter__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535ef40>}>
          --> <tuple 0x7fd8e551f880 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
          --> <tuple 0x7fd8e5351480 rc=1> <(<class 'collections.abc.Mapping'>,)>
          --> <ABCMeta 0x561a2365ce10 rc=34> <<class 'collections.abc.Mapping'>>
            --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
            --> <dict 0x7fd8e535e840 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 775, '__doc__': 'A Mapping is a generic container for associating key/value\npairs.\n\nThis class provides concrete generic implementations of all\nmethods except for __getitem__, __iter__, and __len__.\n', '__slots__': (), '__getitem__': <function Mapping.__getitem__ at 0x7fd8e531f950>, 'get': <function Mapping.get at 0x7fd8e531fa00>, '__contains__': <function Mapping.__contains__ at 0x7fd8e531fab0>, 'keys': <function Mapping.keys at 0x7fd8e531fb60>, 'items': <function Mapping.items at 0x7fd8e531fc10>, 'values': <function Mapping.values at 0x7fd8e531fcc0>, '__eq__': <function Mapping.__eq__ at 0x7fd8e531fd70>, '__reversed__': None, '__static_attributes__': (), '__hash__': None, '__abstractmethods__': frozenset({'__iter__', '__len__', '__getitem__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535e8c0>}>
            --> <tuple 0x7fd8e551f580 rc=1> <(<class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
            --> <tuple 0x7fd8e53511b0 rc=1> <(<class 'collections.abc.Collection'>,)>
            --> <ABCMeta 0x561a2365b360 rc=36> <<class 'collections.abc.Collection'>>
    --> <str 0x7fd8e40b2e20 rc=1> <'HEAD'>
    --> <ByteStream 0x7fd8e3cf57f0 rc=2> <<httpx.ByteStream object at 0x7fd8e3cf57f0>>
      --> <bytes 0x561a014e9218 rc=3221225472> <b''>
      --> <type 0x561a238ede30 rc=18> <<class 'httpx.ByteStream'>>
    --> <URL 0x7fd8e3ce7250 rc=1> <URL('http://www.google.com/')>
      --> <ParseResult 0x7fd8e3bbc430 rc=1> <ParseResult(scheme='http', userinfo='', host='www.google.com', port=None, path='/', query=None, fragment=None)>
        --> <type 0x561a2390fdb0 rc=17> <<class 'httpx._urlparse.ParseResult'>>
        --> <NoneType 0x561a014be300 rc=3221225472> <None>
        --> <NoneType 0x561a014be300 rc=3221225472> <None>
        --> <str 0x561a014f6748 rc=3221225472> <'/'>
        --> <NoneType 0x561a014be300 rc=3221225472> <None>
        --> <str 0x7fd8e3f37070 rc=1> <'www.google.com'>
        --> <str 0x561a014ec530 rc=3221225472> <''>
        --> <str 0x7fd8e3ffbcc0 rc=1> <'http'>
      --> <type 0x561a2390f9c0 rc=25> <<class 'httpx.URL'>>
    --> <type 0x561a23905ca0 rc=30> <<class 'httpx.Request'>>
  --> <str 0x7fd8e52c9650 rc=17> <'utf-8'>
  --> <dict 0x7fd8e3c2c540 rc=1> <{'http_version': b'HTTP/1.1', 'reason_phrase': b'OK', 'network_stream': <httpcore._backends.anyio.AnyIOStream object at 0x7fd8e3c4c590>}>
  --> <Headers 0x7fd8e3d6ae00 rc=1> <Headers([('content-type', 'text/html; charset=ISO-8859-1'), ('content-security-policy-report-only', "object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), ('p3p', 'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), ('date', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('server', 'gws'), ('x-xss-protection', '0'), ('x-frame-options', 'SAMEORIGIN'), ('expires', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('cache-control', 'private'), ('set-cookie', 'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), ('set-cookie', 'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), ('transfer-encoding', 'chunked')])>
    --> <str 0x7fd8e54e26d0 rc=3221225472> <'ascii'>
    --> <list 0x7fd8e3c9eb80 rc=1> <[(b'Content-Type', b'content-type', b'text/html; charset=ISO-8859-1'), (b'Content-Security-Policy-Report-Only', b'content-security-policy-report-only', b"object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), (b'P3P', b'p3p', b'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), (b'Date', b'date', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Server', b'server', b'gws'), (b'X-XSS-Protection', b'x-xss-protection', b'0'), (b'X-Frame-Options', b'x-frame-options', b'SAMEORIGIN'), (b'Expires', b'expires', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Cache-Control', b'cache-control', b'private'), (b'Set-Cookie', b'set-cookie', b'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), (b'Set-Cookie', b'set-cookie', b'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), (b'Transfer-Encoding', b'transfer-encoding', b'chunked')]>
    --> <ABCMeta 0x561a23906090 rc=35> <<class 'httpx.Headers'>>
      --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
      --> <dict 0x7fd8e44e8280 rc=1> <{'__module__': 'httpx', '__firstlineno__': 139, '__doc__': '\nHTTP headers, as a case-insensitive multi-dict.\n', '__init__': <function Headers.__init__ at 0x7fd8e44ef320>, 'encoding': <property object at 0x7fd8e44b98f0>, 'raw': <property object at 0x7fd8e44fdc60>, 'keys': <function Headers.keys at 0x7fd8e44ef8a0>, 'values': <function Headers.values at 0x7fd8e44efa00>, 'items': <function Headers.items at 0x7fd8e44efb60>, 'multi_items': <function Headers.multi_items at 0x7fd8e44efcc0>, 'get': <function Headers.get at 0x7fd8e44efe20>, 'get_list': <function Headers.get_list at 0x7fd8e44f0040>, 'update': <function Headers.update at 0x7fd8e44f01a0>, 'copy': <function Headers.copy at 0x7fd8e44f0300>, '__getitem__': <function Headers.__getitem__ at 0x7fd8e44f0460>, '__setitem__': <function Headers.__setitem__ at 0x7fd8e44f05c0>, '__delitem__': <function Headers.__delitem__ at 0x7fd8e44f0720>, '__contains__': <function Headers.__contains__ at 0x7fd8e44f0880>, '__iter__': <function Headers.__iter__ at 0x7fd8e44f09e0>, '__len__': <function Headers.__len__ at 0x7fd8e44f0b40>, '__eq__': <function Headers.__eq__ at 0x7fd8e44f0ca0>, '__repr__': <function Headers.__repr__ at 0x7fd8e44f0e00>, '__static_attributes__': ('_encoding', '_list'), '__orig_bases__': (typing.MutableMapping[str, str],), '__dict__': <attribute '__dict__' of 'Headers' objects>, '__weakref__': <attribute '__weakref__' of 'Headers' objects>, '__hash__': None, '__parameters__': (), '__abstractmethods__': frozenset(), '_abc_impl': <_abc._abc_data object at 0x7fd8e44e9d80>}>
      --> <tuple 0x7fd8e448ca50 rc=1> <(<class 'httpx.Headers'>, <class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'typing.Generic'>, <class 'object'>)>
      --> <tuple 0x7fd8e44b7940 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'typing.Generic'>)>
      --> <ABCMeta 0x561a2365ddd0 rc=27> <<class 'collections.abc.MutableMapping'>>
        --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
        --> <dict 0x7fd8e535ee80 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 908, '__doc__': 'A MutableMapping is a generic container for associating\nkey/value pairs.\n\nThis class provides concrete generic implementations of all\nmethods except for __getitem__, __setitem__, __delitem__,\n__iter__, and __len__.\n', '__slots__': (), '__setitem__': <function MutableMapping.__setitem__ at 0x7fd8e5364670>, '__delitem__': <function MutableMapping.__delitem__ at 0x7fd8e5364720>, '_MutableMapping__marker': <object object at 0x7fd8e54c01b0>, 'pop': <function MutableMapping.pop at 0x7fd8e53647d0>, 'popitem': <function MutableMapping.popitem at 0x7fd8e5364880>, 'clear': <function MutableMapping.clear at 0x7fd8e5364930>, 'update': <function MutableMapping.update at 0x7fd8e53649e0>, 'setdefault': <function MutableMapping.setdefault at 0x7fd8e5364a90>, '__static_attributes__': (), '__abstractmethods__': frozenset({'__setitem__', '__delitem__', '__getitem__', '__len__', '__iter__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535ef40>}>
        --> <tuple 0x7fd8e551f880 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
        --> <tuple 0x7fd8e5351480 rc=1> <(<class 'collections.abc.Mapping'>,)>
        --> <ABCMeta 0x561a2365ce10 rc=34> <<class 'collections.abc.Mapping'>>
          --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
          --> <dict 0x7fd8e535e840 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 775, '__doc__': 'A Mapping is a generic container for associating key/value\npairs.\n\nThis class provides concrete generic implementations of all\nmethods except for __getitem__, __iter__, and __len__.\n', '__slots__': (), '__getitem__': <function Mapping.__getitem__ at 0x7fd8e531f950>, 'get': <function Mapping.get at 0x7fd8e531fa00>, '__contains__': <function Mapping.__contains__ at 0x7fd8e531fab0>, 'keys': <function Mapping.keys at 0x7fd8e531fb60>, 'items': <function Mapping.items at 0x7fd8e531fc10>, 'values': <function Mapping.values at 0x7fd8e531fcc0>, '__eq__': <function Mapping.__eq__ at 0x7fd8e531fd70>, '__reversed__': None, '__static_attributes__': (), '__hash__': None, '__abstractmethods__': frozenset({'__iter__', '__len__', '__getitem__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535e8c0>}>
          --> <tuple 0x7fd8e551f580 rc=1> <(<class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
          --> <tuple 0x7fd8e53511b0 rc=1> <(<class 'collections.abc.Collection'>,)>
          --> <ABCMeta 0x561a2365b360 rc=36> <<class 'collections.abc.Collection'>>
            --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
            --> <dict 0x7fd8e535dcc0 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 435, '__slots__': (), '__subclasshook__': <classmethod(<function Collection.__subclasshook__ at 0x7fd8e531e2a0>)>, '__static_attributes__': (), '__doc__': None, '__abstractmethods__': frozenset({'__iter__', '__len__', '__contains__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535de40>}>
            --> <tuple 0x7fd8e533a160 rc=1> <(<class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
            --> <tuple 0x7fd8e533d200 rc=1> <(<class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>)>
            --> <ABCMeta 0x561a2365ab80 rc=34> <<class 'collections.abc.Sized'>>
  --> <list 0x7fd8e3c2c900 rc=1> <[]>
  --> <bool 0x561a014a73c0 rc=3221225472> <True>
  --> <bool 0x561a014a73c0 rc=3221225472> <True>
  --> <NoneType 0x561a014be300 rc=3221225472> <None>
  --> <int 0x561a014e8af8 rc=3221225472> <200>
  --> <BoundAsyncStream 0x7fd8e3c4cec0 rc=1> <<httpx._client.BoundAsyncStream object at 0x7fd8e3c4cec0>>
    --> <Response 0x7fd8e3c4cc20 rc=1> <<Response [200 OK]>>
      --> <bytes 0x561a014e9218 rc=3221225472> <b''>
      --> <IdentityDecoder 0x7fd8e3c4d010 rc=1> <<httpx._decoders.IdentityDecoder object at 0x7fd8e3c4d010>>
        --> <type 0x561a238f7ad0 rc=17> <<class 'httpx._decoders.IdentityDecoder'>>
      --> <datetime.timedelta 0x7fd8e551a490 rc=1> <datetime.timedelta(microseconds=329171)>
      --> <int 0x561a014e71f8 rc=3221225472> <0>
      --> <Request 0x7fd8e3cf4ec0 rc=1> <<Request('HEAD', 'http://www.google.com/')>>
        --> <bytes 0x561a014e9218 rc=3221225472> <b''>
        --> <dict 0x7fd8e3bd1200 rc=2> <{'timeout': {'connect': 5.0, 'read': 5.0, 'write': 5.0, 'pool': 5.0}}>
        --> <Headers 0x7fd8e3d68d10 rc=1> <Headers({'host': 'www.google.com', 'accept': '*/*', 'accept-encoding': 'gzip, deflate', 'connection': 'keep-alive', 'user-agent': 'python-httpx/0.28.1'})>
          --> <str 0x7fd8e54e26d0 rc=3221225472> <'ascii'>
          --> <list 0x7fd8e52edb80 rc=1> <[(b'Host', b'host', b'www.google.com'), (b'Accept', b'accept', b'*/*'), (b'Accept-Encoding', b'accept-encoding', b'gzip, deflate'), (b'Connection', b'connection', b'keep-alive'), (b'User-Agent', b'user-agent', b'python-httpx/0.28.1')]>
          --> <ABCMeta 0x561a23906090 rc=35> <<class 'httpx.Headers'>>
            --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
            --> <dict 0x7fd8e44e8280 rc=1> <{'__module__': 'httpx', '__firstlineno__': 139, '__doc__': '\nHTTP headers, as a case-insensitive multi-dict.\n', '__init__': <function Headers.__init__ at 0x7fd8e44ef320>, 'encoding': <property object at 0x7fd8e44b98f0>, 'raw': <property object at 0x7fd8e44fdc60>, 'keys': <function Headers.keys at 0x7fd8e44ef8a0>, 'values': <function Headers.values at 0x7fd8e44efa00>, 'items': <function Headers.items at 0x7fd8e44efb60>, 'multi_items': <function Headers.multi_items at 0x7fd8e44efcc0>, 'get': <function Headers.get at 0x7fd8e44efe20>, 'get_list': <function Headers.get_list at 0x7fd8e44f0040>, 'update': <function Headers.update at 0x7fd8e44f01a0>, 'copy': <function Headers.copy at 0x7fd8e44f0300>, '__getitem__': <function Headers.__getitem__ at 0x7fd8e44f0460>, '__setitem__': <function Headers.__setitem__ at 0x7fd8e44f05c0>, '__delitem__': <function Headers.__delitem__ at 0x7fd8e44f0720>, '__contains__': <function Headers.__contains__ at 0x7fd8e44f0880>, '__iter__': <function Headers.__iter__ at 0x7fd8e44f09e0>, '__len__': <function Headers.__len__ at 0x7fd8e44f0b40>, '__eq__': <function Headers.__eq__ at 0x7fd8e44f0ca0>, '__repr__': <function Headers.__repr__ at 0x7fd8e44f0e00>, '__static_attributes__': ('_encoding', '_list'), '__orig_bases__': (typing.MutableMapping[str, str],), '__dict__': <attribute '__dict__' of 'Headers' objects>, '__weakref__': <attribute '__weakref__' of 'Headers' objects>, '__hash__': None, '__parameters__': (), '__abstractmethods__': frozenset(), '_abc_impl': <_abc._abc_data object at 0x7fd8e44e9d80>}>
            --> <tuple 0x7fd8e448ca50 rc=1> <(<class 'httpx.Headers'>, <class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'typing.Generic'>, <class 'object'>)>
            --> <tuple 0x7fd8e44b7940 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'typing.Generic'>)>
            --> <ABCMeta 0x561a2365ddd0 rc=27> <<class 'collections.abc.MutableMapping'>>
        --> <str 0x7fd8e40b2e20 rc=1> <'HEAD'>
        --> <ByteStream 0x7fd8e3cf57f0 rc=2> <<httpx.ByteStream object at 0x7fd8e3cf57f0>>
          --> <bytes 0x561a014e9218 rc=3221225472> <b''>
          --> <type 0x561a238ede30 rc=18> <<class 'httpx.ByteStream'>>
        --> <URL 0x7fd8e3ce7250 rc=1> <URL('http://www.google.com/')>
          --> <ParseResult 0x7fd8e3bbc430 rc=1> <ParseResult(scheme='http', userinfo='', host='www.google.com', port=None, path='/', query=None, fragment=None)>
            --> <type 0x561a2390fdb0 rc=17> <<class 'httpx._urlparse.ParseResult'>>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <str 0x561a014f6748 rc=3221225472> <'/'>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <str 0x7fd8e3f37070 rc=1> <'www.google.com'>
            --> <str 0x561a014ec530 rc=3221225472> <''>
            --> <str 0x7fd8e3ffbcc0 rc=1> <'http'>
          --> <type 0x561a2390f9c0 rc=25> <<class 'httpx.URL'>>
        --> <type 0x561a23905ca0 rc=30> <<class 'httpx.Request'>>
      --> <str 0x7fd8e52c9650 rc=17> <'utf-8'>
      --> <dict 0x7fd8e3c2c540 rc=1> <{'http_version': b'HTTP/1.1', 'reason_phrase': b'OK', 'network_stream': <httpcore._backends.anyio.AnyIOStream object at 0x7fd8e3c4c590>}>
      --> <Headers 0x7fd8e3d6ae00 rc=1> <Headers([('content-type', 'text/html; charset=ISO-8859-1'), ('content-security-policy-report-only', "object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), ('p3p', 'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), ('date', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('server', 'gws'), ('x-xss-protection', '0'), ('x-frame-options', 'SAMEORIGIN'), ('expires', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('cache-control', 'private'), ('set-cookie', 'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), ('set-cookie', 'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), ('transfer-encoding', 'chunked')])>
        --> <str 0x7fd8e54e26d0 rc=3221225472> <'ascii'>
        --> <list 0x7fd8e3c9eb80 rc=1> <[(b'Content-Type', b'content-type', b'text/html; charset=ISO-8859-1'), (b'Content-Security-Policy-Report-Only', b'content-security-policy-report-only', b"object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), (b'P3P', b'p3p', b'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), (b'Date', b'date', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Server', b'server', b'gws'), (b'X-XSS-Protection', b'x-xss-protection', b'0'), (b'X-Frame-Options', b'x-frame-options', b'SAMEORIGIN'), (b'Expires', b'expires', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Cache-Control', b'cache-control', b'private'), (b'Set-Cookie', b'set-cookie', b'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), (b'Set-Cookie', b'set-cookie', b'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), (b'Transfer-Encoding', b'transfer-encoding', b'chunked')]>
        --> <ABCMeta 0x561a23906090 rc=35> <<class 'httpx.Headers'>>
          --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
          --> <dict 0x7fd8e44e8280 rc=1> <{'__module__': 'httpx', '__firstlineno__': 139, '__doc__': '\nHTTP headers, as a case-insensitive multi-dict.\n', '__init__': <function Headers.__init__ at 0x7fd8e44ef320>, 'encoding': <property object at 0x7fd8e44b98f0>, 'raw': <property object at 0x7fd8e44fdc60>, 'keys': <function Headers.keys at 0x7fd8e44ef8a0>, 'values': <function Headers.values at 0x7fd8e44efa00>, 'items': <function Headers.items at 0x7fd8e44efb60>, 'multi_items': <function Headers.multi_items at 0x7fd8e44efcc0>, 'get': <function Headers.get at 0x7fd8e44efe20>, 'get_list': <function Headers.get_list at 0x7fd8e44f0040>, 'update': <function Headers.update at 0x7fd8e44f01a0>, 'copy': <function Headers.copy at 0x7fd8e44f0300>, '__getitem__': <function Headers.__getitem__ at 0x7fd8e44f0460>, '__setitem__': <function Headers.__setitem__ at 0x7fd8e44f05c0>, '__delitem__': <function Headers.__delitem__ at 0x7fd8e44f0720>, '__contains__': <function Headers.__contains__ at 0x7fd8e44f0880>, '__iter__': <function Headers.__iter__ at 0x7fd8e44f09e0>, '__len__': <function Headers.__len__ at 0x7fd8e44f0b40>, '__eq__': <function Headers.__eq__ at 0x7fd8e44f0ca0>, '__repr__': <function Headers.__repr__ at 0x7fd8e44f0e00>, '__static_attributes__': ('_encoding', '_list'), '__orig_bases__': (typing.MutableMapping[str, str],), '__dict__': <attribute '__dict__' of 'Headers' objects>, '__weakref__': <attribute '__weakref__' of 'Headers' objects>, '__hash__': None, '__parameters__': (), '__abstractmethods__': frozenset(), '_abc_impl': <_abc._abc_data object at 0x7fd8e44e9d80>}>
          --> <tuple 0x7fd8e448ca50 rc=1> <(<class 'httpx.Headers'>, <class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'typing.Generic'>, <class 'object'>)>
          --> <tuple 0x7fd8e44b7940 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'typing.Generic'>)>
          --> <ABCMeta 0x561a2365ddd0 rc=27> <<class 'collections.abc.MutableMapping'>>
            --> <type 0x561a23638a00 rc=158> <<class 'abc.ABCMeta'>>
            --> <dict 0x7fd8e535ee80 rc=1> <{'__module__': 'collections.abc', '__firstlineno__': 908, '__doc__': 'A MutableMapping is a generic container for associating\nkey/value pairs.\n\nThis class provides concrete generic implementations of all\nmethods except for __getitem__, __setitem__, __delitem__,\n__iter__, and __len__.\n', '__slots__': (), '__setitem__': <function MutableMapping.__setitem__ at 0x7fd8e5364670>, '__delitem__': <function MutableMapping.__delitem__ at 0x7fd8e5364720>, '_MutableMapping__marker': <object object at 0x7fd8e54c01b0>, 'pop': <function MutableMapping.pop at 0x7fd8e53647d0>, 'popitem': <function MutableMapping.popitem at 0x7fd8e5364880>, 'clear': <function MutableMapping.clear at 0x7fd8e5364930>, 'update': <function MutableMapping.update at 0x7fd8e53649e0>, 'setdefault': <function MutableMapping.setdefault at 0x7fd8e5364a90>, '__static_attributes__': (), '__abstractmethods__': frozenset({'__setitem__', '__delitem__', '__getitem__', '__len__', '__iter__'}), '_abc_impl': <_abc._abc_data object at 0x7fd8e535ef40>}>
            --> <tuple 0x7fd8e551f880 rc=1> <(<class 'collections.abc.MutableMapping'>, <class 'collections.abc.Mapping'>, <class 'collections.abc.Collection'>, <class 'collections.abc.Sized'>, <class 'collections.abc.Iterable'>, <class 'collections.abc.Container'>, <class 'object'>)>
            --> <tuple 0x7fd8e5351480 rc=1> <(<class 'collections.abc.Mapping'>,)>
            --> <ABCMeta 0x561a2365ce10 rc=34> <<class 'collections.abc.Mapping'>>
      --> <list 0x7fd8e3c2c900 rc=1> <[]>
      --> <bool 0x561a014a73c0 rc=3221225472> <True>
      --> <bool 0x561a014a73c0 rc=3221225472> <True>
      --> <NoneType 0x561a014be300 rc=3221225472> <None>
      --> <int 0x561a014e8af8 rc=3221225472> <200>
      --> <BoundAsyncStream 0x7fd8e3c4cec0 rc=1> <<httpx._client.BoundAsyncStream object at 0x7fd8e3c4cec0>>
        --> <Response 0x7fd8e3c4cc20 rc=1> <<Response [200 OK]>>
          --> <bytes 0x561a014e9218 rc=3221225472> <b''>
          --> <IdentityDecoder 0x7fd8e3c4d010 rc=1> <<httpx._decoders.IdentityDecoder object at 0x7fd8e3c4d010>>
            --> <type 0x561a238f7ad0 rc=17> <<class 'httpx._decoders.IdentityDecoder'>>
          --> <datetime.timedelta 0x7fd8e551a490 rc=1> <datetime.timedelta(microseconds=329171)>
          --> <int 0x561a014e71f8 rc=3221225472> <0>
          --> <Request 0x7fd8e3cf4ec0 rc=1> <<Request('HEAD', 'http://www.google.com/')>>
            --> <bytes 0x561a014e9218 rc=3221225472> <b''>
            --> <dict 0x7fd8e3bd1200 rc=2> <{'timeout': {'connect': 5.0, 'read': 5.0, 'write': 5.0, 'pool': 5.0}}>
            --> <Headers 0x7fd8e3d68d10 rc=1> <Headers({'host': 'www.google.com', 'accept': '*/*', 'accept-encoding': 'gzip, deflate', 'connection': 'keep-alive', 'user-agent': 'python-httpx/0.28.1'})>
            --> <str 0x7fd8e40b2e20 rc=1> <'HEAD'>
            --> <ByteStream 0x7fd8e3cf57f0 rc=2> <<httpx.ByteStream object at 0x7fd8e3cf57f0>>
            --> <URL 0x7fd8e3ce7250 rc=1> <URL('http://www.google.com/')>
            --> <type 0x561a23905ca0 rc=30> <<class 'httpx.Request'>>
          --> <str 0x7fd8e52c9650 rc=17> <'utf-8'>
          --> <dict 0x7fd8e3c2c540 rc=1> <{'http_version': b'HTTP/1.1', 'reason_phrase': b'OK', 'network_stream': <httpcore._backends.anyio.AnyIOStream object at 0x7fd8e3c4c590>}>
          --> <Headers 0x7fd8e3d6ae00 rc=1> <Headers([('content-type', 'text/html; charset=ISO-8859-1'), ('content-security-policy-report-only', "object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), ('p3p', 'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), ('date', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('server', 'gws'), ('x-xss-protection', '0'), ('x-frame-options', 'SAMEORIGIN'), ('expires', 'Thu, 18 Dec 2025 19:58:31 GMT'), ('cache-control', 'private'), ('set-cookie', 'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), ('set-cookie', 'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), ('transfer-encoding', 'chunked')])>
            --> <str 0x7fd8e54e26d0 rc=3221225472> <'ascii'>
            --> <list 0x7fd8e3c9eb80 rc=1> <[(b'Content-Type', b'content-type', b'text/html; charset=ISO-8859-1'), (b'Content-Security-Policy-Report-Only', b'content-security-policy-report-only', b"object-src 'none';base-uri 'self';script-src 'nonce-wYhXOYJqh4UnZEs4DxhhSQ' 'strict-dynamic' 'report-sample' 'unsafe-eval' 'unsafe-inline' https: http:;report-uri https://csp.withgoogle.com/csp/gws/other-hp"), (b'P3P', b'p3p', b'CP="This is not a P3P policy! See g.co/p3phelp for more info."'), (b'Date', b'date', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Server', b'server', b'gws'), (b'X-XSS-Protection', b'x-xss-protection', b'0'), (b'X-Frame-Options', b'x-frame-options', b'SAMEORIGIN'), (b'Expires', b'expires', b'Thu, 18 Dec 2025 19:58:31 GMT'), (b'Cache-Control', b'cache-control', b'private'), (b'Set-Cookie', b'set-cookie', b'AEC=AaJma5s1zLZQ5IPrCVHEYXZ36YmVgcr5-AEKPVg2jUl76XZ8_WnBc1Hi1w; expires=Tue, 16-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; Secure; HttpOnly; SameSite=lax'), (b'Set-Cookie', b'set-cookie', b'NID=527=3iGMZI4Y7equ8Mm7al5iqN6B8scsLZWhqvIoNekMv-XogyGDxHpXAHp8u2R7PENdMzUoboTQ7K13pz2v4F0T8rlk6STI8Svq1M9C7y6OLyXFuBVwhfA_oueOg7wWzo6cpBGVLZ9BpJe_FZHsEQ4jhdiw2YpY-QI93PWn8l50q_Dc0WuOh4JESyPMws1qZbp1efSUN5aVp_zlV4Gn8Cu-Kf48hx8H; expires=Fri, 19-Jun-2026 19:58:31 GMT; path=/; domain=.google.com; HttpOnly'), (b'Transfer-Encoding', b'transfer-encoding', b'chunked')]>
            --> <ABCMeta 0x561a23906090 rc=35> <<class 'httpx.Headers'>>
          --> <list 0x7fd8e3c2c900 rc=1> <[]>
          --> <bool 0x561a014a73c0 rc=3221225472> <True>
          --> <bool 0x561a014a73c0 rc=3221225472> <True>
          --> <NoneType 0x561a014be300 rc=3221225472> <None>
          --> <int 0x561a014e8af8 rc=3221225472> <200>
          --> <BoundAsyncStream 0x7fd8e3c4cec0 rc=1> <<httpx._client.BoundAsyncStream object at 0x7fd8e3c4cec0>>
            --> <Response 0x7fd8e3c4cc20 rc=1> <<Response [200 OK]>>
            --> <float 0x7fd8e3cdd3b0 rc=1> <39037.104077293>
            --> <AsyncResponseStream 0x7fd8e3cf6120 rc=1> <<httpx._transports.default.AsyncResponseStream object at 0x7fd8e3cf6120>>
            --> <type 0x561a23924160 rc=15> <<class 'httpx._client.BoundAsyncStream'>>
          --> <type 0x561a239058b0 rc=30> <<class 'httpx.Response'>>
        --> <float 0x7fd8e3cdd3b0 rc=1> <39037.104077293>
        --> <AsyncResponseStream 0x7fd8e3cf6120 rc=1> <<httpx._transports.default.AsyncResponseStream object at 0x7fd8e3cf6120>>
          --> <PoolByteStream 0x7fd8e3c4cd70 rc=1> <<httpcore._async.connection_pool.PoolByteStream object at 0x7fd8e3c4cd70>>
            --> <bool 0x561a014a73c0 rc=3221225472> <True>
            --> <AsyncConnectionPool 0x7fd8e3cf4c20 rc=1> <<AsyncConnectionPool [Requests: 0 active, 0 queued | Connections: 0 active, 0 idle]>>
            --> <AsyncPoolRequest 0x7fd8e3cf6270 rc=1> <<httpcore._async.connection_pool.AsyncPoolRequest object at 0x7fd8e3cf6270>>
            --> <HTTP11ConnectionByteStream 0x7fd8e3cf7770 rc=1> <<httpcore._async.http11.HTTP11ConnectionByteStream object at 0x7fd8e3cf7770>>
            --> <type 0x561a23ad0b10 rc=17> <<class 'httpcore._async.connection_pool.PoolByteStream'>>
          --> <type 0x561a2391c5d0 rc=15> <<class 'httpx._transports.default.AsyncResponseStream'>>
        --> <type 0x561a23924160 rc=15> <<class 'httpx._client.BoundAsyncStream'>>
      --> <type 0x561a239058b0 rc=30> <<class 'httpx.Response'>>
    --> <float 0x7fd8e3cdd3b0 rc=1> <39037.104077293>
    --> <AsyncResponseStream 0x7fd8e3cf6120 rc=1> <<httpx._transports.default.AsyncResponseStream object at 0x7fd8e3cf6120>>
      --> <PoolByteStream 0x7fd8e3c4cd70 rc=1> <<httpcore._async.connection_pool.PoolByteStream object at 0x7fd8e3c4cd70>>
        --> <bool 0x561a014a73c0 rc=3221225472> <True>
        --> <AsyncConnectionPool 0x7fd8e3cf4c20 rc=1> <<AsyncConnectionPool [Requests: 0 active, 0 queued | Connections: 0 active, 0 idle]>>
          --> <list 0x7fd8e3ee70c0 rc=1> <[]>
          --> <bool 0x561a014a73c0 rc=3221225472> <True>
          --> <bool 0x561a014a73e0 rc=3221225472> <False>
          --> <float 0x7fd8e48bd710 rc=113> <5.0>
          --> <NoneType 0x561a014be300 rc=3221225472> <None>
          --> <int 0x561a014e7e78 rc=3221225472> <100>
          --> <int 0x561a014e7478 rc=3221225472> <20>
          --> <AutoBackend 0x7fd8e3cf4ad0 rc=2> <<httpcore._backends.auto.AutoBackend object at 0x7fd8e3cf4ad0>>
            --> <AnyIOBackend 0x7fd8e3cf7380 rc=1> <<httpcore.AnyIOBackend object at 0x7fd8e3cf7380>>
            --> <type 0x561a23accb80 rc=18> <<class 'httpcore._backends.auto.AutoBackend'>>
          --> <AsyncThreadLock 0x7fd8e3cf4d70 rc=1> <<httpcore._synchronization.AsyncThreadLock object at 0x7fd8e3cf4d70>>
            --> <type 0x561a23a532c0 rc=19> <<class 'httpcore._synchronization.AsyncThreadLock'>>
          --> <NoneType 0x561a014be300 rc=3221225472> <None>
          --> <list 0x7fd8e40aa540 rc=1> <[]>
          --> <int 0x561a014e71f8 rc=3221225472> <0>
          --> <NoneType 0x561a014be300 rc=3221225472> <None>
          --> <SSLContext 0x7fd8e3caa3c0 rc=2> <<ssl.SSLContext object at 0x7fd8e3caa3c0>>
            --> <type 0x561a23809210 rc=20> <<class 'ssl.SSLContext'>>
          --> <NoneType 0x561a014be300 rc=3221225472> <None>
          --> <type 0x561a23ad0410 rc=22> <<class 'httpcore.AsyncConnectionPool'>>
        --> <AsyncPoolRequest 0x7fd8e3cf6270 rc=1> <<httpcore._async.connection_pool.AsyncPoolRequest object at 0x7fd8e3cf6270>>
          --> <AsyncEvent 0x7fd8e3cf63c0 rc=1> <<httpcore._synchronization.AsyncEvent object at 0x7fd8e3cf63c0>>
            --> <Event 0x7fd8e3cf6900 rc=1> <<anyio._backends._asyncio.Event object at 0x7fd8e3cf6900>>
            --> <str 0x7fd8e5353630 rc=3221225472> <'asyncio'>
            --> <type 0x561a23a97010 rc=19> <<class 'httpcore._synchronization.AsyncEvent'>>
          --> <AsyncHTTPConnection 0x7fd8e3cf6660 rc=1> <<AsyncHTTPConnection ['http://www.google.com:80', HTTP/1.1, CLOSED, Request Count: 1]>>
            --> <bool 0x561a014a73e0 rc=3221225472> <False>
            --> <AsyncHTTP11Connection 0x7fd8e3c4c6e0 rc=2> <<AsyncHTTP11Connection ['http://www.google.com:80', CLOSED, Request Count: 1]>>
            --> <bool 0x561a014a73c0 rc=3221225472> <True>
            --> <bool 0x561a014a73e0 rc=3221225472> <False>
            --> <float 0x7fd8e48bd710 rc=113> <5.0>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <AutoBackend 0x7fd8e3cf4ad0 rc=2> <<httpcore._backends.auto.AutoBackend object at 0x7fd8e3cf4ad0>>
            --> <Origin 0x7fd8e3cf6510 rc=2> <<httpcore.Origin object at 0x7fd8e3cf6510>>
            --> <AsyncLock 0x7fd8e3cf67b0 rc=1> <<httpcore._synchronization.AsyncLock object at 0x7fd8e3cf67b0>>
            --> <int 0x561a014e71f8 rc=3221225472> <0>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <SSLContext 0x7fd8e3caa3c0 rc=2> <<ssl.SSLContext object at 0x7fd8e3caa3c0>>
            --> <NoneType 0x561a014be300 rc=3221225472> <None>
            --> <type 0x561a23acf3a0 rc=20> <<class 'httpcore.AsyncHTTPConnection'>>
          --> <Request 0x7fd8e3cf5d30 rc=2> <<Request [b'HEAD']>>
            --> <dict 0x7fd8e3bd1200 rc=2> <{'timeout': {'connect': 5.0, 'read': 5.0, 'write': 5.0, 'pool': 5.0}}>
            --> <list 0x7fd8e3d534c0 rc=1> <[(b'Host', b'www.google.com'), (b'Accept', b'*/*'), (b'Accept-Encoding', b'gzip, deflate'), (b'Connection', b'keep-alive'), (b'User-Agent', b'python-httpx/0.28.1')]>
            --> <bytes 0x7fd8e3c967f0 rc=1> <b'HEAD'>
            --> <ByteStream 0x7fd8e3cf57f0 rc=2> <<httpx.ByteStream object at 0x7fd8e3cf57f0>>
            --> <URL 0x7fd8e3cf5fd0 rc=1> <URL(scheme=b'http', host=b'www.google.com', port=None, target=b'/')>
            --> <type 0x561a23a8b5f0 rc=30> <<class 'httpcore.Request'>>
          --> <type 0x561a23ada870 rc=18> <<class 'httpcore._async.connection_pool.AsyncPoolRequest'>>
        --> <HTTP11ConnectionByteStream 0x7fd8e3cf7770 rc=1> <<httpcore._async.http11.HTTP11ConnectionByteStream object at 0x7fd8e3cf7770>>
          --> <bool 0x561a014a73c0 rc=3221225472> <True>
          --> <AsyncHTTP11Connection 0x7fd8e3c4c6e0 rc=2> <<AsyncHTTP11Connection ['http://www.google.com:80', CLOSED, Request Count: 1]>>
            --> <float 0x7fd8e3cdd470 rc=1> <39042.433826994>
            --> <Connection 0x7fd8e3c4c830 rc=1> <<h11._connection.Connection object at 0x7fd8e3c4c830>>
            --> <float 0x7fd8e48bd710 rc=113> <5.0>
            --> <AnyIOStream 0x7fd8e3c4c590 rc=2> <<httpcore._backends.anyio.AnyIOStream object at 0x7fd8e3c4c590>>
            --> <Origin 0x7fd8e3cf6510 rc=2> <<httpcore.Origin object at 0x7fd8e3cf6510>>
            --> <int 0x561a014e7218 rc=3221225472> <1>
            --> <HTTPConnectionState 0x7fd8e3d50c50 rc=16> <<HTTPConnectionState.CLOSED: 3>>
            --> <AsyncLock 0x7fd8e3bd8910 rc=1> <<httpcore._synchronization.AsyncLock object at 0x7fd8e3bd8910>>
            --> <type 0x561a23ace1b0 rc=19> <<class 'httpcore.AsyncHTTP11Connection'>>
          --> <Request 0x7fd8e3cf5d30 rc=2> <<Request [b'HEAD']>>
            --> <dict 0x7fd8e3bd1200 rc=2> <{'timeout': {'connect': 5.0, 'read': 5.0, 'write': 5.0, 'pool': 5.0}}>
            --> <list 0x7fd8e3d534c0 rc=1> <[(b'Host', b'www.google.com'), (b'Accept', b'*/*'), (b'Accept-Encoding', b'gzip, deflate'), (b'Connection', b'keep-alive'), (b'User-Agent', b'python-httpx/0.28.1')]>
            --> <bytes 0x7fd8e3c967f0 rc=1> <b'HEAD'>
            --> <ByteStream 0x7fd8e3cf57f0 rc=2> <<httpx.ByteStream object at 0x7fd8e3cf57f0>>
            --> <URL 0x7fd8e3cf5fd0 rc=1> <URL(scheme=b'http', host=b'www.google.com', port=None, target=b'/')>
            --> <type 0x561a23a8b5f0 rc=30> <<class 'httpcore.Request'>>
          --> <type 0x561a23acebc0 rc=17> <<class 'httpcore._async.http11.HTTP11ConnectionByteStream'>>
        --> <type 0x561a23ad0b10 rc=17> <<class 'httpcore._async.connection_pool.PoolByteStream'>>
      --> <type 0x561a2391c5d0 rc=15> <<class 'httpx._transports.default.AsyncResponseStream'>>
    --> <type 0x561a23924160 rc=15> <<class 'httpx._client.BoundAsyncStream'>>
  --> <type 0x561a239058b0 rc=30> <<class 'httpx.Response'>>

[nascheme]

I was wrong about SSL not being involved. Here is a memray flame graph. Nearly all the memory is allocated from load_verify_locations(), which is implemented in the _ssl.c module, it seems. Given that explicitly calling gc.collect() fixes the memory usage, I would suspect it's not memory leak in the _ssl.c code but due to reference cycles keeping the default context alive.

Edit: passing a context to the verify parameter drastically reduces the memory usage, e.g.

    ssl_ctx = ssl.create_default_context()
    for i in range(2000):
        async with httpx.AsyncClient(verify=ssl_ctx) as client:
            resp = await client.get(URL)

[nascheme]

One example of the cycle:

Response -> BoundAsyncStream -> Response -> ...

Good detective work! If I break that cycle, the high memory usage goes away. E.g. add this line at the end of the "with" block: resp.stream = None. It would be good if httpx would break that cycle when the "with" block is exited. However, we should also try to fix the GC so that these cycles are cleaned more quickly.

[nascheme]

I don't know if this is an accurate minimization of the test case but it does use more memory in 3.14 vs 3.13 (and more yet in 3.14.2). It's just reference cycles that contain the value from ssl.create_default_context(). To me this looks like a case of relatively few objects involved in a reference cycle keeping quite a bit of memory in use. A single value from create_default_context() uses 800 kB, at least on my machine. So it doesn't take many Response->Stream cycles to chew up a lot of memory.

import gc
import ssl


def get_rss():
    with open(f"/proc/self/status") as fp:
        for line in fp:
            if line.startswith("VmRSS"):
                rss = line.split(":", 1)[1].rstrip()
                rss = rss.removesuffix(" kB")
                return int(rss)
    return None


class Cell:
    pass


def main():
    for i in range(100_000):
        c = Cell()
        c.ctx = ssl.create_default_context()
        c.cell = c
        if (i % 100) == 0:
            print(f'{get_rss():,} kB')
            # gc.collect()


if __name__ == '__main__':
    main()