Over the years, the urllib3 HTTP client maintained its SSL cipher suites based on Hynek's evergreen list. However, with the release of urllib3 version 2.0, we decided to require OpenSSL 1.1.1 or above and rely on the default cipher suites, which are secure starting with OpenSSL 1 …
read moreThere are comments.
urllib3 is a HTTP client written in Python. We just got 7 billions downloads, 1 million GitHub repositories depend on us, and popular libraries like requests and boto3 use us for performing secure HTTP requests.
We’re currently working on a v2 version that is more modern and secure. It …
read moreThere are comments.
urllib3 is a HTTP client written in Python. We get millions of downloads every day, 500k GitHub repositories depend on us, and popular libraries like requests and boto3 use us for performing secure HTTP requests. In short, urllib3 is Python’s most depended-on HTTP client library.
We’re currently working …
read moreThere are comments.
I spent the last five days working full-time on urllib3, currently the most downloaded Python package, and a dependency of requests. I can’t believe I’m writing this sentence! But it’s true. A dream come true. This was made possible thanks to:
There are comments.
asyncio has brought support for asynchronous I/O with an event loop to
Python 3. Since Python 3.5, we can use the async and await
keywords:
async def get(self):
await self.wait_for_token()
return self.client.get()
But Python 3.5 also brought support for asynchronous context managers
(async …
There are comments.
If you're coming from a JavaScript background, it's tempting to try to use the promises that you know and love with Python. That's what I tried to initially too, and I was surprised to see that promises were very rarely used in Python. It turns out that promises are not …
read moreThere are comments.
When you first switch away from synchronous sequential code, you realize that you can send hundreds of HTTP requests per second because you don't wait for one request to finish before sending the next one. These requests can be anything, but I focus on HTTP requests in this post because …
read moreThere are comments.
At work, I recently wanted to visualize how we annotated documents that we store as JSON in Elasticsearch. We annotate substrings with different codes, and I wanted to show those codes using colors in a terminal. This basically boils down to outputting JSON with ANSI escape codes. The easier part …
read moreThere are comments.
One of the first hurdles that you can encounter when trying out asyncio is "asyncio eats all my memory!". Indeed, to keep your CPU busy, you're encouraged to launch a lot of coroutines simultaneously. Coroutines don't use a lot of memory by themselves, but what you're doing inside them can …
read moreThere are comments.
Since I've began to program, I've heard everywhere that synchronous, sequential programming is a way of the past. In a manycore future, we will need to write programs that execute in parallel. But aside from embarassingly parallel programs, this has always been considered to be too difficult to be worth …
read moreThere are comments.