At the end of the year, Tryo Labs launched a list of the top ten Python libraries in 2016 as it did last year. For the screening criteria for this list, Tryo Labs wrote: We have avoided the already successful projects of Django, Flask, etc. that have become today's standard libraries. In addition, the libraries in this list were created before 2016, but they have seen a surge in popularity this year or we think they are very good so they can enter this list. Here is the list details:
Zappa
Link: https://
Since the release of AWS Lambda (and other subsequent projects), people's focus has shifted to serverless architecture. These architectures allow us to deploy microservices to the cloud and deploy them into a fully manageable environment; in such an environment, people don't care about any servers, they just need to allocate stateless, short-lived compute containers. (computing container) - a single service provider can be fully managed. Through this paradigm, events (such as traffic spikes) can trigger the execution of more of these containers, so it is possible to handle "infinite" horizontal expansion.
Zappa is a serverless framework for Python, although (at least for now) it only supports AWS Lambda and AWS API Gateway. It makes it easy to develop applications like this, freeing you from the tedious configuration of using AWS Console or APIs, and it has a variety of commands to simplify deployment and manage different environments.
2. Sanic + uvloop
link:
Sanic: https://github.com/channelcat/sanic
Uvloop: https://magic.io/blog/uvloop-blazing-fast-python-networking/
Who says Python can't be fast? Sanic is not only the best software library name ever, but also the fastest Python web framework ever, and it seems to be far more than other frameworks. It is a Flask Python 3.5+ web server designed for speed. Another library, uvloop, is an ultra-fast plug-in replacement for asyncio's event loop (which uses libuv at its bottom). These two add up is a powerful combination!
According to Sanic's author's benchmark, uvloop can drive Sanic to process more than 33,000 requests per second, which is too strong! (faster than node.js). Your code can benefit from this new async/await syntax -- they look neat; we also like the Flask-style API. You must try Sanic, and if you are also using asyncio, you can benefit from uvloop without much modification to your code.
3. asyncpg
Link: https://github.com/MagicStack/asyncpg
Following the latest developments in the asyncio framework, people from MagicStack brought us this highly efficient asynchronous (currently only CPython 3.5) database interface library, which was designed specifically for PostgreSQL. It has zero correlation, which means there is no need to install libpq. In contrast, withpsycopg2 (the most popular Python PostgreSQL adapter) needs to exchange data with the database server in text format; asyncpg implements the PostgreSQL binary I/O protocol, which allows it to support not only generic types, but also many other features. The benefits.
The benchmark is clear: asyncpg is at least three times faster than psycopg2 (or aiopg) and faster than node.js and Go.
4. boto3
Link: https://github.com/boto/boto3
If your infrastructure is deployed on AWS or uses their services (like S3), then you should be happy to see that boto (the Python interface for the AWS API) has been completely rewritten from beginning to end. And you don't have to completely migrate your app in one go: you can use both boto3 and boto(2) at the same time; for example, use boto3 only in new parts of your application.
This new implementation will be much more consistent across different services, and because it uses a data-driven approach to generating classes from a JSON description file at runtime, it can always be updated quickly. No longer have to lag behind the new Amazon API features, use bot3!
Ic Data Acquisition,Data Acquisition Analog Front End,Data Acquisition Analog To Digital Converters,Data Acquisition Digital To Analog
Shenzhen Kaixuanye Technology Co., Ltd. , https://www.icoilne.com