【SparkData 博客文章】 A Brief Summary of Advantages and Dr (转载)# DataSciences - 数据科学
d*9
1 楼
【 以下文字转载自 Statistics 讨论区 】
发信人: dido2009 (music fan), 信区: Statistics
标 题: 【SparkData 博客文章】 A Brief Summary of Advantages and Drawbacks of Python
发信站: BBS 未名空间站 (Fri Nov 13 11:30:53 2020, 美东)
Python is the most popular programming language nowadays. Based on my own
experience, here is a list of advantages and limitations/drawbacks of Python:
Advantages:
1. Easy to learn: Simplicity makes the learning curve of Python is low.
Basically one can pick it up from a few days to a few weeks.
2. Versatile: Python code can be found almost everywhere, such as web
applications, desktop applications, data science, big data, spark, cloud,
etc.
3. Flexible: Flexible syntax makes its code easy to write.
4. Python is a hybrid of several programming language types. It supports
Object-oriented programming, Procedural programming, and Functional
programming.
5. Generator is a very useful feature.
6. Python has a comprehensive ecosystem, supporting community, extensive
libraries, and 3rd party modules.
Limitations or Drawbacks:
1. Some syntax brings inconvenience when writing code or even makes its code
error-prone. Example A: Python code uses indentation to define scope,
instead of brackets. Example B: If you want to call a function, the
definition of the function must be defined before the caller, otherwise the
function cannot be found.
2. Python is interpreted instead of compiled. You may modify code on-the-fly
, but the result may be not what you expected, as the old code is cached.
3. Python is not strongly typed language. Type is determined at run-time
only. Thus it is not type-safe before running.
4. As an interpreted programming language, Python’s performance is not as
good as compiled ones, such as C-family languages (C, C++, Java, C#, etc.).
Usually it is slower.
5. By design, there is no explicit pass-by-reference parameter. Although
there are some mechanisms to mimic pass-by-reference implicitly, for example
, use dictionary as a parameter. Such implicit pass-by-reference behavior
may confuse people. People need to be very clear what data type is pass-by-
value, and what data type is pass-by-reference.
6. If you want to write multi-threading code, probably Python is not a good
choice for it.
===================================
有空来坐坐，SparkData 欢迎您！
http://www.sparkdata.ca
入门介绍，行业科普
技术讨论，技能培训
职业规划，工作信息
====================================
发信人: dido2009 (music fan), 信区: Statistics
标 题: 【SparkData 博客文章】 A Brief Summary of Advantages and Drawbacks of Python
发信站: BBS 未名空间站 (Fri Nov 13 11:30:53 2020, 美东)
Python is the most popular programming language nowadays. Based on my own
experience, here is a list of advantages and limitations/drawbacks of Python:
Advantages:
1. Easy to learn: Simplicity makes the learning curve of Python is low.
Basically one can pick it up from a few days to a few weeks.
2. Versatile: Python code can be found almost everywhere, such as web
applications, desktop applications, data science, big data, spark, cloud,
etc.
3. Flexible: Flexible syntax makes its code easy to write.
4. Python is a hybrid of several programming language types. It supports
Object-oriented programming, Procedural programming, and Functional
programming.
5. Generator is a very useful feature.
6. Python has a comprehensive ecosystem, supporting community, extensive
libraries, and 3rd party modules.
Limitations or Drawbacks:
1. Some syntax brings inconvenience when writing code or even makes its code
error-prone. Example A: Python code uses indentation to define scope,
instead of brackets. Example B: If you want to call a function, the
definition of the function must be defined before the caller, otherwise the
function cannot be found.
2. Python is interpreted instead of compiled. You may modify code on-the-fly
, but the result may be not what you expected, as the old code is cached.
3. Python is not strongly typed language. Type is determined at run-time
only. Thus it is not type-safe before running.
4. As an interpreted programming language, Python’s performance is not as
good as compiled ones, such as C-family languages (C, C++, Java, C#, etc.).
Usually it is slower.
5. By design, there is no explicit pass-by-reference parameter. Although
there are some mechanisms to mimic pass-by-reference implicitly, for example
, use dictionary as a parameter. Such implicit pass-by-reference behavior
may confuse people. People need to be very clear what data type is pass-by-
value, and what data type is pass-by-reference.
6. If you want to write multi-threading code, probably Python is not a good
choice for it.
===================================
有空来坐坐，SparkData 欢迎您！
http://www.sparkdata.ca
入门介绍，行业科普
技术讨论，技能培训
职业规划，工作信息
====================================