What is the best way to read in a txt file that is organized with different sections from top to bottom?

Question

Jake Marvin 2017-9-11

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/356117-what-is-the-best-way-to-read-in-a-txt-file-that-is-organized-with-different-sections-from-top-to-bot

编辑： José-Luis 2017-9-11

I currently have been using the textscan function to read in a set of data. The problem is that will result in lots of hard coding and delimiters for sorting through the data and assigning variables. I am interested in using the readtable function but that reads text from left to right.

Note- Each section name on a new line starts with '#'. To give a preview of the data:

# number of x points
  2
#coordinates of points
# x     y     z
 0.0  0.1   0.2
 0.4  0.3   0.6

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

José-Luis 2017-9-11

编辑：José-Luis 2017-9-11

Do you generate the text file yourself? If yes, then you could put it in whatever structured format you want and avoid this problem. There are very many possibilities and the better approach depends on the nature of your problem: xml, json, databases. There's even an API to create .mat files from other languages.

In this case, you might need a custom parser.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Guillaume 2017-9-11

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/356117-what-is-the-best-way-to-read-in-a-txt-file-that-is-organized-with-different-sections-from-top-to-bot#answer_281112

readtable is not designed for this kind of files and will not work.

With these kind of file, unfortunately, you do have to write your own parser, using lower level functions such as textscan, sscanf, and co.

It does not mean that you can't avoid most hard coding if you design your code efficiently. If you show us your current code, we may be able to suggest improvements. Note that the most important thing is defining correctly the grammar of the text file.