Skip to content

Latest commit

 

History

History
47 lines (41 loc) · 2.34 KB

04-latex.rst

File metadata and controls

47 lines (41 loc) · 2.34 KB
Raw

Best practice with LaTeX

Objectives

  • Compile LaTeX from command line;
  • Have a consistent and meaningful directory structure for LaTeX;
  • Inputting software outputs automatically/directly in to LaTeX.

Plan

  1. Discuss LaTeX: ask if everyone has used it before? If anyone hasn't perhaps make sure a tutor is closer to them. For those who have used it before, explain that we're going to not talk about how to write specific things but talk about best practice.

  2. Open up editor and create the basic file saved in rsd-workshop: compile it.

  3. Describe calling one document from another. Explain why we do this: clearer directory structure and further modularisation. Helps separate: content (tex/) from instructions (main.tex).

  4. Talk about the directory structure: move all python files to src. Create/copy the file to create a scatter plot and show how to run that and use output directly in code. Note the name of the python: same name for pdf graph, same name for label in latex document. This helps ensure users won't be left wondering where one thing comes from.

  5. Go through other examples. For all of these, invite participants to download the files: no need to write them. The purpose of this workshop is not to teach advanced Python, but emphasise that similar functionality exists in most languages. (For example, if you don't have functionality to write the LaTeX code for a table directly you can just use the programming language to write the LaTeX using for loops etc...)

  6. Discuss how editors can be used for various tasks. In this workshop we have covered two, writing code and LaTeX. Editors come with plugins to help users with their different tasks, for LaTeX the following plugins exist:

FAQ

  • Question: Why do can I not use Word? Answer: Using a GUI implies a lot of the things we've discussed here is not possible (dynamically reading input from file for example) and thus leaves the possibility for human error. Also: portability, sharing research using a closed commercial solution implies that it can only be consumed by those that have the same closed commercial solution.