12. Programming with Larch

This chapter describes some of the key concepts for programming with Larch. Larch is a fairly complete programming language for data analysis, but it is also easy to extend Larch with new functionality. There are two distinct ways to do this. First, modules are files of Larch or Python code that can be imported and used from Larch. These are primarily intended to organize programming text, and to allow re-usable code. Second, plugins are Python modules that have special methods to export the functionality to Larch. Whereas modules run from within the Larch interpreter, plugins are run in Python, and can closely interact with a running Larch interpreter to add or manipulate the existing session. Plugins should be the main approach for extending the capabilities of Larch, and indeed essentially all the real scientific code in Larch is implemented as plugins. As an important note, the functions defined in the Larch can be accessed from Python, without relying on the syntax of the Larch language at all.

The reader here is assumed to be familiar with computer programming, and to have some exposure to working with Larch and/or Python, at least from the Tutorial (Chapter 5). While the present chapter is aimed primarily at developers and advanced users, Larch (and Python) makes such developments easy enough that neither extensive effort nor training are required to begin writing programs with Larch.

• 12.1. Differences between Larch and Python
  • 12.1.1. Code Block Ends
  • 12.1.2. Groups vs Modules
  • 12.1.3. Symbol Lookup Rules
  • 12.1.4. Unimplemented features
• 12.2. Modules
• 12.3. Plugins
  • 12.3.1. Locations of Plugins
  • 12.3.2. Customizing Plugins and Folders of Plugins
• 12.4. Using Larch from Python