Re: Lark - A surface syntax for Common Lisp
"Greg Menke (as gregm32768 at comcast dot net)" <[email protected]>
| Newsgroups | gmane.lisp.lispworks.general |
|---|---|
| Message-ID | <[email protected]> |
On 2/16/2026 5:11 PM, Tim Bradshaw wrote: > > >> On 16 Feb 2026, at 13:32, Greg Menke (as gregm32768 at comcast dot >> net) <[email protected]> wrote: >> >> But how can an editor handle substantial levels of nesting cleanly? >> It is ever so easy to add/delete a space even by accident and then >> things get ambiguous. > > Even if that were true which in my experience it is not when using any > slightly competent editor, it is surprisingly hard for this to happen > in Python for instance. > > * You can only increase indentation where the syntax allows it, and > then it is not optional to do so. > * Assuming you use n-space indentation steps then a line which is > outdented by 1 to n-1 spaces from its expected indentation is a > syntax error. > * You are not allowed to use any mixture of tabs and spaces for > indentation which depends on the width of a tab, which in effect > outlaws any indentation scheme which is not either pure tabs or > pure spaces. > > The people who may find this hard are those who use single-space > indent steps. OK. > > --tim I still don't get it, not trying to be difficult I just don't understand how python is not fragile. I don't use single-space indents, use Emacs and 3 or 4 spaces for indents. Yet its crazy easy to accidentally move code following a conditional up into the conditional, or bring code out of the conditional just by accidentally hitting a tab, and then ambiguity can spread if the region of code is deeply nested. The block move keystrokes are a lifesaver, but I keep getting into little indenting problems when code is complex. Its not a problem I ever have with C/C++ or Lisp, or Java, heaven help us.