Here's a list of features necessary for me to switch to this library as a daily driver.
[x] Parse the snippet inside expand_snippet(snippet). The argument should either be a snippet AST, or a string to be parsed.
[ ] Handle escapes in the take_until combinator. Namely \$, \\\$ etc. Need scan backwards from the match and count the number of slashes (odd = escaped).
[ ] Match the indentation level. Figure out the current indentation level, then apply it to all the snippet lines.
[ ] Handle those internal stops being overwritten (TextChanged should remove entries that have a nil range == missing marks)
[ ] Same thing for stops being deleted by the user (filter these inside jump)
[ ] Handle $0, the last tabstop to jump to when finalize_snippet() is called. Should not be reachable if jumping from $1 backwards.
[ ] Snippet loading mechanism, check how to integrate honza/vim-snippets. Need to populate into memory based on filetype, parsing out the .snippets file format.
[ ] Correctly handle expanding a second snippet while modifying a tabstop. VSCode merges the snippet into place, inserting tabstops at the current position, that way we can correctly jump back and forth over tabstops from both snippets.
Here's a list of features necessary for me to switch to this library as a daily driver.
expand_snippet(snippet). The argument should either be a snippet AST, or a string to be parsed.take_untilcombinator. Namely\$,\\\$etc. Need scan backwards from the match and count the number of slashes (odd = escaped).$0, the last tabstop to jump to whenfinalize_snippet()is called. Should not be reachable if jumping from$1backwards..snippetsfile format.Lastly
expand_snippet(string)