I'm getting contacted occasionally by people who look up the
package's author from npm's registry. Package.json doesn't have a
dedicated 'maintainer' field, so 'author' should work, I think.
* first pass at accessibility report in checkTheme
* resolve CSS color names when necessary
* (enh) default Dark theme is WCAG AAA wrt contrast
* (enh) improve accessibility of default theme
- drops the legacy `lib/highlight.js`
- build ES modules in `es` folder side-by-side with `lib` for CJS
- adds `lib/common`
- loader is no longer supported, languages must `export default` or equiv
- add --no-esm build option (for speed)
- add `.js.js` stubs for maximum compatible with requires using extensions
- Move AUTHORS.txt to CONTRIBUTORS.md
- Removes the manually maintained list of contributors
- Links to GitHub contributor graph information instead
- Update README links
- `contributors` in package.json is now core team (current and alumni)
- remove this step from the pull request template
* enh(tests) analyze regex for catastrophic backtracking
* allow testing individual languages
* fix(routeros) resolve potential exponential backtracking issue
* fix(powershell) resolve potential exponential backtracking issue
* fix(erlang-repl) resolve backtracking issue
* fix(r) resolve backtracking issue
* fix(jboss-cli) resolve backtracking issue
* (lint) perl
* fix(perl) resolve exponential backtracking issue
* fix(gams) resolve exponential backtracking issue
* (lint) c-like
* fix(handlebars) resolve exponential backtracking issue
* fix(cpp) resolve exponential backtracking issue
* fix(sqf) fix poly backtracking issue
- The `_` case should already be handled by `_+`
* fix(xquery) fix poly backtracking issue
* fix(ruleslanguage) fix poly backtracking issue
The extra expression here does not matter since it was purely
optional (`*`). The important thing is gobbling up the `#` to prevent
it from stealing relevance. And `\s+` gets that done on it's own.
* fix(markdown) fix poly backtracking issue
- Fix poly backtracking with code blocks
* fix(dsconfig) fix poly backtracking issue
Not 100% sure what the original intent of \S was since it seems that
valueless properties wouldn't have a trailing `:` and why should we
include the letter after as part of the match?
That means that now there is an edge case that we handle slightly
differently
property:"quoted value"
The `:` will not be highlighted (as it was before). This is a
problem for another day when someone who knows about dsconfig
shows up.
* fix(x86asm) fix poly backtracking issue
* fix(yaml) fix poly backtracking issue
* fix(livecodeserver) fix poly backtracking issue
* fix(fortran) fix poly backtracking issue
* fix(fortran/irpf90) fix poly backtracking issue
* fix(ruby) fix poly backtracking issues
- Fix poly backtracking issue with RVM_PROMPT.
- Fix poly backtracking issue with heredocs.
* fix(ebnf) fix poly backtracking issue
* fix(basic) fix poly backtracking issue
* fix(elixir) fix poly backtracking issue
* fix(crystal) fix poly backtracking issue
* fix(scilab) fix poly backtracking issue
* fix(csharp) fix poly backtracking issue
* fix(coffee/livescript) fix poly backtracking issue
- Fix issue with optional params for anonymous functions
* fix(moonscript) fix poly backtracking issue
* fix(aspectj) fix poly backtracking issue
* fix(d) fix poly backtracking issue
* fix(gcode) fix poly backtracking issue
* fix(kotlin) fix poly backtracking issue
No explanation for what this `illegal` is trying to accomplish to
without that data, just remove it.
* fix(kotlin) fix poly backtracking issue
- Use same numeric mode rules as for Java
* fix(asciidoc) fix poly backtracking issue
* fix(javascript/typescript) fix poly backtracking issue
- Fix poly backtracking issue in gnarly `()` counting regex
* fix(latex) fix poly backtracking issue
* fix(reasonml) fix poly backtracking issue
- fix typo/bug with using `s` vs `\s` (string vs regex mistake)
- simply `[pattern]?[pattern]?` to just `[pattern]{0,2}`
- fix ambiguous `\s*` poly issues
* enh(ci): Add tests for polynomial regex issues
Stephen Li