* Fix for PartialStyleSheet getting stuck in loop
Adds a clause for detecting `//` as a comment marker when working backwards through a file, preventing the parser from getting stuck when the comment starts at 1:1
* Modify change to detect start of file instead of skipping additional `/` characters now that we know the issue only occurs at offset 0.
Co-authored-by: Tyler McEnaney <tharbakim@hotmail.com>
For match tag, update tag, and balance in/out, we now have the following:
- vscode-html-languageservice parse is being used to speed up file parse times
- Balance balance-in after balance-out behaviour is now more consistent
- These commands now work in markup files with unclosed tags
- Implemented a cache for these commands to prevent multiple document parses on the same version of the document
* Enfoce node boundries in places where it matters
* Flip defualt emmet includeBoundries to true
Explicitly set getNode's include boundries where needed
* Remove defualt papameter
* Add update image boundry tests
* Add tests for bondries on some of the tag actions
* Rest of tag tests
* Allow Emmet "Go to matching pair" to work inside script tags
* Refactor template script detection and node extraction.
* Futher refactoring of Emmet template identification logic.
* Emmet: select entire script tag when unable to idnetify an inner element
* Remove redundant nullity check
* Refactor parsePartialStylesheet
* More refactoring
* To avoid errors, parse line comments char by char and no when getting to previous line.
* Revert "To avoid errors, parse line comments char by char and no when getting to previous line."
This reverts commit f353dba4c7.
* Fix for inline comments inside block comments
* First draft at partial parsing of css for emmet
* Partial parsing of stylesheets for Emmet
* Addressing feedback
* Refactoring and fixing little bug
* Liitle fix
* Equals are now equals
* We don't want emmet in selectors
* Skip comments when preparsing
* Addressing feedback, refactoring and cleaning up code.
* Update endPosition only if not eof
* Use function to find closing comment
* Don't check more than 5000 characters back
