Resync the lexers with Scintillua

- Resync the lexers with Scintillua

- Update the lexer readme

- Update `zenburn` theme to fix some highlighting issues

- lexers: redirect print function to vis:info()

- Fix support for custom style names

- As per error message "lexer.delimited_range() is deprecated, use lexer.range()".

- Remove remaining `lexer.delimited_range()` call

- Set syntax to `nil` if the file type has no matching lexer

- Updated Go lexer for Go 1.18.

- lexers/dsv: convert to new lexer format

  (cherry picked from commit 9edbc3cd9ea1d7142b1305840432a3d2739e755a)

- lexers/gemini: disable legacy gemini lexer

  This reverts commit 468f9ee1b027a7ce98b1a249fa1af5888feeb989.

  It is in legacy format and of questionable quality. Ideally it
  should be contributed upstream from where it will eventually
  trickle down to us.

- lexers/git-rebase: convert to new lexer format

  (cherry picked from commit 4000a4cc9ac4a4c2869dfae772b977a82aee8d8c)

- lexers/strace: convert to new lexer format

  (cherry picked from commit e420451320d97eb164f5629c1bcfab0b595be29d)

- lexers/typescript: add new upstream lexer revision 28e2b60

  (cherry picked from commit 7326e6deecdaa75fa94ae9ebdb653f9f907b33f2)

- use `package.searchpath` instead of a local `searchpath` function

- Restore `filetype: support filetype detection via hashbang`

- Remove redundant comment

- Restore gemini lexer

1 files changed, 62 insertions, 74 deletions

diff --git a/lua/lexers/wsf.lua b/lua/lexers/wsf.lua
index 37cb33e..6972cfe 100644
--- a/lua/lexers/wsf.lua
+++ b/lua/lexers/wsf.lua
@@ -1,102 +1,90 @@
--- Copyright 2006-2017 Mitchell mitchell.att.foicica.com. See LICENSE.
+-- Copyright 2006-2022 Mitchell. See LICENSE.
 -- WSF LPeg lexer (based on XML).
 -- Contributed by Jeff Stone.
 
-local l = require('lexer')
-local token, word_match = l.token, l.word_match
-local P, R, S, V = lpeg.P, lpeg.R, lpeg.S, lpeg.V
+local lexer = require('lexer')
+local token, word_match = lexer.token, lexer.word_match
+local P, S = lpeg.P, lpeg.S
 
-local M = {_NAME = 'wsf'}
+local lex = lexer.new('wsf')
 
 -- Whitespace.
-local ws = token(l.WHITESPACE, l.space^1)
+local ws = token(lexer.WHITESPACE, lexer.space^1)
+lex:add_rule('whitespace', ws)
 
 -- Comments.
-local comment = token(l.COMMENT, '<!--' * (l.any - '-->')^0 * P('-->')^-1)
-
--- Strings.
-local sq_str = l.delimited_range("'", false, true)
-local dq_str = l.delimited_range('"', false, true)
-local string = #S('\'"') * l.last_char_includes('=') *
-               token(l.STRING, sq_str + dq_str)
-
-local in_tag = #P((1 - S'><')^0 * '>')
-
--- Numbers.
-local number = #l.digit * l.last_char_includes('=') *
-               token(l.NUMBER, l.digit^1 * P('%')^-1) * in_tag
-
-local alpha = R('az', 'AZ', '\127\255')
-local word_char = l.alnum + S('_-:.??')
-local identifier = (l.alpha + S('_-:.??')) * word_char^0
+lex:add_rule('comment', token(lexer.COMMENT, lexer.range('<!--', '-->')))
 
 -- Elements.
+local alpha = lpeg.R('az', 'AZ', '\127\255')
+local word_char = lexer.alnum + S('_-:.?')
+local identifier = (alpha + S('_-:.?')) * word_char^0
 local element = token('element', '<' * P('/')^-1 * identifier)
-
--- Attributes.
-local attribute = token('attribute', identifier) * #(l.space^0 * '=')
+lex:add_rule('element', element)
+lex:add_style('element', lexer.styles.keyword)
 
 -- Closing tags.
 local tag_close = token('element', P('/')^-1 * '>')
+lex:add_rule('tag_close', tag_close)
+
+-- Attributes.
+local attribute = token('attribute', identifier) * #(lexer.space^0 * '=')
+lex:add_rule('attribute', attribute)
+lex:add_style('attribute', lexer.styles.type)
 
 -- Equals.
-local equals = token(l.OPERATOR, '=') * in_tag
+local in_tag = P(function(input, index)
+  local before = input:sub(1, index - 1)
+  local s, e = before:find('<[^>]-$'), before:find('>[^<]-$')
+  if s and e then return s > e and index or nil end
+  if s then return index end
+  return input:find('^[^<]->', index) and index or nil
+end)
+
+local equals = token(lexer.OPERATOR, '=') * in_tag
+lex:add_rule('equals', equals)
+
+-- Strings.
+local sq_str = lexer.range("'", false, false)
+local dq_str = lexer.range('"', false, false)
+local string = #S('\'"') * lexer.last_char_includes('=') * token(lexer.STRING, sq_str + dq_str)
+lex:add_rule('string', string)
+
+-- Numbers.
+local number = token(lexer.NUMBER, lexer.dec_num * P('%')^-1)
+lex:add_rule('number', #lexer.digit * lexer.last_char_includes('=') * number * in_tag)
 
 -- Entities.
-local entity = token('entity', '&' * word_match{
-  'lt', 'gt', 'amp', 'apos', 'quot'
-} * ';')
-
-M._rules = {
-  {'whitespace', ws},
-  {'comment', comment},
-  {'element', element},
-  {'tag_close', tag_close},
-  {'attribute', attribute},
-  {'equals', equals},
-  {'string', string},
-  {'number', number},
-  {'entity', entity}
-}
-
-M._tokenstyles = {
-  element = l.STYLE_KEYWORD,
-  attribute = l.STYLE_TYPE,
-  entity = l.STYLE_OPERATOR
-}
-
-M._foldsymbols = {
-  _patterns = {'</?', '/>', '<!%-%-', '%-%->'},
-  element = {['<'] = 1, ['/>'] = -1, ['</'] = -1},
-  [l.COMMENT] = {['<!--'] = 1, ['-->'] = -1},
-}
+lex:add_rule('entity', token('entity', '&' * word_match('lt gt amp apos quot') * ';'))
+lex:add_style('entity', lexer.styles.operator)
+
+-- Fold points.
+local function disambiguate_lt(text, pos, line, s) return not line:find('^</', s) and 1 or -1 end
+lex:add_fold_point('element', '<', disambiguate_lt)
+lex:add_fold_point('element', '/>', -1)
+lex:add_fold_point(lexer.COMMENT, '<!--', '-->')
 
 -- Finally, add JavaScript and VBScript as embedded languages
 
 -- Tags that start embedded languages.
-M.embed_start_tag = element *
-                    (ws^1 * attribute * ws^0 * equals * ws^0 * string)^0 *
-                    ws^0 * tag_close
-M.embed_end_tag = element * tag_close
+local embed_start_tag = element * (ws^1 * attribute * ws^0 * equals * ws^0 * string)^0 * ws^0 *
+  tag_close
+local embed_end_tag = element * tag_close
 
 -- Embedded JavaScript.
-local js = l.load('javascript')
+local js = lexer.load('javascript')
 local js_start_rule = #(P('<script') * (P(function(input, index)
-  if input:find('^%s+language%s*=%s*(["\'])[jJ][ava]*[sS]cript%1', index) then
-    return index
-  end
-end) + '>')) * M.embed_start_tag -- <script language="javascript">
-local js_end_rule = #('</script' * ws^0 * '>') * M.embed_end_tag -- </script>
-l.embed_lexer(M, js, js_start_rule, js_end_rule)
+  if input:find('^%s+language%s*=%s*(["\'])[jJ][ava]*[sS]cript%1', index) then return index end
+end) + '>')) * embed_start_tag -- <script language="javascript">
+local js_end_rule = #('</script' * ws^0 * '>') * embed_end_tag -- </script>
+lex:embed(js, js_start_rule, js_end_rule)
 
 -- Embedded VBScript.
-local vbs = l.load('vbscript')
+local vbs = lexer.load('vb', 'vbscript')
 local vbs_start_rule = #(P('<script') * (P(function(input, index)
-  if input:find('^%s+language%s*=%s*(["\'])[vV][bB][sS]cript%1', index) then
-    return index
-  end
-end) + '>')) * M.embed_start_tag -- <script language="vbscript">
-local vbs_end_rule = #('</script' * ws^0 * '>') * M.embed_end_tag -- </script>
-l.embed_lexer(M, vbs, vbs_start_rule, vbs_end_rule)
-
-return M
+  if input:find('^%s+language%s*=%s*(["\'])[vV][bB][sS]cript%1', index) then return index end
+end) + '>')) * embed_start_tag -- <script language="vbscript">
+local vbs_end_rule = #('</script' * ws^0 * '>') * embed_end_tag -- </script>
+lex:embed(vbs, vbs_start_rule, vbs_end_rule)
+
+return lex