@@ -29,45 +29,60 @@ let s:julia_spellcheck_comments = get(g:, "julia_spellcheck_comments", 1)
2929
3030let s: julia_highlight_operators= get (g: " julia_highlight_operators" 1 )
3131
32- " characters which cannot be used in identifiers. This list is very incomplete:
33- " 1) it only cares about charactes below 256
34- " 2) it doesn't distinguish between what's allowed as the 1st char vs in the
35- " rest of an identifier (e.g. digits and `!`)
36- " Despite these shortcomings, it seems to do a decent job.
37- " note: \U5B and \U5D are '[' and ']'
38- let s: nonid_chars= " \U01 -\U07 " " \U0E -\U1F "
39- \ " ?\" #$'(,.:;=@`\\ U5B{"
40- \ " \U80 -\UA1 " " \UA7\UA8\UAB\UAD\UAF\UB4 " " \UB6 -\UB8 " " \UBB\UBF "
41-
42- let s: nonidS_chars= " [:space:])\\ U5D}" s: nonid_chars
43-
44- " the following excludes '!' since it can be used as an identifier,
45- " and '$' since it can be used in interpolations
46- " note that \U2D is '-'
47- let s: uniop_chars= " +\\ U2D~¬√∛∜⋆"
48-
49- let s: binop_chars= " =+\\ U2D*/\\ %÷^&|⊻<>≤≥≡≠≢∈∉⋅×∪∩⊆⊈⊂⊄⊊←→∋∌⊕⊖⊞⊟∘∧⊗⊘↑↓∨⊠±⟂⋆"
50-
51- " the following is a list of all remainig valid operator chars,
52- " but it's more efficient when expressed with ranges (see below)
53- " let s:binop_chars_extra = "↔↚↛↠↣↦↮⇎⇏⇒⇔⇴⇶⇷⇸⇹⇺⇻⇼⇽⇾⇿⟵⟶⟷⟷⟹⟺⟻⟼⟽⟾⟿⤀⤁⤂⤃⤄⤅⤆⤇⤌⤍⤎⤏⤐⤑⤔⤕⤖⤗⤘⤝⤞⤟⤠⥄⥅⥆⥇⥈⥊⥋⥎⥐⥒⥓⥖⥗⥚⥛⥞⥟⥢⥤⥦⥧⥨⥩⥪⥫⥬⥭⥰⧴⬱⬰⬲⬳⬴⬵⬶⬷⬸⬹⬺⬻⬼⬽⬾⬿⭀⭁⭂⭃⭄⭇⭈⭉⭊⭋⭌￩￫" .
54- " \ "∝∊∍∥∦∷∺∻∽∾≁≃≄≅≆≇≈≉≊≋≌≍≎≐≑≒≓≔≕≖≗≘≙≚≛≜≝≞≟≣≦≧≨≩≪≫≬≭≮≯≰≱≲≳≴≵≶≷≸≹≺≻≼≽≾≿⊀⊁⊃⊅⊇⊉⊋⊏⊐⊑⊒⊜⊩⊬⊮⊰⊱⊲⊳⊴⊵⊶⊷⋍⋐⋑⋕⋖⋗⋘⋙⋚⋛⋜⋝⋞⋟⋠⋡⋢⋣⋤⋥⋦⋧⋨⋩⋪⋫⋬⋭⋲⋳⋴⋵⋶⋷⋸⋹⋺⋻⋼⋽⋾⋿⟈⟉⟒⦷⧀⧁⧡⧣⧤⧥⩦⩧⩪⩫⩬⩭⩮⩯⩰⩱⩲⩳⩴⩵⩶⩷⩸⩹⩺⩻⩼⩽⩾⩿⪀⪁⪂⪃⪄⪅⪆⪇⪈⪉⪊⪋⪌⪍⪎⪏⪐⪑⪒⪓⪔⪕⪖⪗⪘⪙⪚⪛⪜⪝⪞⪟⪠⪡⪢⪣⪤⪥⪦⪧⪨⪩⪪⪫⪬⪭⪮⪯⪰⪱⪲⪳⪴⪵⪶⪷⪸⪹⪺⪻⪼⪽⪾⪿⫀⫁⫂⫃⫄⫅⫆⫇⫈⫉⫊⫋⫌⫍⫎⫏⫐⫑⫒⫓⫔⫕⫖⫗⫘⫙⫷⫸⫹⫺⊢⊣" .
55- " \ "⊔∓∔∸≂≏⊎⊽⋎⋓⧺⧻⨈⨢⨣⨤⨥⨦⨧⨨⨩⨪⨫⨬⨭⨮⨹⨺⩁⩂⩅⩊⩌⩏⩐⩒⩔⩖⩗⩛⩝⩡⩢⩣" .
56- " \ "⊙⊚⊛⊡⊓∗∙∤⅋≀⊼⋄⋇⋉⋊⋋⋌⋏⋒⟑⦸⦼⦾⦿⧶⧷⨇⨰⨱⨲⨳⨴⨵⨶⨷⨸⨻⨼⨽⩀⩃⩄⩋⩍⩎⩑⩓⩕⩘⩚⩜⩞⩟⩠⫛⊍▷⨝⟕⟖⟗" .
57- " \ "⇵⟰⟱⤈⤉⤊⤋⤒⤓⥉⥌⥍⥏⥑⥔⥕⥘⥙⥜⥝⥠⥡⥣⥥⥮⥯￪￬"
58-
59- " same as above, but with character ranges, for performance
60- let s: binop_chars_extra= " \\ U214B\\ U2190-\\ U2194\\ U219A\\ U219B\\ U21A0\\ U21A3\\ U21A6\\ U21AE\\ U21CE\\ U21CF\\ U21D2\\ U21D4\\ U21F4-\\ U21FF\\ U2208-\\ U220D\\ U2213\\ U2214\\ U2217-\\ U2219\\ U221D\\ U2224-\\ U222A\\ U2237\\ U2238\\ U223A\\ U223B\\ U223D\\ U223E\\ U2240-\\ U228B\\ U228D-\\ U229C\\ U229E-\\ U22A3\\ U22A9\\ U22AC\\ U22AE\\ U22B0-\\ U22B7\\ U22BB-\\ U22BD\\ U22C4-\\ U22C7\\ U22C9-\\ U22D3\\ U22D5-\\ U22ED\\ U22F2-\\ U22FF\\ U25B7\\ U27C8\\ U27C9\\ U27D1\\ U27D2\\ U27D5-\\ U27D7\\ U27F0\\ U27F1\\ U27F5-\\ U27F7\\ U27F7\\ U27F9-\\ U27FF\\ U2900-\\ U2918\\ U291D-\\ U2920\\ U2944-\\ U2970\\ U29B7\\ U29B8\\ U29BC\\ U29BE-\\ U29C1\\ U29E1\\ U29E3-\\ U29E5\\ U29F4\\ U29F6\\ U29F7\\ U29FA\\ U29FB\\ U2A07\\ U2A08\\ U2A1D\\ U2A22-\\ U2A2E\\ U2A30-\\ U2A3D\\ U2A40-\\ U2A45\\ U2A4A-\\ U2A58\\ U2A5A-\\ U2A63\\ U2A66\\ U2A67\\ U2A6A-\\ U2AD9\\ U2ADB\\ U2AF7-\\ U2AFA\\ U2B30-\\ U2B44\\ U2B47-\\ U2B4C\\ UFFE9-\\ UFFEC"
61-
62- " a Julia identifier, sort of
63- let s: idregex= ' \%([^' s: nonidS_chars' 0-9!' s: uniop_charss: binop_chars' ][^' s: nonidS_charss: uniop_charss: binop_charss: binop_chars_extra' ]*\)'
64-
65- let s: operators= ' \%(' ' \.\%([-+*/^÷%|&!]\|//\|\\\|<<\|>>>\?\)\?='
66- \ ' \|' ' [:$<>]=\|||\|&&\||>\|<|\|<:\|>:\|::\|<<\|>>>\?\|//\|[-=]>\|\.\{3\}'
67- \ ' \|' ' \.\?[' s: uniop_chars' !]'
68- \ ' \|' ' \.\?[' s: binop_charss: binop_chars_extra' ]'
32+ " List of characters, up to \UFF, which cannot be used in identifiers.
33+ " (It includes operator characters; we don't consider them identifiers.)
34+ " This is used mostly in lookbehinds with `\@<=`, e.g. when we need to check
35+ " that that we're not in the middle of an identifier.
36+ " It doesn't include a few characters (spaces and all closing parentheses)
37+ " because those may or may not be valid in the lookbehind on a case-by-case
38+ " basis.
39+ let s: nonid_chars= ' \U00-\U08' ' \U0A-\U1F'
40+ \ ' \U21-\U28' ' \U2A-\U2F' ' \U3A-\U40' ' \U5B-\U5E' ' \U60' ' \U7B\U7C'
41+ \ ' \U7E-\UA1' ' \UA7\UA8' ' \UAB-\UAD' ' \UAF\UB1\UB4' ' \UB6-\UB8' ' \UBB\UBF' ' \UD7\UF7'
42+
43+ " The complete list
44+ let s: nonidS_chars= ' [:space:])\U5D}' s: nonid_chars
45+
46+
47+ " List of all valid operator chars up to \UFF (NOTE: they must all be included
48+ " in s:nonidS_chars, so that if we include that, then this is redundant)
49+ " It does not include '!' since it can be used in an identifier.
50+ " The list contains the following characters: '%&*+-/<=>\\^|~¬±×÷'
51+ let s: op_chars= ' \U25\U26\U2A\U2B\U2D\U2F\U3C-\U3E\U5C\U5E\U7C\U7E\UAC\UB1\UD7\UF7'
52+
53+ " List of all valid operator chars above \UFF
54+ " Written with ranges for performance reasons
55+ " The list contains the following characters: '…⁝⅋←↑→↓↔↚↛↜↝↞↠↢↣↤↦↩↪↫↬↮↶↷↺↻↼↽⇀⇁⇄⇆⇇⇉⇋⇌⇍⇎⇏⇐⇒⇔⇚⇛⇜⇝⇠⇢⇴⇵⇶⇷⇸⇹⇺⇻⇼⇽⇾⇿∈∉∊∋∌∍∓∔∗∘∙√∛∜∝∤∥∦∧∨∩∪∷∸∺∻∽∾≀≁≂≃≄≅≆≇≈≉≊≋≌≍≎≏≐≑≒≓≔≕≖≗≘≙≚≛≜≝≞≟≠≡≢≣≤≥≦≧≨≩≪≫≬≭≮≯≰≱≲≳≴≵≶≷≸≹≺≻≼≽≾≿⊀⊁⊂⊃⊄⊅⊆⊇⊈⊉⊊⊋⊍⊎⊏⊐⊑⊒⊓⊔⊕⊖⊗⊘⊙⊚⊛⊜⊞⊟⊠⊡⊢⊣⊩⊬⊮⊰⊱⊲⊳⊴⊵⊶⊷⊻⊼⊽⋄⋅⋆⋇⋉⋊⋋⋌⋍⋎⋏⋐⋑⋒⋓⋕⋖⋗⋘⋙⋚⋛⋜⋝⋞⋟⋠⋡⋢⋣⋤⋥⋦⋧⋨⋩⋪⋫⋬⋭⋮⋯⋰⋱⋲⋳⋴⋵⋶⋷⋸⋹⋺⋻⋼⋽⋾⋿⌿▷⟂⟈⟉⟑⟒⟕⟖⟗⟰⟱⟵⟶⟷⟹⟺⟻⟼⟽⟾⟿⤀⤁⤂⤃⤄⤅⤆⤇⤈⤉⤊⤋⤌⤍⤎⤏⤐⤑⤒⤓⤔⤕⤖⤗⤘⤝⤞⤟⤠⥄⥅⥆⥇⥈⥉⥊⥋⥌⥍⥎⥏⥐⥑⥒⥓⥔⥕⥖⥗⥘⥙⥚⥛⥜⥝⥞⥟⥠⥡⥢⥣⥤⥥⥦⥧⥨⥩⥪⥫⥬⥭⥮⥯⥰⦷⦸⦼⦾⦿⧀⧁⧡⧣⧤⧥⧴⧶⧷⧺⧻⨇⨈⨝⨟⨢⨣⨤⨥⨦⨧⨨⨩⨪⨫⨬⨭⨮⨰⨱⨲⨳⨴⨵⨶⨷⨸⨹⨺⨻⨼⨽⩀⩁⩂⩃⩄⩅⩊⩋⩌⩍⩎⩏⩐⩑⩒⩓⩔⩕⩖⩗⩘⩚⩛⩜⩝⩞⩟⩠⩡⩢⩣⩦⩧⩪⩫⩬⩭⩮⩯⩰⩱⩲⩳⩴⩵⩶⩷⩸⩹⩺⩻⩼⩽⩾⩿⪀⪁⪂⪃⪄⪅⪆⪇⪈⪉⪊⪋⪌⪍⪎⪏⪐⪑⪒⪓⪔⪕⪖⪗⪘⪙⪚⪛⪜⪝⪞⪟⪠⪡⪢⪣⪤⪥⪦⪧⪨⪩⪪⪫⪬⪭⪮⪯⪰⪱⪲⪳⪴⪵⪶⪷⪸⪹⪺⪻⪼⪽⪾⪿⫀⫁⫂⫃⫄⫅⫆⫇⫈⫉⫊⫋⫌⫍⫎⫏⫐⫑⫒⫓⫔⫕⫖⫗⫘⫙⫛⫷⫸⫹⫺⬰⬱⬲⬳⬴⬵⬶⬷⬸⬹⬺⬻⬼⬽⬾⬿⭀⭁⭂⭃⭄⭇⭈⭉⭊⭋⭌￩￪￫￬'
56+ let s: op_chars_wc= ' \U2026\U205D\U214B\U2190-\U2194\U219A-\U219E\U21A0\U21A2-\U21A4\U21A6\U21A9-\U21AC\U21AE\U21B6\U21B7\U21BA-\U21BD\U21C0\U21C1\U21C4\U21C6\U21C7\U21C9\U21CB-\U21D0\U21D2\U21D4\U21DA-\U21DD\U21E0\U21E2\U21F4-\U21FF\U2208-\U220D\U2213\U2214\U2217-\U221D\U2224-\U222A\U2237\U2238\U223A\U223B\U223D\U223E\U2240-\U228B\U228D-\U229C\U229E-\U22A3\U22A9\U22AC\U22AE\U22B0-\U22B7\U22BB-\U22BD\U22C4-\U22C7\U22C9-\U22D3\U22D5-\U22FF\U233F\U25B7\U27C2\U27C8\U27C9\U27D1\U27D2\U27D5-\U27D7\U27F0\U27F1\U27F5-\U27F7\U27F9-\U27FF\U2900-\U2918\U291D-\U2920\U2944-\U2970\U29B7\U29B8\U29BC\U29BE-\U29C1\U29E1\U29E3-\U29E5\U29F4\U29F6\U29F7\U29FA\U29FB\U2A07\U2A08\U2A1D\U2A1F\U2A22-\U2A2E\U2A30-\U2A3D\U2A40-\U2A45\U2A4A-\U2A58\U2A5A-\U2A63\U2A66\U2A67\U2A6A-\U2AD9\U2ADB\U2AF7-\U2AFA\U2B30-\U2B44\U2B47-\U2B4C\UFFE9-\UFFEC'
57+
58+ " Operator characters (including '!' this time). When highlighting, we mostly
59+ " can just recognize one character at a time, since longer operators are
60+ " basically made up of characters from this list. An exception is the dot
61+ " character which must be treated a bit specially.
62+ let s: operators1= ' \.\?[!' s: op_charss: op_chars_wc' ]'
63+ let s: operatorsd= ' \.\.\.\?'
64+
65+ " Full operators regex, to be used when we actually need to recognize when an
66+ " operator has ended (e.g. in quoted symbols)
67+ let s: operators= ' \%(' ' \.\%([-+*/^÷%|&⊻]\|//\|\\\|>>\|>>>\?\)\?='
68+ \ ' \|' ' [:<>]=\|||\|&&\||>\|<|\|[<>:]:\|<<\|>>>\?\|//\|[-=]>\|\.\.\.\?'
69+ \ ' \|' s: operators1
6970 \ ' \)'
7071
72+
73+ " Characters that can be used to start an identifier. Above \UBF we don't
74+ " bother checking. (If a UTF8 operator is used, it will take precedence anyway.)
75+ let s: id_charsH= ' \%([A-Za-z_\UA2-\UA6\UA9\UAA\UAE\UB0\UB5\UBA]\|[^\U00-\UBF]\)'
76+ " Characters that can appear in an identifier, starting in 2nd position. Above
77+ " \UBF we check for operators since we need to stop the identifier if one
78+ " appears. We don't check for invalid characters though.
79+ let s: id_charsW= ' \%([0-9A-Za-z_!\UA2-\UA6\UA9\UAA\UAE-\UB0\UB2-\UB5\UB8-\UBA\UBC-\UBE]\|[^\U00-\UBF]\@=[^' s: op_chars_wc' ]\)'
80+
81+ " A valid julia identifier, more or less
82+ let s: idregex= ' \%(' s: id_charsHs: id_charsW' *\)'
83+
84+
85+
7186syn case match
7287
7388syntax cluster juliaExpressions contains =@juliaParItems,@juliaStringItems,@juliaKeywordItems,@juliaBlocksItems,@juliaTypesItems,@juliaConstItems,@juliaMacroItems,@juliaSymbolItems,@juliaOperatorItems,@juliaNumberItems,@juliaCommentItems,@juliaErrorItems,@juliaSyntaxRegions
@@ -208,7 +223,7 @@ syntax match juliaConstNum display "\%(\<\%(\%(NaN\|Inf\)\%(16\|32\|64\)\?\|p
208223" (This also tries to detect preceding number constants; it does so in a crude
209224" way.)
210225syntax match juliaPossibleEuler " ℯ" contains =juliaEuler
211- exec ' syntax match juliaEuler contained "\%(\%(^\|[' s: nonidS_chars' ]\| ' . s: operators ' \)\%([ .0-9eEf_]*\d\)\?\)\@' s: d80 ).' <=ℯ\ze\%($\| [' s: nonidS_chars' ]\| ' . s: operators ' \) "'
226+ exec ' syntax match juliaEuler contained "\%(\%(^\|[' s: nonidS_charss: op_chars_wc ' ] \)\%(.\?[0-9][ .0-9eEf_]*\d\)\?\)\@' s: d80 ).' <=ℯ\ze[' s: nonidS_charss: op_chars_wc ' ] "'
212227syntax match juliaConstBool display " \<\% (true\| false\)\> "
213228syntax match juliaConstEnv display " \<\% (ARGS\| ENV\| ENDIAN_BOM\| LOAD_PATH\| VERSION\| PROGRAM_FILE\| DEPOT_PATH\)\> "
214229syntax match juliaConstIO display " \<\% (std\% (out\| in\| err\)\| devnull\)\> "
@@ -266,10 +281,12 @@ exec 'syntax match juliaFloat contained "' . s:float_regex . '" contains=juli
266281syntax match juliaComplexUnit display contained " \< im\> "
267282
268283syntax match juliaRangeOperator display " :"
269- exec ' syntax match juliaOperator "' s: operators' "'
284+ exec ' syntax match juliaOperator display "' s: operators1' "'
285+ exec ' syntax match juliaOperator "' s: operatorsd' "'
286+
270287exec ' syntax region juliaTernaryRegion matchgroup=juliaTernaryOperator start="\s\zs?\ze\s" skip="\%(:\(:\|[^:[:space:]' " '" ' "({[]\+\s*\ze:\)\|\%(?\s*\)\@' s: d6 ).' <=:(\)" end=":" contains=@juliaExpressions,juliaErrorSemicol'
271288
272- let s: interp_dollar= ' \([' s: nonidS_charss: uniop_chars . s: binop_chars ' !]\|^\)\@' s: d1 ).' <=\$'
289+ let s: interp_dollar= ' \([' s: nonidS_charss: op_chars_wc ' !]\|^\)\@' s: d1 ).' <=\$'
273290
274291exec ' syntax match juliaDollarVar display contained "' s: interp_dollars: idregex' "'
275292exec ' syntax region juliaDollarPar matchgroup=juliaDollarVar contained start="' s: interp_dollar' (" end=")" contains=@juliaExpressions'
@@ -281,7 +298,7 @@ syntax match juliaChar display "'\\x\x\{2\}'" contains=juliaHexEscapeChar
281298syntax match juliaChar display " '\\ u\x\{ 1,4\} '" contains =juliaUniCharSmall
282299syntax match juliaChar display " '\\ U\x\{ 1,8\} '" contains =juliaUniCharLarge
283300
284- exec ' syntax match juliaCTransOperator "[[:space:]}' s: nonid_charss: uniop_chars . s: binop_chars ' !]\@' s: d1 ).' <!\.\?' " '" ' ᵀ\?"'
301+ exec ' syntax match juliaCTransOperator "[[:space:]}' s: nonid_charss: op_chars_wc ' !]\@' s: d1 ).' <!\.\?' " '" ' ᵀ\?"'
285302
286303" TODO: some of these might be specialized; the rest could be just left to the
287304" generic juliaStringPrefixed fallback
@@ -332,8 +349,8 @@ syntax match juliaPrintfFmt display contained "\\%%"hs=s+1
332349" (for performance reasons)
333350syntax match juliaPossibleSymbol transparent " :\z e[^:]" contains =juliaSymbol,juliaQuotedParBlock,juliaQuotedQMarkPar,juliaColon
334351
335- let s: quotable= ' \%(' s: idregex' \|?\| ' s: operators' \|' s: float_regex' \|' s: int_regex' \)'
336- let s: quoting_colon= ' \%(\%(^\s*\|\s\{6,\}\|[' s: nonid_charss: uniop_chars . s: binop_chars ' ]\s*\)\@' s: d6 ).' <=\|\%(\<\%(return\|if\|else\%(if\)\?\|while\|try\|begin\)\s\+\)\@' s: d9 ).' <=\)\zs:'
352+ let s: quotable= ' \%(' s: idregex' \|' s: operators' \|[?.] \|' s: float_regex' \|' s: int_regex' \)'
353+ let s: quoting_colon= ' \%(\%(^\s*\|\s\{6,\}\|[' s: nonid_charss: op_chars_wc ' ]\s*\)\@' s: d6 ).' <=\|\%(\<\%(return\|if\|else\%(if\)\?\|while\|try\|begin\)\s\+\)\@' s: d9 ).' <=\)\zs:'
337354let s: quoting_colonS= ' \s\@' s: d1 ).' <=:'
338355
339356" note: juliaSymbolS only works within whitespace-sensitive contexts,
0 commit comments