Function List not working in v7.2
-
I have a functionList.xml file, it doesn’t work. The function list is empty for JavaScript files. Earlier versions worked.
-
This is a clean install of Notepad++ 7.2.2 32-bit on Windows 10.
-
Just try to modify you functionList.xml file this way.
add :
<association langID="19" id="js_function"/>and add the parser
<parser id="js_function" displayName="Javascript" commentExpr="(?s:/\*.*?\*/)|(?m-s://.*?$)"> <function mainExpr="((^|\s+|[;\}\.])([A-Za-z_]\w*\.)*[A-Za-z_]\w*\s*[=:]|^|[\s;\}]+)\s*function(\s+[A-Za-z_]?\w*\([^\)\(]*\)|\([^\)\(]*\))[\n\s]*\{" > <functionName> <nameExpr expr="[A-Za-z_]\w*\s*[=:]|[A-Za-z_]?\w*\s*\(" /> <nameExpr expr="[A-Za-z_]?\w*" /> </functionName> <className> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*\." /> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*" /> </className> </function> </parser>It’s works don’t ask me why !
-
Notepad++ should assign the “L_JAVASCRIPT” language ID (= 58) to JavaScript files. It appears to assign the “old” language ID “L_JS” (= 19).
@donho is this a regression or am I missing something?
-
my function.xml has even a note
Don't use L_JS (19) use L_JAVASCRIPT (58) instead!not to use 19 and as I’m 100% sure that I never touched javascript configuration,
because I don’t like java at all (I don’t want to start flame war - java has its validity)
I have to assume that this is the standard configuration.Cheers
Claudia -
19 or 58, nothing works. Where can I find a corrected functionList.xml?
-
There are two different files named functionList.xml. One is in Program Files (x86)\Notepad++. The other is in User\XXX\AppData\Roaming\Notepad++.
The latter one is much bigger and more comprehensive, and by default, uses 58 for the association ID. This file also has the apparently incorrect statement “Don’t use L_JS (19) use L_JAVASCRIPT (58) instead!”
When I change the association to 19, the function list works!
-
depending on your configuration whether to use %APPDATA% configuration or not the respective
functionlist gets loaded. I’m still confused that your saying that after changing the id it is working for you.
Could it be that you use an embedded js version as in html?Cheers
Claudia -
No, I am using straight-up JS files, Claudia.
-
ok - I can confirm. After complete deinstall and a wiping of all files and folders in
my installation directory and under %APPDATA% (which I thought I never used)
I reinstalled npp7.2.2. 32bit. Now I have the same behavior. The reason why it was working before is that I run into the doLocalConf bug.
I can’t say which version started ignoring the file but it seems 7.2.2 is doing it.
So @MAPJe71 was completely correct and I have to apologize.Cheers
Claudia -
@MAPJe71 said:
Notepad++ should assign the “L_JAVASCRIPT” language ID (= 58) to JavaScript files. It appears to assign the “old” language ID “L_JS” (= 19).
@donho is this a regression or am I missing something?
– Ver 7.4.1 32bits. for me the problem is still here. so I patched again my functionList.xml
-
Hi @all,
I also found a bug in the regular expression of the JavaScript parser for matching the pattern of a function header. If between the function name and the left parenthesis of the parameter list have been inserted one or more blanks the parser failed to find the function header.
Npp version
v7.4.1, 32 bit
File
C:\Program Files (x86)\Notepad++\functionList.xml
copied to
C:\Users<UserName>\AppData\Roaming\Notepad++\functionList.xmlCorrected JavaScript parser
<parser displayName="JavaScript" id ="javascript_function" commentExpr="(?s:/\*.*?\*/)|(?m-s://.*?$)" > <function mainExpr="((^|\s+|[;\}\.])([A-Za-z_]\w*\.)*[A-Za-z_]\w*\s*[=:]|^|[\s;\}]+)\s*function(\s+[A-Za-z_]?\w*\s*\([^\)\(]*\)|\([^\)\(]*\))[\n\s]*\{" > <functionName> <nameExpr expr="[A-Za-z_]\w*\s*[=:]|[A-Za-z_]?\w*\s*\(" /> <nameExpr expr="[A-Za-z_]?\w*" /> </functionName> <className> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*\." /> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*" /> </className> </function> </parser>The bug was located in the value of
mainExpr. -
I improved the parser for Inno Setup scripts too. Now it is also able to find procedures as well as functions/procedures with no or an empty parameter list. Furthermore forward declarations are no longer matched.
<!-- ================================================== [ Inno Setup ] --> <parser displayName="Inno Setup" id ="innosetup_syntax" > <classRange mainExpr ="(?x) # Utilize inline comments (see `RegEx - Pattern Modifiers`) (?ms) (?'SECTION_HEADER' ^ # header starts at beginning of a line \[ # start of section header (?-i:Code) # `Code` section name ] # end of section header ) .*? # whatever, up till... (?= # ...next valid section header or... ^ # +-- header starts at beginning of a line \[ # +-- start-of-header indicator (?-i: Components|(?:Custom)?Messages | Dirs | Files | I(?:cons|nstallDelete) | Languages | R(?:egistry|un) | Setup | T(?:asks|ypes) | Uninstall(?:Delete|Run) ) # +-- valid section name ] # \-- end-of-header indicator | \Z # ...end-of-file ) " > <className> <nameExpr expr="^\[\K[^\h\]]+(?=])" /> </className> <function mainExpr="(?x) # Utilize inline comments (see `RegEx - Pattern Modifiers`) (?m-s) # multi-line mode on, single-line mode off ^\h* # optional leading whitespace (?i: ( (?'FUNCTION_HEADER' # function FUNCTION ) | # or (?'PROCEDURE_HEADER' # procedure PROCEDURE ) )\h+ ) (?'VALID_ID' # function/procedure identifier [A-Za-z_]\w* ) ( # optional parameter list \s*\( # start-of-parameter-list indicator [^()]* # parameter list \) # end-of-parameter-list indicator )* (?('FUNCTION_HEADER') # functions have a return type ( \s*: # type indicator \s*[A-Za-z_]\w* # type identifier \s*; # end-of-statement indicator ) ) (?('PROCEDURE_HEADER') # procedures don't have a return type ( \s*; # end-of-statement indicator ) ) (?!(?i:\s*FORWARD\s*;)) # prevent matching forward declarations " > <functionName> <funcNameExpr expr="(?i:(FUNCTION|PROCEDURE)\h+)\K[A-Za-z_]\w*\s*(\([^()]*\))*" /> <!-- comment out the following node to display the method with its parameters --> <funcNameExpr expr="[A-Za-z_]\w*" /> </functionName> </function> </classRange> <function mainExpr="(?x) # Utilize inline comments (see `RegEx - Pattern Modifiers`) (?ms) (?'SECTION_HEADER' ^ # header starts at beginning of a line \[ # start-of-header indicator (?-i: Components|(?:Custom)?Messages | Dirs | Files | I(?:cons|nstallDelete) | Languages | R(?:egistry|un) | Setup | T(?:asks|ypes) | Uninstall(?:Delete|Run) ) # valid section name ] # end-of-header indicator ) .*? # whatever, up till... (?= (?&SECTION_HEADER) # ...next valid section header,... | ^\[(?-i:Code)] # ...`Code` section header or... | \Z # ...end-of-file ) " > <functionName> <nameExpr expr="^\[\K[^\]]+(?=])" /> </functionName> </function> </parser> -
doesn’t work anymore for 7.4.2, why is this even an issue? this is the most basic ui test possible, very disappointing. But looking at the “new” xml structure, no wonder people have problems navigating to this custom xml interpretation.
Switching ID to 19, even though a random comment says not to, fixed this for me… -
`
<parser
displayName=“JavaScript”
id =“javascript_function”
commentExpr=“(?s:/.?*/)|(?m-s://.*?$)”bemerkung1="für Kommentar zwischen function header und {. [\n\s]*\{ am Ende gelöscht." bemerkung2="für new Function(...) (?:new\s*)? eingefügt." bemerkung3="für function (a,b,c) ohne Funktionsname ein \s* eingefügt." comment1="comment between function header and function body {, removed [\n\s]*\{" comment2="new Function() : inserted (?:new\s*)? " comment3="function (a,b,c) without function name: inserted \s*" comment4="comment within function header still not recognized. e.g. function /*returns real*/ a(b) { }" comment99="there must be something wrong with processing of commentExpr" > <function mainExpr="((^|\s+|[;\}\.])([A-Za-z_]\w*\.)*[A-Za-z_]\w*\s*[=:]|^|[\s;\}]+)\s*(?:new\s*)?function(\s+[A-Za-z_]?\w*\s*\([^\)\(]*\)|s*\([^\)\(]*\))" > <functionName> <nameExpr expr="[A-Za-z_]\w*\s*[=:]|[A-Za-z_]?\w*\s*\(" /> <nameExpr expr="[A-Za-z_]?\w*" /> </functionName> <className> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*\." /> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*" /> </className> </function> </parser>` -
trying again, by content and by form
<parser displayName="JavaScript" id ="javascript_function" XcommentExpr="(?s:/\*.*?\*/)|(?m-s://.*?$)" bemerkung1="für Kommentar zwischen function header und {. [\n\s]*\{ am Ende gelöscht." bemerkung2="für new Function(...) (?:new\s*)? eingefügt." bemerkung3="für function (a,b,c) ohne Funktionsname ein \s* eingefügt." bemerkung4="Kommentar im Funktionsheader wird nicht behandelt, z.B. function /*returns real*/ a(b) { } " bemerkung5="commentExpr überhaupt X-ed." bemerkung99="Es muss etwas falsch sein mit commentExpr durch nichts ersetzen." comment1="Comment between function header and function body {: removed [\n\s]*\{ " comment2="new Function(): inserted (?:new\s*)? " comment3="function (a,b,c) without function name: inserted \s*" comment4="Comment within function header still not handled. e.g. function /*returns real*/ a(b) { } " comment5="commentExpr disabled: X-ed." comment99="There must be something wrong in NPP when replacing commentExpr by nothing." > <function mainExpr="((^|\s+|[;\}\.])([A-Za-z_]\w*\.)*[A-Za-z_]\w*\s*[=:]|^|[\s;\}]+)\s*(?:new\s*)?function(\s+[A-Za-z_]?\w*\s*\([^\)\(]*\)|s*\([^\)\(]*\))" > <functionName> <nameExpr expr="[A-Za-z_]\w*\s*[=:]|[A-Za-z_]?\w*\s*\(" /> <nameExpr expr="[A-Za-z_]?\w*" /> </functionName> <className> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*\." /> <nameExpr expr="([A-Za-z_]\w*\.)*[A-Za-z_]\w*" /> </className> </function> </parser> -
Notepad++ uses the
commentExprto determine which parts of the text to skip when searching for classes, methods and/or functions.
“Embedded” comments are not supported. -
@t12122013-Fink
See also issues #344 and #1023 on GitHub. -
C# function list fix for v7.4.2
Basically copy the function section inside classRange and paste it at the same level of classRange.
<!-- ========================================================== [ C# ] --><!-- | Based on: | http://sourceforge.net/p/notepad-plus/patches/613/ \--> <parser displayName="C#" id ="csharp_class" commentExpr="(?s:/\*.*?\*/)|(?m-s://.*?$)" > <classRange mainExpr ="^[\t\x20]*((public|protected|private|internal)\s+)?(\w+\s*)?(class|struct|interface)[\t\x20]+[^\{]+\{" openSymbole ="\{" closeSymbole="\}" > <className> <nameExpr expr="(class|struct|interface)[\t\x20]+\w+" /> <nameExpr expr="[\t\x20]+\w+" /> <nameExpr expr="\w+" /> </className> <function mainExpr="^[^\S\r\n]*(?<modifier1>(?:public|protected|internal|private)\s*)?(?<modifier2>(?:new|static|virtual|sealed|override|abstract|extern)\s*)?(partial\s*)?(?<type>(?!(return|if|else))\w+(?<genericType><[\w,\s<>]+>)?\s+)(?<name>\w+(?<genericNameType><[\w,\s<>]+>)?\s?)\((?<params>[\w\s,<>\[\]\:=\.]*)\)(?<ctorChain>\s*\:\s*(?:base|this)\s*\((?<ctorParams>[\w\s,<>\[\]\:=\.]*)\))?[\w\s<>\:,\(\)\[\]]*(?:\{|;)" > <functionName> <funcNameExpr expr="(\w+(<[\w,\s<>]+>)?\s?)\(" /> <funcNameExpr expr="(\w+(<[\w,\s<>]+>)?\s?)" /> </functionName> </function> </classRange> <function mainExpr="^[^\S\r\n]*(?<modifier1>(?:public|protected|internal|private)\s*)?(?<modifier2>(?:new|static|virtual|sealed|override|abstract|extern)\s*)?(partial\s*)?(?<type>(?!(return|if|else))\w+(?<genericType><[\w,\s<>]+>)?\s+)(?<name>\w+(?<genericNameType><[\w,\s<>]+>)?\s?)\((?<params>[\w\s,<>\[\]\:=\.]*)\)(?<ctorChain>\s*\:\s*(?:base|this)\s*\((?<ctorParams>[\w\s,<>\[\]\:=\.]*)\))?[\w\s<>\:,\(\)\[\]]*(?:\{|;)" > <functionName> <funcNameExpr expr="(\w+(<[\w,\s<>]+>)?\s?)\(" /> <funcNameExpr expr="(\w+(<[\w,\s<>]+>)?\s?)" /> </functionName> </function> </parser> -
@ MAPJe71
Thank you for the explanation about comment handling. Of course I assumed Notepad++ would strip comments lexically, replacing them by an empty string, and only then apply mainExpr, but it seems not to be the case.For my situation and needs I prefer to disable commentExpr at all, and do the other minor improvements written in my comments. I do get all functions, and do not get false functions. Reasoning, I do not have functions inside comments, and I do not place comments in function headers except between ) and { . Besides that, functions temporarely commented out, may still count as functions.
At least the <parser> above is a partial proposal to my and other’s issues.By the way do you have an idea how to remove the mis-written earlier posts. I was not fluent with this BB and with markup language, and is there a preview?
