MISSING_SEMI false positive in if/else

[TheRealAgentK]

<cfscript>
	a = false;
	b = "";

	if (isDefined("c") && d == "registration") {
		b = "e.f";
	}
	else if (structKeyExists(variables,'g') && variables.g) {
		b = "y.j";
		myFusebox.do("k.l");
	}
	else if (session.sf.get('n.m') == "p") {
		b = "l.k";
		session.sf.set('n.m',false);
	}
</cfscript>

Result:

<issue severity="ERROR" id="MISSING_SEMI" message="MISSING_SEMI" category="CFLint" abbrev="MS"><location file="/Users/kai/Documents/Code/....cfm" fileName="ppp.cfm" function="" column="1" line="8" message="End of statement(;) expected instead of else" variable=""><Expression><![CDATA[else]]></Expression></location>
</issue>

<issue severity="ERROR" id="MISSING_SEMI" message="MISSING_SEMI" category="CFLint" abbrev="MS"><location file="/Users/kai/Documents/Code/.....cfm" fileName="ppp.cfm" function="" column="14" line="10" message="End of statement(;) expected instead of (" variable=""><Expression><![CDATA[(]]></Expression></location>
</issue>

The offending lines are:

else if (structKeyExists(variables,'g') && variables.g) {

and

myFusebox.do("k.l");

[TheRealAgentK]

Trying to ignore the MISSING_SEMI here leads to the same issue as reported in #285.

I've tried to put

// cflint ignore:MISSING_SEMI on lines 8 and 10 to the right of the code - doesn't do anything.

Tried to put it above each of the offending/reported lines - doesn't do anything either.

[TheRealAgentK]

Interesting enough, putting

<!--- @CFLintIgnore MISSING_SEMI --->

above the <cfscript> tag doesn't help and it's still not being ignored.

[TheRealAgentK]

Ahh, actually the same file triggers this, too:

<issue severity="ERROR" id="PARSE_ERROR" message="PARSE_ERROR" category="CFLint" abbrev="PE"><location file="/Users/kai/Documents/Code/...cfm" fileName="act_processModalRules.cfm" function="" column="1" line="11" message="Unable to parse" variable=""><Expression><![CDATA[mismatched input '}' expecting WHILE]]></Expression></location>
</issue>

which now makes me think that the core of this whole issue might be that CFLint thinks

myFusebox.do("k.l");

is actually starting a do-while loop?

[ryaneberly]

created this issue in the cfparser project
cfparser/cfparser#76

[ryaneberly]

@TheRealAgentK . Yes, you are correct. Do is a reserved word. I am changing the parser to be aware of the differences between a member function call and an unqualified function call.

You could work around it today if you are able to rename your do() function.

[ryaneberly]

fixed in cfparser 2.4.6

[TheRealAgentK]

Wow, you've been on a roll! :) Will give all of your fixes and changes a good test against my offending code during the weekend.

Does cfparser get pulled into the cflint maven build automatically?

[ryaneberly]

Thanks.

No, to get the parser fixes:
checkout and build cfparser locally, and then change the version of cfparser in the CFLint pom to 2.4.6.

or wait a week on those few, I'll push cfparser out to maven soon.

[TheRealAgentK]

I just built it locally and it seems to run fine with my local cfparser 2.4.6

I do get a message about ANTLR 4.6 vs 4.7 though when running cflint.

...
JSON Output false
HTML Output false
log4j:WARN No appenders could be found for logger (com.cflint.tools.CFLintFilter).
log4j:WARN Please initialize the log4j system properly.
log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info.
ERROR StatusLogger No log4j2 configuration file found. Using default configuration: logging only errors to the console.
ANTLR Tool version 4.6 used for code generation does not match the current runtime version 4.7
...

Have updated all references in the pom to Antlr 4.7 though.

[TheRealAgentK]

The MISSING_SEMI issue is actually working fine now with that myFusebox.do(), too :)