AjaxMinDll/JavaScript/ResolutionVisitor.cs (9 lines): - line 255: // TODO: we could get rid of the ConstantWrapper restriction and use an Evaluate visitor - line 270: // TODO: what if we are assigning it to a variable? Should we track that variable and see if we call it? - line 399: // TODO: this really should be a LIST of ghosted fields, since multiple - line 487: // TODO: this really should be a LIST of ghosted fields, since multiple - line 783: // TODO: throw a warning if there's nothing defined in this scope, because there - line 1023: // TODO: try resolving the module, and if we can, use its exports to add to our exports. - line 1033: // TODO: add the specifiers to the exports of this module - line 1043: // TODO: add the references or definitions to the export list for the module - line 1499: // TODO: what about initializers INSIDE the bindings? AjaxMinDll/JavaScript/jsparser.cs (4 lines): - line 959: // TODO: error node? - line 3504: // TODO: we want to error if we aren't immediately preceeded by a comma operator, - line 4064: // TODO: not sure if this is the right place to hook for the ES6 YieldExpression semantics! - line 4174: // TODO: figure out how to get a context on just the literal part! AjaxMinDll/JavaScript/AnalyzeNodeVisitor.cs (4 lines): - line 429: // TODO: would be nice to properly remove unused declarations from binding parameters - line 889: // TODO: we *could* also do it is the expr1 assignments are to lookups that are - line 1439: // TODO: if we removed a labeled continue, do we need to fix up some label references? - line 2035: // TODO: verify that there are no naming collisions in the scope AjaxMinDll/JavaScript/declaration.cs (3 lines): - line 130: // TODO: what should we do if we try to add a const to a var, or a var to a const??? - line 161: // TODO: what should we do if we try to add a const to a var, or a var to a const??? - line 178: // TODO: for now, just do this logic for simple binding identifiers. AjaxMinDll/JavaScript/EvaluateLiteralVisitor.cs (2 lines): - line 1998: // TODO: shouldn't we check if they BOTH are binary operators? (a*6)*(5*b) ==> a*30*b (for instance) - line 2466: // TODO: (someday) we'd RATHER eliminate the statement altogether if the condition is always false, AjaxMinDll/JavaScript/activationobject.cs (2 lines): - line 573: // TODO: try to reuse some pre-existing variable, or maybe replace - line 782: // TODO: should we at least throw a warning? AjaxMinDll/Css/CssParser.cs (2 lines): - line 2274: // TODO: error? shouldn't we ALWAYS have a selector list inside a not()/matches()/any() function? - line 3368: // TODO: make sure there is whitespace before the ( if it would cause AjaxMinDll/JavaScript/block.cs (1 line): - line 102: // TODO: if there are no statements -- empty block -- then is is an expression? DllUnitTest/ManifestTask.cs (1 line): - line 104: // TODO: verify map file contents AjaxMinDll/JavaScript/OutputVisitor.cs (1 line): - line 423: // TODO: the plus operator: if we can prove that it is a numeric operator AjaxMinDll/CommonSettings.cs (1 line): - line 384: // TODO: how can I escape any commas? AjaxMinDll/JavaScript/ReorderScopeVisitor.cs (1 line): - line 860: // TODO: or should we leave them to preserve the "error"? AjaxMinDll/JavaScript/jsscanner.cs (1 line): - line 2293: // TODO: we want to flag this string as unterminated *and having issues*,