|
| 1 | +/** |
| 2 | + * Provides JavaScript-specific definitions for use in the NfaUtils module. |
| 3 | + */ |
| 4 | + |
| 5 | +private import codeql.regex.nfa.NfaUtils as NfaUtils |
| 6 | +private import codeql.regex.RegexTreeView |
| 7 | + |
| 8 | +/** An implementation that parses a regular expression into a tree of `RegExpTerm`s. */ |
| 9 | +module RegexTreeView implements RegexTreeViewSig { |
| 10 | + import javascript |
| 11 | + |
| 12 | + class Top = Locatable; |
| 13 | + |
| 14 | + /** |
| 15 | + * Holds if `term` is an escape class representing e.g. `\d`. |
| 16 | + * `clazz` is which character class it represents, e.g. "d" for `\d`. |
| 17 | + */ |
| 18 | + predicate isEscapeClass(RegExpTerm term, string clazz) { |
| 19 | + exists(RegExpCharacterClassEscape escape | term = escape | escape.getValue() = clazz) |
| 20 | + } |
| 21 | + |
| 22 | + /** |
| 23 | + * Holds if `term` is a possessive quantifier. |
| 24 | + * As javascript's regexes do not support possessive quantifiers, this never holds, but is used by the shared library. |
| 25 | + */ |
| 26 | + predicate isPossessive(RegExpQuantifier term) { none() } |
| 27 | + |
| 28 | + /** |
| 29 | + * Holds if the regex that `term` is part of is used in a way that ignores any leading prefix of the input it's matched against. |
| 30 | + * Not yet implemented for JavaScript. |
| 31 | + */ |
| 32 | + predicate matchesAnyPrefix(RegExpTerm term) { any() } |
| 33 | + |
| 34 | + /** |
| 35 | + * Holds if the regex that `term` is part of is used in a way that ignores any trailing suffix of the input it's matched against. |
| 36 | + * Not yet implemented for JavaScript. |
| 37 | + */ |
| 38 | + predicate matchesAnySuffix(RegExpTerm term) { any() } |
| 39 | + |
| 40 | + /** |
| 41 | + * Holds if the regular expression should not be considered. |
| 42 | + * |
| 43 | + * For javascript we make the pragmatic performance optimization to ignore minified files. |
| 44 | + */ |
| 45 | + predicate isExcluded(RegExpParent parent) { parent.(Expr).getTopLevel().isMinified() } |
| 46 | + |
| 47 | + /** |
| 48 | + * Holds if `root` has the `i` flag for case-insensitive matching. |
| 49 | + */ |
| 50 | + predicate isIgnoreCase(RegExpTerm root) { RegExp::isIgnoreCase(getFlags(root)) } |
| 51 | + |
| 52 | + /** |
| 53 | + * Gets the flags for `root`, or the empty string if `root` has no flags. |
| 54 | + */ |
| 55 | + private string getFlags(RegExpTerm root) { |
| 56 | + root.isRootTerm() and |
| 57 | + exists(DataFlow::RegExpCreationNode node | node.getRoot() = root | |
| 58 | + result = node.getFlags() |
| 59 | + or |
| 60 | + not exists(node.getFlags()) and |
| 61 | + result = "" |
| 62 | + ) |
| 63 | + or |
| 64 | + exists(RegExpPatternSource source | source.getRegExpTerm() = root | |
| 65 | + result = source.getARegExpObject().(DataFlow::RegExpCreationNode).getFlags() |
| 66 | + ) |
| 67 | + } |
| 68 | + |
| 69 | + /** |
| 70 | + * Holds if `root` has the `s` flag for multi-line matching. |
| 71 | + */ |
| 72 | + predicate isDotAll(RegExpTerm root) { RegExp::isDotAll(getFlags(root)) } |
| 73 | +} |
0 commit comments