Add a primop for regular expression pattern matching

The function ‘builtins.match’ takes a POSIX extended regular
expression and an arbitrary string. It returns ‘null’ if the string
does not match the regular expression. Otherwise, it returns a list
containing substring matches corresponding to parenthesis groups in
the regex. The regex must match the entire string (i.e. there is an
implied "^<pat>$" around the regex).  For example:

  match "foo" "foobar" => null
  match "foo" "foo" => []
  match "f(o+)(.*)" "foooobar" => ["oooo" "bar"]
  match "(.*/)?([^/]*)" "/dir/file.nix" => ["/dir/" "file.nix"]
  match "(.*/)?([^/]*)" "file.nix" => [null "file.nix"]

The following example finds all regular files with extension .nix or
.patch underneath the current directory:

  let

    findFiles = pat: dir: concatLists (mapAttrsToList (name: type:
      if type == "directory" then
        findFiles pat (dir + "/" + name)
      else if type == "regular" && match pat name != null then
        [(dir + "/" + name)]
      else []) (readDir dir));

  in findFiles ".*\\.(nix|patch)" (toString ./.)

5 files changed, 84 insertions, 5 deletions

diff --git a/src/libexpr/primops.cc b/src/libexpr/primops.cc
index ed50c8091..b0596dad9 100644
--- a/src/libexpr/primops.cc
+++ b/src/libexpr/primops.cc
@@ -1430,7 +1430,34 @@ static void prim_hashString(EvalState & state, const Pos & pos, Value * * args,
     string s = state.forceString(*args[1], context, pos);
 
     mkString(v, printHash(hashString(ht, s)), context);
-};
+}
+
+
+/* Match a regular expression against a string and return either
+   ‘null’ or a list containing substring matches. */
+static void prim_match(EvalState & state, const Pos & pos, Value * * args, Value & v)
+{
+    Regex regex(state.forceStringNoCtx(*args[0], pos), true);
+
+    PathSet context;
+    string s = state.forceString(*args[1], context, pos);
+
+    Regex::Subs subs;
+    if (!regex.matches(s, subs)) {
+        mkNull(v);
+        return;
+    }
+
+    unsigned int len = subs.empty() ? 0 : subs.rbegin()->first + 1;
+    state.mkList(v, len);
+    for (unsigned int n = 0; n < len; ++n) {
+        auto i = subs.find(n);
+        if (i == subs.end())
+            mkNull(*(v.list.elems[n] = state.allocValue()));
+        else
+            mkString(*(v.list.elems[n] = state.allocValue()), i->second);
+    }
+}
 
 
 /*************************************************************
@@ -1584,6 +1611,7 @@ void EvalState::createBaseEnv()
     addPrimOp("__unsafeDiscardStringContext", 1, prim_unsafeDiscardStringContext);
     addPrimOp("__unsafeDiscardOutputDependency", 1, prim_unsafeDiscardOutputDependency);
     addPrimOp("__hashString", 2, prim_hashString);
+    addPrimOp("__match", 2, prim_match);
 
     // Versions
     addPrimOp("__parseDrvName", 1, prim_parseDrvName);
diff --git a/src/libutil/regex.cc b/src/libutil/regex.cc
index 36c8458ce..84274b3e1 100644
--- a/src/libutil/regex.cc
+++ b/src/libutil/regex.cc
@@ -1,13 +1,16 @@
 #include "regex.hh"
 #include "types.hh"
 
+#include <algorithm>
+
 namespace nix {
 
-Regex::Regex(const string & pattern)
+Regex::Regex(const string & pattern, bool subs)
 {
     /* Patterns must match the entire string. */
-    int err = regcomp(&preg, ("^(" + pattern + ")$").c_str(), REG_NOSUB | REG_EXTENDED);
-    if (err) throw Error(format("compiling pattern ‘%1%’: %2%") % pattern % showError(err));
+    int err = regcomp(&preg, ("^(" + pattern + ")$").c_str(), (subs ? 0 : REG_NOSUB) | REG_EXTENDED);
+    if (err) throw RegexError(format("compiling pattern ‘%1%’: %2%") % pattern % showError(err));
+    nrParens = subs ? std::count(pattern.begin(), pattern.end(), '(') : 0;
 }
 
 Regex::~Regex()
@@ -23,6 +26,20 @@ bool Regex::matches(const string & s)
     throw Error(format("matching string ‘%1%’: %2%") % s % showError(err));
 }
 
+bool Regex::matches(const string & s, Subs & subs)
+{
+    regmatch_t pmatch[nrParens + 2];
+    int err = regexec(&preg, s.c_str(), nrParens + 2, pmatch, 0);
+    if (err == 0) {
+        for (unsigned int n = 2; n < nrParens + 2; ++n)
+            if (pmatch[n].rm_eo != -1)
+                subs[n - 2] = string(s, pmatch[n].rm_so, pmatch[n].rm_eo - pmatch[n].rm_so);
+        return true;
+    }
+    else if (err == REG_NOMATCH) return false;
+    throw Error(format("matching string ‘%1%’: %2%") % s % showError(err));
+}
+
 string Regex::showError(int err)
 {
     char buf[256];
diff --git a/src/libutil/regex.hh b/src/libutil/regex.hh
index aa012b721..53e31f4ed 100644
--- a/src/libutil/regex.hh
+++ b/src/libutil/regex.hh
@@ -5,16 +5,23 @@
 #include <sys/types.h>
 #include <regex.h>
 
+#include <map>
+
 namespace nix {
 
+MakeError(RegexError, Error)
+
 class Regex
 {
 public:
-    Regex(const string & pattern);
+    Regex(const string & pattern, bool subs = false);
     ~Regex();
     bool matches(const string & s);
+    typedef std::map<unsigned int, string> Subs;
+    bool matches(const string & s, Subs & subs);
 
 private:
+    unsigned nrParens;
     regex_t preg;
     string showError(int err);
 };
diff --git a/tests/lang/eval-okay-regex-match.exp b/tests/lang/eval-okay-regex-match.exp
new file mode 100644
index 000000000..27ba77dda
--- /dev/null
+++ b/tests/lang/eval-okay-regex-match.exp
@@ -0,0 +1 @@
+true
diff --git a/tests/lang/eval-okay-regex-match.nix b/tests/lang/eval-okay-regex-match.nix
new file mode 100644
index 000000000..ae6501532
--- /dev/null
+++ b/tests/lang/eval-okay-regex-match.nix
@@ -0,0 +1,26 @@
+with builtins;
+
+let
+
+  matches = pat: s: match pat s != null;
+
+  splitFN = match "((.*)/)?([^/]*)\\.(nix|cc)";
+
+in
+
+assert  matches "foobar" "foobar";
+assert  matches "fo*" "f";
+assert !matches "fo+" "f";
+assert  matches "fo*" "fo";
+assert  matches "fo*" "foo";
+assert  matches "fo+" "foo";
+assert  matches "fo{1,2}" "foo";
+assert !matches "fo{1,2}" "fooo";
+assert !matches "fo*" "foobar";
+
+assert match "(.*)\\.nix" "foobar.nix" == [ "foobar" ];
+
+assert splitFN "/path/to/foobar.nix" == [ "/path/to/" "/path/to" "foobar" "nix" ];
+assert splitFN "foobar.cc" == [ null null "foobar" "cc" ];
+
+true