Use Boost's regex implementation instead of libstdc++'s

The latter runs in unbounded stack space so it crashes on large
inputs.

Fixes #4550.

doc/manual/generate-manpage.nix

@@ -7,7 +7,10 @@ let
 
   showCommand =
     { command, def, filename }:
-    "# Name\n\n"
+    ''
+      **Warning**: This program is **experimental** and its interface is subject to change.
+    ''
+    + "# Name\n\n"
     + "`${command}` - ${def.description}\n\n"
     + "# Synopsis\n\n"
     + showSynopsis { inherit command; args = def.args; }

doc/manual/src/command-ref/nix-shell.md

@@ -232,22 +232,23 @@ terraform apply
 > in a nix-shell shebang.
 
 Finally, using the merging of multiple nix-shell shebangs the following
-Haskell script uses a specific branch of Nixpkgs/NixOS (the 18.03 stable
+Haskell script uses a specific branch of Nixpkgs/NixOS (the 20.03 stable
 branch):
 
 ```haskell
 #! /usr/bin/env nix-shell
-#! nix-shell -i runghc -p "haskellPackages.ghcWithPackages (ps: [ps.HTTP ps.tagsoup])"
-#! nix-shell -I nixpkgs=https://github.com/NixOS/nixpkgs/archive/nixos-18.03.tar.gz
+#! nix-shell -i runghc -p "haskellPackages.ghcWithPackages (ps: [ps.download-curl ps.tagsoup])"
+#! nix-shell -I nixpkgs=https://github.com/NixOS/nixpkgs/archive/nixos-20.03.tar.gz
 
-import Network.HTTP
+import Network.Curl.Download
 import Text.HTML.TagSoup
+import Data.Either
+import Data.ByteString.Char8 (unpack)
 
 -- Fetch nixos.org and print all hrefs.
 main = do
-  resp <- Network.HTTP.simpleHTTP (getRequest "http://nixos.org/")
-  body <- getResponseBody resp
-  let tags = filter (isTagOpenName "a") $ parseTags body
+  resp <- openURI "https://nixos.org/"
+  let tags = filter (isTagOpenName "a") $ parseTags $ unpack $ fromRight undefined resp
   let tags' = map (fromAttrib "href") tags
   mapM_ putStrLn $ filter (/= "") tags'
 ```

flake.nix

@@ -135,7 +135,7 @@
               # Copy libboost_context so we don't get all of Boost in our closure.
               # https://github.com/NixOS/nixpkgs/issues/45462
               mkdir -p $out/lib
-              cp -pd ${boost}/lib/{libboost_context*,libboost_thread*,libboost_system*} $out/lib
+              cp -pd ${boost}/lib/{libboost_context*,libboost_thread*,libboost_system*,libboost_regex*} $out/lib
               rm -f $out/lib/*.a
               ${lib.optionalString stdenv.isLinux ''
                 chmod u+w $out/lib/*.so.*

src/libutil/regex.hh

@@ -0,0 +1,6 @@
+#pragma once
+
+#include <boost/regex.hpp>
+
+namespace regex = boost;
+

src/nix/develop.cc

@@ -6,8 +6,7 @@
 #include "derivations.hh"
 #include "affinity.hh"
 #include "progress-bar.hh"
-
-#include <regex>
+#include "regex.hh"
 
 using namespace nix;
 
@@ -64,42 +63,42 @@ BuildEnvironment readEnvironment(const Path & path)
     static std::string indexedArrayRegex =
         R"re((?:\(( *\[[0-9]+\]="(?:[^"\\]|\\.)*")*\)))re";
 
-    static std::regex declareRegex(
+    static regex::regex declareRegex(
         "^declare -a?x (" + varNameRegex + ")(=(" +
         dquotedStringRegex + "|" + indexedArrayRegex + "))?\n");
 
-    static std::regex varRegex(
+    static regex::regex varRegex(
         "^(" + varNameRegex + ")=(" + simpleStringRegex + "|" + squotedStringRegex + "|" + indexedArrayRegex + ")\n");
 
     /* Note: we distinguish between an indexed and associative array
        using the space before the closing parenthesis. Will
        undoubtedly regret this some day. */
-    static std::regex assocArrayRegex(
+    static regex::regex assocArrayRegex(
         "^(" + varNameRegex + ")=" + R"re((?:\(( *\[[^\]]+\]="(?:[^"\\]|\\.)*")* *\)))re" + "\n");
 
-    static std::regex functionRegex(
+    static regex::regex functionRegex(
         "^" + varNameRegex + " \\(\\) *\n");
 
     while (pos != file.end()) {
 
-        std::smatch match;
+        regex::smatch match;
 
-        if (std::regex_search(pos, file.cend(), match, declareRegex, std::regex_constants::match_continuous)) {
+        if (regex::regex_search(pos, file.cend(), match, declareRegex, regex::regex_constants::match_continuous)) {
             pos = match[0].second;
             exported.insert(match[1]);
         }
 
-        else if (std::regex_search(pos, file.cend(), match, varRegex, std::regex_constants::match_continuous)) {
+        else if (regex::regex_search(pos, file.cend(), match, varRegex, regex::regex_constants::match_continuous)) {
             pos = match[0].second;
             res.env.insert({match[1], Var { .exported = exported.count(match[1]) > 0, .quoted = match[2] }});
         }
 
-        else if (std::regex_search(pos, file.cend(), match, assocArrayRegex, std::regex_constants::match_continuous)) {
+        else if (regex::regex_search(pos, file.cend(), match, assocArrayRegex, regex::regex_constants::match_continuous)) {
             pos = match[0].second;
             res.env.insert({match[1], Var { .associative = true, .quoted = match[2] }});
         }
 
-        else if (std::regex_search(pos, file.cend(), match, functionRegex, std::regex_constants::match_continuous)) {
+        else if (regex::regex_search(pos, file.cend(), match, functionRegex, regex::regex_constants::match_continuous)) {
             res.bashFunctions = std::string(pos, file.cend());
             break;
         }

src/nix/local.mk

@@ -18,7 +18,7 @@ nix_CXXFLAGS += -I src/libutil -I src/libstore -I src/libfetchers -I src/libexpr
 
 nix_LIBS = libexpr libmain libfetchers libstore libutil libcmd
 
-nix_LDFLAGS = -pthread $(SODIUM_LIBS) $(EDITLINE_LIBS) $(BOOST_LDFLAGS) -llowdown
+nix_LDFLAGS = -pthread $(SODIUM_LIBS) $(EDITLINE_LIBS) $(BOOST_LDFLAGS) -llowdown -lboost_regex
 
 $(foreach name, \
   nix-build nix-channel nix-collect-garbage nix-copy-closure nix-daemon nix-env nix-hash nix-instantiate nix-prefetch-url nix-shell nix-store, \