refactor: Replace Boost trim and to_lower with libxrpl helpers (#7995)

include/xrpl/basics/StringUtilities.h

@@ -125,9 +125,31 @@ struct ParsedUrl
 bool
 parseUrl(ParsedUrl& pUrl, std::string const& strUrl);
 
+/**
+ * Remove leading and trailing ASCII whitespace.
+ *
+ * Whitespace is the fixed set " \t\n\v\f\r"; the current locale is not
+ * consulted, so the result depends only on the input.
+ *
+ * @param str The string to trim.
+ * @return @p str without leading or trailing whitespace.
+ */
 std::string
 trimWhitespace(std::string str);
 
+/**
+ * Fold ASCII upper case letters to lower case.
+ *
+ * Only 'A' through 'Z' are remapped; every other byte is left alone and the
+ * current locale is not consulted, so the result depends only on the input.
+ *
+ * @param str The string to fold.
+ * @return @p str with each ASCII upper case letter replaced by its lower case
+ *         equivalent.
+ */
+std::string
+toLower(std::string str);
+
 std::optional<std::uint64_t>
 toUInt64(std::string const& s);
 

src/libxrpl/basics/StringUtilities.cpp

@@ -5,15 +5,15 @@
 #include <xrpl/beast/net/IPEndpoint.h>
 
 #include <boost/algorithm/hex.hpp>
-#include <boost/algorithm/string/case_conv.hpp>
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/regex/v5/regbase.hpp>
 #include <boost/regex/v5/regex.hpp>
 #include <boost/regex/v5/regex_match.hpp>
 
+#include <algorithm>
 #include <cstdint>
 #include <iterator>
 #include <optional>
+#include <ranges>
 #include <string>
 #include <string_view>
 
@@ -67,7 +67,7 @@ parseUrl(ParsedUrl& pUrl, std::string const& strUrl)
     }
 
     pUrl.scheme = smMatch[1];
-    boost::algorithm::to_lower(pUrl.scheme);
+    pUrl.scheme = toLower(pUrl.scheme);
     pUrl.username = smMatch[2];
     pUrl.password = smMatch[3];
     std::string const domain = smMatch[4];
@@ -93,10 +93,42 @@ parseUrl(ParsedUrl& pUrl, std::string const& strUrl)
     return true;
 }
 
+namespace {
+
+// Deliberately not std::isspace / std::tolower: those consult the current C
+// locale, so the same input could trim or fold differently depending on
+// process-wide state set by something else entirely. Everything these helpers
+// are used on (config keys and values, URL schemes, hex digests) is ASCII, and
+// the callers want a fixed answer, so spell the ASCII rules out.
+
+constexpr bool
+isAsciiSpace(char c)
+{
+    return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r';
+}
+
+constexpr char
+toAsciiLower(char c)
+{
+    return (c >= 'A' && c <= 'Z') ? static_cast<char>(c - 'A' + 'a') : c;
+}
+
+}  // namespace
+
 std::string
 trimWhitespace(std::string str)
 {
-    boost::trim(str);
+    auto const end = std::ranges::find_if_not(str | std::views::reverse, isAsciiSpace).base();
+    str.erase(end, str.end());
+    str.erase(str.begin(), std::ranges::find_if_not(str, isAsciiSpace));
+
+    return str;
+}
+
+std::string
+toLower(std::string str)
+{
+    std::ranges::transform(str, str.begin(), toAsciiLower);
     return str;
 }
 

src/libxrpl/crypto/RFC1751.cpp

@@ -1,11 +1,11 @@
 #include <xrpl/crypto/RFC1751.h>
 
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/beast/utility/instrumentation.h>
 
 #include <boost/algorithm/string/classification.hpp>
 #include <boost/algorithm/string/constants.hpp>
 #include <boost/algorithm/string/split.hpp>
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/range/adaptor/copied.hpp>
 
 #include <cctype>
@@ -397,7 +397,7 @@ RFC1751::getKeyFromEnglish(std::string& strKey, std::string const& strHuman)
 
     std::string strTrimmed(strHuman);
 
-    boost::algorithm::trim(strTrimmed);
+    strTrimmed = trimWhitespace(strTrimmed);
 
     boost::algorithm::split(
         vWords, strTrimmed, boost::algorithm::is_space(), boost::algorithm::token_compress_on);

src/libxrpl/server/Manifest.cpp

@@ -23,8 +23,6 @@
 #include <xrpl/rdb/DatabaseCon.h>
 #include <xrpl/server/Wallet.h>
 
-#include <boost/algorithm/string/trim.hpp>
-
 #include <cstddef>
 #include <cstdint>
 #include <exception>
@@ -277,7 +275,7 @@ loadValidatorToken(std::vector<std::string> const& blob, beast::Journal journal)
                 [](std::size_t init, std::string const& s) { return init + s.size(); }));
 
         for (auto const& line : blob)
-            tokenStr += boost::algorithm::trim_copy(line);
+            tokenStr += trimWhitespace(line);
 
         tokenStr = base64Decode(tokenStr);
 
@@ -653,7 +651,7 @@ ManifestCache::load(
                 [](std::size_t init, std::string const& s) { return init + s.size(); }));
 
         for (auto const& line : configRevocation)
-            revocationStr += boost::algorithm::trim_copy(line);
+            revocationStr += trimWhitespace(line);
 
         auto mo = deserializeManifest(base64Decode(revocationStr));
 

src/libxrpl/server/Port.cpp

@@ -1,5 +1,6 @@
 #include <xrpl/server/Port.h>
 
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/contract.h>
 #include <xrpl/basics/safe_cast.h>
 #include <xrpl/beast/core/LexicalCast.h>
@@ -9,7 +10,6 @@
 #include <xrpl/config/Constants.h>
 
 #include <boost/algorithm/string/predicate.hpp>
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/asio/ip/address.hpp>
 #include <boost/asio/ip/impl/network_v4.ipp>
 #include <boost/asio/ip/impl/network_v6.ipp>
@@ -98,7 +98,7 @@ populate(
 
     while (std::getline(ss, ip, ','))
     {
-        boost::algorithm::trim(ip);
+        ip = trimWhitespace(ip);
         bool v4 = false;
         boost::asio::ip::network_v4 v4Net;
         boost::asio::ip::network_v6 v6Net;

src/tests/libxrpl/basics/StringUtilities.cpp

@@ -290,4 +290,44 @@ TEST_F(StringUtilitiesTest, to_string)
     EXPECT_EQ(result, "hello");
 }
 
+TEST_F(StringUtilitiesTest, trimWhitespace)
+{
+    EXPECT_EQ(trimWhitespace(""), "");
+    EXPECT_EQ(trimWhitespace("   "), "");
+    EXPECT_EQ(trimWhitespace("abc"), "abc");
+    EXPECT_EQ(trimWhitespace("  abc"), "abc");
+    EXPECT_EQ(trimWhitespace("abc  "), "abc");
+    EXPECT_EQ(trimWhitespace(" \t\n\v\f\r abc \t\n\v\f\r "), "abc");
+
+    // Interior whitespace is preserved.
+    EXPECT_EQ(trimWhitespace("  a b\tc  "), "a b\tc");
+}
+
+TEST_F(StringUtilitiesTest, toLower)
+{
+    EXPECT_EQ(toLower(""), "");
+    EXPECT_EQ(toLower("ABC"), "abc");
+    EXPECT_EQ(toLower("AbC123"), "abc123");
+    EXPECT_EQ(toLower("already lower"), "already lower");
+
+    // Only 'A'-'Z' are remapped. Neighbouring punctuation and digits, which a
+    // buggy range check could catch, must survive untouched.
+    EXPECT_EQ(toLower("@[`{_^"), "@[`{_^");
+}
+
+// Both helpers are documented as depending only on their input. Guard that by
+// checking the bytes just outside ASCII, which a locale-aware isspace/tolower
+// could classify differently.
+TEST_F(StringUtilitiesTest, trimAndLowerIgnoreLocale)
+{
+    // 0xA0 is NO-BREAK SPACE in Latin-1 and is whitespace to some locales.
+    std::string const nbsp("\xA0", 1);
+    EXPECT_EQ(trimWhitespace(nbsp), nbsp);
+    EXPECT_EQ(trimWhitespace(" " + nbsp + " "), nbsp);
+
+    // 0xC0 is LATIN CAPITAL LETTER A WITH GRAVE in Latin-1.
+    std::string const agrave("\xC0", 1);
+    EXPECT_EQ(toLower(agrave), agrave);
+}
+
 }  // namespace xrpl

src/xrpld/app/main/GRPCServer.cpp

@@ -9,6 +9,7 @@
 
 #include <xrpl/basics/FileUtilities.h>
 #include <xrpl/basics/Log.h>
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/contract.h>
 #include <xrpl/beast/core/CurrentThreadName.h>
 #include <xrpl/beast/net/IPAddressConversion.h>
@@ -24,7 +25,6 @@
 #include <xrpl/resource/Fees.h>
 #include <xrpl/server/InfoSub.h>
 
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/asio/ip/address.hpp>
 #include <boost/asio/ip/tcp.hpp>
 #include <boost/icl/interval_set.hpp>
@@ -371,7 +371,7 @@ GRPCServerImpl::GRPCServerImpl(Application& app)
                 std::string ip;
                 while (std::getline(ss, ip, ','))
                 {
-                    boost::algorithm::trim(ip);
+                    ip = trimWhitespace(ip);
                     auto const addr = boost::asio::ip::make_address(ip);
 
                     if (addr.is_unspecified())

src/xrpld/app/main/Main.cpp

@@ -6,6 +6,7 @@
 
 #include <xrpl/basics/Log.h>
 #include <xrpl/basics/SlabAllocator.h>
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/base_uint.h>
 #include <xrpl/beast/core/CurrentThreadName.h>
 #include <xrpl/beast/net/IPEndpoint.h>
@@ -21,7 +22,6 @@
 
 #include <boost/algorithm/string/classification.hpp>
 #include <boost/algorithm/string/split.hpp>
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/process/v1/args.hpp>
 #include <boost/process/v1/child.hpp>  // IWYU pragma: keep
 #include <boost/process/v1/exe.hpp>
@@ -211,7 +211,7 @@ public:
         boost::split(v, patterns, boost::algorithm::is_any_of(","));
         selectors_.reserve(v.size());
         std::ranges::for_each(v, [this](std::string s) {
-            boost::trim(s);
+            s = trimWhitespace(s);
             if (selectors_.empty() || !s.empty())
                 selectors_.emplace_back(beast::unit_test::Selector::ModeT::Automatch, s);
         });
@@ -614,7 +614,7 @@ run(int argc, char** argv)
                 std::vector<std::uint32_t> result;
                 for (auto& s : strVec)
                 {
-                    boost::trim(s);
+                    s = trimWhitespace(s);
                     if (!s.empty())
                         result.push_back(std::stoi(s));
                 }

src/xrpld/core/detail/Config.cpp

@@ -20,7 +20,6 @@
 #include <boost/algorithm/string/predicate.hpp>
 #include <boost/algorithm/string/replace.hpp>
 #include <boost/algorithm/string/split.hpp>
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/filesystem/operations.hpp>
 #include <boost/filesystem/path.hpp>
 #include <boost/format/free_funcs.hpp>
@@ -185,7 +184,7 @@ parseIniFile(std::string const& strInput, bool const bTrim)
     for (auto& strValue : vLines)
     {
         if (bTrim)
-            boost::algorithm::trim(strValue);
+            strValue = trimWhitespace(strValue);
 
         if (strValue.empty() || strValue[0] == '#')
         {

src/xrpld/rpc/detail/ServerHandler.cpp

@@ -8,6 +8,7 @@
 #include <xrpld/rpc/detail/WSInfoSub.h>
 
 #include <xrpl/basics/Log.h>
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/base64.h>
 #include <xrpl/basics/contract.h>
 #include <xrpl/basics/make_SSLContext.h>
@@ -44,7 +45,6 @@
 #include <xrpl/server/WSSession.h>
 #include <xrpl/server/detail/JSONRPCUtil.h>
 
-#include <boost/algorithm/string/trim.hpp>
 #include <boost/asio/buffer.hpp>
 #include <boost/asio/io_context.hpp>
 #include <boost/asio/ip/tcp.hpp>
@@ -113,7 +113,7 @@ authorized(Port const& port, std::map<std::string, std::string> const& h)
     if ((it == h.end()) || (!it->second.starts_with("Basic ")))
         return false;
     std::string strUserPass64 = it->second.substr(6);
-    boost::trim(strUserPass64);
+    strUserPass64 = trimWhitespace(strUserPass64);
     std::string const strUserPass = base64Decode(strUserPass64);
     std::string::size_type const nColon = strUserPass.find(':');
     if (nColon == std::string::npos)

src/xrpld/rpc/handlers/account/AccountInfo.cpp

@@ -5,6 +5,7 @@
 
 #include <xrpl/basics/Blob.h>
 #include <xrpl/basics/Slice.h>
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/strHex.h>
 #include <xrpl/beast/utility/instrumentation.h>
 #include <xrpl/json/json_forwards.h>
@@ -22,7 +23,6 @@
 #include <xrpl/protocol/XRPAmount.h>
 #include <xrpl/protocol/jss.h>
 
-#include <boost/algorithm/string/case_conv.hpp>
 #include <boost/format/free_funcs.hpp>
 
 #include <array>
@@ -57,7 +57,7 @@ injectSLE(json::Value& jv, SLE const& sle)
         auto const& hash = sle.getFieldH128(sfEmailHash);
         Blob const b(hash.begin(), hash.end());
         std::string md5 = strHex(makeSlice(b));
-        boost::to_lower(md5);
+        md5 = toLower(md5);
         // VFALCO TODO Give a name to this constant and move it
         //             to a more visible location.
         jv[jss::urlgravatar] = str(boost::format("https://www.gravatar.com/avatar/%s") % md5);

src/xrpld/rpc/handlers/admin/data/CanDelete.cpp

@@ -3,14 +3,13 @@
 #include <xrpld/app/misc/SHAMapStore.h>
 #include <xrpld/rpc/Context.h>
 
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/base_uint.h>
 #include <xrpl/beast/core/LexicalCast.h>
 #include <xrpl/json/json_value.h>
 #include <xrpl/protocol/ErrorCodes.h>
 #include <xrpl/protocol/jss.h>
 
-#include <boost/algorithm/string/case_conv.hpp>
-
 #include <cstdint>
 #include <limits>
 #include <string>
@@ -38,7 +37,7 @@ doCanDelete(rpc::JsonContext& context)
         else
         {
             std::string canDeleteStr = canDelete.asString();
-            boost::to_lower(canDeleteStr);
+            canDeleteStr = toLower(canDeleteStr);
 
             if (canDeleteStr.find_first_not_of("0123456789") == std::string::npos)
             {

src/xrpld/rpc/handlers/server_info/ServerDefinitions.cpp

@@ -2,6 +2,7 @@
 
 #include <xrpld/rpc/Context.h>
 
+#include <xrpl/basics/StringUtilities.h>
 #include <xrpl/basics/base_uint.h>
 #include <xrpl/json/json_value.h>
 #include <xrpl/json/json_writer.h>
@@ -14,7 +15,6 @@
 #include <xrpl/protocol/digest.h>
 #include <xrpl/protocol/jss.h>
 
-#include <boost/algorithm/string/case_conv.hpp>
 #include <boost/algorithm/string/replace.hpp>
 
 #include <cstddef>
@@ -106,7 +106,7 @@ ServerDefinitions::translate(std::string const& inp)
         std::string token = inpToProcess.substr(0, pos);
         if (token.size() > 1)
         {
-            boost::algorithm::to_lower(token);
+            token = toLower(token);
             token[0] -= ('a' - 'A');
             out += token;
         }