Merge remote-tracking branch 'origin/topic/timw/cpp20-starts-and-ends-with'

* origin/topic/timw/cpp20-starts-and-ends-with:
  Use std::string/string_view versions of starts_with/ends_with where appropriate

.clang-tidy

@@ -62,7 +62,6 @@ Checks: [-*,
 
          # These are features in newer version of C++ that we don't have
          # access to yet.
-         -modernize-use-starts-ends-with,
          -modernize-use-std-format,
          -modernize-use-std-numbers,
          -modernize-use-std-print,

CHANGES

@@ -1,3 +1,12 @@
+8.0.0-dev.682 | 2025-07-17 09:09:28 -0700
+
+  * Use std::string/string_view versions of starts_with/ends_with where appropriate (Tim Wojtulewicz, Corelight)
+
+    The util:: versions of these methods remain as a thin wrapper around them so
+    they can be used with const char* arguments. Otherwise callers have to manually
+    make string_view objects from the input.
+    s Please enter the commit message for your changes. Lines starting
+
 8.0.0-dev.680 | 2025-07-17 08:40:49 -0700
 
   * Add nolint for enum size for HILTI_RT_ENUM use (Tim Wojtulewicz, Corelight)

NEWS

@@ -348,12 +348,17 @@ Deprecated Functionality
   tuples. The ConnTuple struct, used by this deprecated Connection constructor,
   is now deprecated as well.
 
+
 - The ``zeek::filesystem`` namespace alias is deprecated in favor of using
   ``std::filesystem`` directly. Similarly, the ``ghc::filesystem`` submodule stored in
   ``auxil/filessytem`` has been removed and the files included from it in the Zeek
   installation will no longer be installed. Builds won't warn about the deprecation of
   ``zeek::filesystem`` due to limitations of how we can mark deprecations in C++.
 
+- The ``zeek::util::starts_with`` and ``zeek::util::ends_with`` functions are deprecated.
+  ``std::string`` and ``std::string_view`` added ``begins_with`` and ``ends_with`` methods
+  in C++ 20, and those should be used instead.
+
 Zeek 7.2.0
 ==========
 

VERSION

@@ -1 +1 @@
-8.0.0-dev.680
+8.0.0-dev.682

src/DebugLogger.cc

@@ -111,7 +111,7 @@ void DebugLogger::EnableStreams(const char* s) {
             exit(0);
         }
 
-        if ( util::starts_with(ltok, "plugin-") ) {
+        if ( std::string_view{ltok}.starts_with("plugin-") ) {
             // Cannot verify this at this time, plugins may not
             // have been loaded.
             enabled_streams.insert(std::move(ltok));
@@ -145,7 +145,7 @@ bool DebugLogger::CheckStreams(const std::set<std::string>& plugin_names) {
         available_plugin_streams.insert(PluginStreamName(p));
 
     for ( const auto& stream : enabled_streams ) {
-        if ( ! util::starts_with(stream, "plugin-") )
+        if ( ! stream.starts_with("plugin-") )
             continue;
 
         if ( available_plugin_streams.count(stream) == 0 ) {

src/Options.cc

@@ -319,13 +319,8 @@ Options parse_cmdline(int argc, char** argv) {
     }
     else {
         if ( argc > 1 ) {
-            auto endsWith = [](const std::string& str, const std::string& suffix) {
-                return str.size() >= suffix.size() &&
-                       0 == str.compare(str.size() - suffix.size(), suffix.size(), suffix);
-            };
-
             auto i = 0;
-            for ( ; i < argc && ! endsWith(argv[i], "--"); ++i ) {
+            for ( ; i < argc && ! util::ends_with(argv[i], "--"); ++i ) {
                 zeek_args.emplace_back(argv[i]);
             }
 
@@ -333,17 +328,17 @@ Options parse_cmdline(int argc, char** argv) {
                 // If a script is invoked with Zeek as the interpreter, the arguments provided
                 // directly in the interpreter line of the script won't be broken apart in the
                 // argv on Linux so we split it up here.
-                if ( endsWith(argv[i], "--") && zeek_args.size() == 1 ) {
+                if ( util::ends_with(argv[i], "--") && zeek_args.size() == 1 ) {
                     std::istringstream iss(argv[i]);
                     for ( std::string s; iss >> s; ) {
-                        if ( ! endsWith(s, "--") ) {
+                        if ( ! s.ends_with("--") ) {
                             zeek_args.emplace_back(s);
                         }
                     }
                 }
 
                 // There is an additional increment here to skip over the "--" if it was found.
-                if ( endsWith(argv[i], "--") )
+                if ( util::ends_with(argv[i], "--") )
                     ++i;
 
                 // The first argument after the double hyphens in implicitly a script name.

src/Scope.cc

@@ -91,10 +91,10 @@ TraversalCode Scope::Traverse(TraversalCallback* cb) const {
 
 const IDPtr& lookup_ID(const char* name, const char* curr_module, bool no_global, bool same_module_only,
                        bool check_export) {
-    bool explicit_global = zeek::util::starts_with(name, "::");
+    bool explicit_global = std::string_view{name}.starts_with("::");
 
     static std::string global_prefix = util::fmt("%s::", GLOBAL_MODULE_NAME);
-    if ( zeek::util::starts_with(name, global_prefix) )
+    if ( std::string_view{name}.starts_with(global_prefix) )
         reporter->Error("Using GLOBAL:: as a prefix is invalid. Use :: instead (%s)", name);
 
     std::string fullname = make_full_var_name(curr_module, name);

src/Val.cc

@@ -1053,8 +1053,7 @@ static zeek::expected<ValPtr, std::string> BuildVal(const rapidjson::Value& j, c
                 candidate.erase(candidate.size() - 1);
             }
             // Remove any surrounding "^?(" and ")$?", automatically added below.
-            if ( candidate.size() > 6 && candidate.substr(0, 3) == "^?(" &&
-                 candidate.substr(candidate.size() - 3, 3) == ")$?" ) {
+            if ( candidate.size() > 6 && candidate.starts_with("^?(") && candidate.ends_with(")$?") ) {
                 candidate.erase(0, 3);
                 candidate.erase(candidate.size() - 3);
             }

src/analyzer/protocol/irc/IRC.cc

@@ -533,7 +533,7 @@ void IRC_Analyzer::DeliverStream(int length, const u_char* line, bool orig) {
             message = message.substr(1); // DCC
 
         // Check for DCC messages.
-        if ( message.size() > 3 && message.substr(0, 3) == "DCC" ) {
+        if ( message.size() > 3 && message.starts_with("DCC") ) {
             if ( message[message.size() - 1] == 1 )
                 message = message.substr(0, message.size() - 1);
 

src/broker/Data.cc

@@ -767,7 +767,7 @@ std::optional<broker::data> val_to_data(const Val* v) {
             broker::vector rval;
             rval.emplace_back(f->GetName());
 
-            if ( f->GetName().find("lambda_<") == 0 ) {
+            if ( f->GetName().starts_with("lambda_<") ) {
                 // Only ScriptFuncs have closures.
                 if ( auto b = dynamic_cast<const zeek::detail::ScriptFunc*>(f) ) {
                     auto bc = b->SerializeCaptures();

src/input/readers/ascii/Ascii.cc

@@ -241,7 +241,7 @@ bool Ascii::GetLine(string& str) {
         if ( str[0] != '#' )
             return true;
 
-        if ( (str.length() > 8) && (str.compare(0, 7, "#fields") == 0) && (str[7] == separator[0]) ) {
+        if ( (str.length() > 8) && str.starts_with("#fields") && (str[7] == separator[0]) ) {
             str = str.substr(8);
             return true;
         }

src/logging/writers/ascii/Ascii.h

@@ -40,7 +40,7 @@ private:
 
     static void RotateLeftoverLogs();
 
-    bool IsSpecial(const std::string& path) { return path.find("/dev/") == 0; }
+    bool IsSpecial(const std::string& path) { return path.starts_with("/dev/"); }
     bool WriteHeader(const std::string& path);
     bool WriteHeaderField(const std::string& key, const std::string& value);
     void CloseFile(double t);

src/script_opt/ProfileFunc.cc

@@ -1467,7 +1467,7 @@ std::string func_name_at_loc(std::string fname, const Location* loc) {
 
     auto mod_prefix = module + "::";
 
-    if ( fname.find(mod_prefix) == 0 )
+    if ( fname.starts_with(mod_prefix) )
         return fname; // it already has the module name
 
     return mod_prefix + fname;

src/storage/backend/redis/Redis.cc

@@ -635,7 +635,7 @@ void Redis::HandleInfoResult(redisReply* reply) {
                 continue;
 
             // We only care about the redis_version entry. Skip anything else.
-            if ( ! util::starts_with(e, "redis_version:") )
+            if ( ! e.starts_with("redis_version:") )
                 continue;
 
             auto splits = util::split(e, ':');

src/telemetry/Manager.cc

@@ -189,7 +189,7 @@ RecordValPtr Manager::GetMetricOptsRecord(const prometheus::MetricFamily& metric
     record_val->Assign(prefix_idx, make_intrusive<zeek::StringVal>(prefix));
 
     // Assume that a metric ending with _total is always a summed metric so we can set that.
-    record_val->Assign(is_total_idx, val_mgr->Bool(util::ends_with(metric_family.name, "_total")));
+    record_val->Assign(is_total_idx, val_mgr->Bool(metric_family.name.ends_with("_total")));
 
     if ( metric_family.type == prometheus::MetricType::Counter )
         record_val->Assign(metric_type_idx, zeek::BifType::Enum::Telemetry::MetricType->GetEnumVal(

src/telemetry/Utils.cc

@@ -26,9 +26,9 @@ std::string BuildFullPrometheusName(std::string_view prefix, std::string_view na
     });
 
     // Suffixes of full metric names of _total are reserved by Prometheus. Disallow their use here.
-    if ( util::ends_with(fn, "_total") )
+    if ( fn.ends_with("_total") )
         reporter->FatalError("Metric names cannot end with '_total'");
-    else if ( unit == "total" || util::ends_with(unit, "_total") )
+    else if ( unit == "total" || unit.ends_with("_total") )
         reporter->FatalError("Metric units cannot end with '_total'");
 
     // We were previously using "1" to mean "no unit value" for whatever reason, so we have to handle that now

src/util.cc

@@ -91,7 +91,7 @@ std::string extract_ip(const std::string& i) {
     if ( s.size() > 0 && s[0] == '[' )
         s.erase(0, 1);
 
-    if ( s.size() > 1 && s.substr(0, 2) == "0x" )
+    if ( s.starts_with("0x") )
         s.erase(0, 2);
 
     if ( size_t pos = s.find(']'); pos != std::string::npos )
@@ -599,7 +599,7 @@ string normalize_path(std::string_view path) {
     }
     // "//" interferes with std::weakly_canonical
     string stringPath = string(path);
-    if ( stringPath._Starts_with("//") ) {
+    if ( stringPath.starts_with("//") ) {
         stringPath.erase(0, 2);
     }
     return std::filesystem::path(stringPath).lexically_normal().string();
@@ -672,7 +672,7 @@ string without_zeekpath_component(std::string_view path) {
     for ( const auto& p : paths ) {
         string common = normalize_path(p);
 
-        if ( rval.find(common) != 0 )
+        if ( ! rval.starts_with(common) )
             continue;
 
         // Found the containing directory.
@@ -1034,31 +1034,8 @@ TEST_CASE("util streq") {
 
 bool streq(const char* s1, const char* s2) { return strcmp(s1, s2) == 0; }
 
-bool starts_with(std::string_view s, std::string_view beginning) {
-    if ( beginning.size() > s.size() )
-        return false;
-
-    return std::equal(beginning.begin(), beginning.end(), s.begin());
-}
-
-TEST_CASE("util starts_with") {
-    CHECK(starts_with("abcde", "ab") == true);
-    CHECK(starts_with("abcde", "de") == false);
-    CHECK(starts_with("abcde", "abcedf") == false);
-}
-
-bool ends_with(std::string_view s, std::string_view ending) {
-    if ( ending.size() > s.size() )
-        return false;
-
-    return std::equal(ending.rbegin(), ending.rend(), s.rbegin());
-}
-
-TEST_CASE("util ends_with") {
-    CHECK(ends_with("abcde", "de") == true);
-    CHECK(ends_with("abcde", "fg") == false);
-    CHECK(ends_with("abcde", "abcedf") == false);
-}
+bool starts_with(std::string_view s, std::string_view beginning) { return s.starts_with(beginning); }
+bool ends_with(std::string_view s, std::string_view ending) { return s.ends_with(ending); }
 
 char* skip_whitespace(char* s) {
     while ( *s == ' ' || *s == '\t' )

src/zeekygen/IdentifierInfo.cc

@@ -37,7 +37,7 @@ void IdentifierInfo::AddRedef(const string& script, zeek::detail::InitClass ic,
                               vector<string> comments) {
     bool omit_value = false;
     for ( auto it = comments.begin(); it != comments.end(); ++it ) {
-        if ( it->find("@docs-omit-value") == 0 ) {
+        if ( it->starts_with("@docs-omit-value") ) {
             comments.erase(it);
             omit_value = true;
             break;

src/zeekygen/utils.cc

@@ -173,7 +173,7 @@ std::optional<std::string> source_code_range(const zeek::detail::ID* id) {
             if ( id->IsType() ) {
                 loc = type->GetLocationInfo();
 
-                if ( zeek::util::ends_with(loc->FileName(), ".bif.zeek") )
+                if ( std::string_view{loc->FileName()}.ends_with(".bif.zeek") )
                     // Source code won't be available to reference, so fall back
                     // to identifier location which may actually be in a regular
                     // .zeek script.