util: ParseByteUnits - Parse a string with suffix unit [k|K|m|M|g|G|t|T]

A convenience utility for human readable arguments/config e.g. -maxuploadtarget=500g

2 files changed, 76 insertions, 0 deletions

diff --git a/src/util/strencodings.cpp b/src/util/strencodings.cpp
index 15bd07b374..5e7c99f140 100644
--- a/src/util/strencodings.cpp
+++ b/src/util/strencodings.cpp
@@ -11,6 +11,7 @@
 #include <algorithm>
 #include <cstdlib>
 #include <cstring>
+#include <limits>
 #include <optional>
 
 static const std::string CHARS_ALPHA_NUM = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
@@ -526,3 +527,48 @@ std::string HexStr(const Span<const uint8_t> s)
     assert(it == rv.end());
     return rv;
 }
+
+std::optional<uint64_t> ParseByteUnits(const std::string& str, ByteUnit default_multiplier)
+{
+    if (str.empty()) {
+        return std::nullopt;
+    }
+    auto multiplier = default_multiplier;
+    char unit = str.back();
+    switch (unit) {
+    case 'k':
+        multiplier = ByteUnit::k;
+        break;
+    case 'K':
+        multiplier = ByteUnit::K;
+        break;
+    case 'm':
+        multiplier = ByteUnit::m;
+        break;
+    case 'M':
+        multiplier = ByteUnit::M;
+        break;
+    case 'g':
+        multiplier = ByteUnit::g;
+        break;
+    case 'G':
+        multiplier = ByteUnit::G;
+        break;
+    case 't':
+        multiplier = ByteUnit::t;
+        break;
+    case 'T':
+        multiplier = ByteUnit::T;
+        break;
+    default:
+        unit = 0;
+        break;
+    }
+
+    uint64_t unit_amount = static_cast<uint64_t>(multiplier);
+    auto parsed_num = ToIntegral<uint64_t>(unit ? str.substr(0, str.size() - 1) : str);
+    if (!parsed_num || parsed_num > std::numeric_limits<uint64_t>::max() / unit_amount) { // check overflow
+        return std::nullopt;
+    }
+    return *parsed_num * unit_amount;
+}
diff --git a/src/util/strencodings.h b/src/util/strencodings.h
index eedb5ec2f8..bc0a6233ab 100644
--- a/src/util/strencodings.h
+++ b/src/util/strencodings.h
@@ -30,6 +30,23 @@ enum SafeChars
 };
 
 /**
+ * Used by ParseByteUnits()
+ * Lowercase base 1000
+ * Uppercase base 1024
+*/
+enum class ByteUnit : uint64_t {
+    NOOP = 1ULL,
+    k = 1000ULL,
+    K = 1024ULL,
+    m = 1'000'000ULL,
+    M = 1ULL << 20,
+    g = 1'000'000'000ULL,
+    G = 1ULL << 30,
+    t = 1'000'000'000'000ULL,
+    T = 1ULL << 40,
+};
+
+/**
 * Remove unsafe chars. Safe chars chosen to allow simple messages/URLs/email
 * addresses, but avoid anything even possibly remotely dangerous like & or >
 * @param[in] str    The string to sanitize
@@ -305,4 +322,17 @@ std::string ToUpper(const std::string& str);
  */
 std::string Capitalize(std::string str);
 
+/**
+ * Parse a string with suffix unit [k|K|m|M|g|G|t|T].
+ * Must be a whole integer, fractions not allowed (0.5t), no whitespace or +-
+ * Lowercase units are 1000 base. Uppercase units are 1024 base.
+ * Examples: 2m,27M,19g,41T
+ *
+ * @param[in] str                  the string to convert into bytes
+ * @param[in] default_multiplier   if no unit is found in str use this unit
+ * @returns                        optional uint64_t bytes from str or nullopt
+ *                                 if ToIntegral is false, str is empty, trailing whitespace or overflow
+ */
+std::optional<uint64_t> ParseByteUnits(const std::string& str, ByteUnit default_multiplier);
+
 #endif // BITCOIN_UTIL_STRENCODINGS_H