From a9ac3da508239be8a736e1ec19cc0f9766a834d1 Mon Sep 17 00:00:00 2001
From: Jurjen Bokma <j.bokma@rug.nl>
Date: Sun, 11 May 2025 12:20:42 +0200
Subject: [PATCH] Accept sizes with k/M/G/T/E suffixes.

---
 CmdlineParser.cc | 69 ++++++++++++++++++++++++++++++++++++++++++++++++
 CmdlineParser.hh |  4 +++
 rdfind.1         |  3 +++
 rdfind.cc        |  9 ++++---
 4 files changed, 82 insertions(+), 3 deletions(-)

diff --git a/CmdlineParser.cc b/CmdlineParser.cc
index 6a930df..ea9f5e0 100644
--- a/CmdlineParser.cc
+++ b/CmdlineParser.cc
@@ -10,6 +10,7 @@
 #include <cstdlib>
 #include <cstring>
 #include <iostream>
+#include <stdexcept>
 
 // project
 #include "CmdlineParser.hh"
@@ -97,3 +98,71 @@ Parser::current_arg_is(const char* what) const
 {
   return 0 == std::strcmp(get_current_arg(), what);
 }
+
+namespace {
+
+    enum FILESIZE : long long
+    {
+        k = 1024,
+        K = k,
+        M = k * k,
+        G = k * M,
+        T = k * G,
+        P = k * T,
+        E = k * P,
+        // too big for signed long long // Z = k * E,
+        // too big for signed long long // Y = k * Z,
+        // too big for signed long long // R = k * Y,
+        // too big for signed long long // Q = k * R,
+    };
+
+    
+    Fileinfo::filesizetype value_of_file_size_suffix(char suffix)
+    {
+        switch (toupper(suffix))
+        {
+        // No 'B': too ambiguous.
+        case 'K': return FILESIZE::k;
+        case 'M': return FILESIZE::M;
+        case 'G': return FILESIZE::G;
+        case 'T': return FILESIZE::T;
+        case 'P': return FILESIZE::P;
+        case 'E': return FILESIZE::E;
+        // case 'Z': return FILESIZE::Z;
+        // case 'Y': return FILESIZE::Y;
+        // case 'R': return FILESIZE::R;
+        // case 'Q': return FILESIZE::Q;
+        default: throw std::runtime_error(
+            "The program only understands file size suffixes "
+            "K, M, G, T, P or E. Lower case works too.");
+        };
+    }
+
+}
+
+Fileinfo::filesizetype long Parser::read_file_size(char const *text)
+{
+
+    using filesizetype = Fileinfo::filesizetype;
+    
+    size_t pos = 0;
+    filesizetype const without_suffix = std::stoll(text, &pos);
+
+    size_t const len = strlen(text);
+    filesizetype const multiplier = 
+        (len - pos > 1) ?
+        throw std::runtime_error(
+            "The program only understands single-letter file size suffixes.")
+        :
+        (len > pos) ?
+        value_of_file_size_suffix(text[pos])
+        :
+        1;
+
+    filesizetype const retval = multiplier * without_suffix;
+
+    if (retval / multiplier != without_suffix)
+        throw std::runtime_error("File size overflows long long.");
+
+    return retval;
+}
diff --git a/CmdlineParser.hh b/CmdlineParser.hh
index 84210e0..54ea74b 100644
--- a/CmdlineParser.hh
+++ b/CmdlineParser.hh
@@ -7,6 +7,8 @@
 #ifndef RDFIND_CMDLINEPARSER_HH_
 #define RDFIND_CMDLINEPARSER_HH_
 
+#include "Fileinfo.hh"    //for filesizetype
+
 /**
  * Command line parser, designed to be easy to use for rdfind.
  * It will signal user errors by a helpful printout followed by
@@ -49,6 +51,8 @@ public:
 
   bool current_arg_is(const char* what) const;
 
+  static Fileinfo::filesizetype read_file_size(char const *text);
+    
 private:
   const int m_argc{};
   const char** m_argv{};
diff --git a/rdfind.1 b/rdfind.1
index 17a50ee..5e250bb 100644
--- a/rdfind.1
+++ b/rdfind.1
@@ -64,6 +64,9 @@ Ignore empty files. Setting this to true (the default) is equivalent to
 .BR \-minsize " "\fIN\fR
 Ignores files with less than N bytes. Default is 1, meaning empty files
 are ignored.
+
+File and buffer sizes can be given with suffix k, M, G etc. up to E,
+meaning powers of 1024.
 .TP
 .BR \-maxsize " "\fIN\fR
 Ignores files with N bytes or more. Default is 0, which means this check
diff --git a/rdfind.cc b/rdfind.cc
index 71d7588..06002c9 100644
--- a/rdfind.cc
+++ b/rdfind.cc
@@ -58,6 +58,9 @@ usage()
     << "                                  false implies -minsize 0)\n"
     << " -minsize N        (N=1)          ignores files with size less than N "
        "bytes\n"
+    << "                                  For -minsize, -maxsize and -buffersize,\n"
+       "                                  Suffixes 'K', 'M', 'G' etc. up to 'E'\n"
+       "                                  are accepted. Factors are 1024, not 1000.\n"
     << " -maxsize N        (N=0)          ignores files with size N "
        "bytes and larger (use 0 to disable this check).\n"
     << " -followsymlinks    true |(false) follow symlinks\n"
@@ -153,13 +156,13 @@ parseOptions(Parser& parser)
         o.minimumfilesize = 0;
       }
     } else if (parser.try_parse_string("-minsize")) {
-      const long long minsize = std::stoll(parser.get_parsed_string());
+        const long long minsize = Parser::read_file_size(parser.get_parsed_string());
       if (minsize < 0) {
         throw std::runtime_error("negative value of minsize not allowed");
       }
       o.minimumfilesize = minsize;
     } else if (parser.try_parse_string("-maxsize")) {
-      const long long maxsize = std::stoll(parser.get_parsed_string());
+        const long long maxsize = Parser::read_file_size(parser.get_parsed_string());
       if (maxsize < 0) {
         throw std::runtime_error("negative value of maxsize not allowed");
       }
@@ -199,7 +202,7 @@ parseOptions(Parser& parser)
         std::exit(EXIT_FAILURE);
       }
     } else if (parser.try_parse_string("-buffersize")) {
-      const long buffersize = std::stoll(parser.get_parsed_string());
+      const long buffersize = Parser::read_file_size(parser.get_parsed_string());
       constexpr long max_buffersize = 128 << 20;
       if (buffersize <= 0) {
         std::cerr << "a negative or zero buffersize is not allowed\n";