Skip firstbytes / lastbytes

.gitignore

@@ -23,3 +23,4 @@ test-driver
 nettle32bit/
 *.log
 rdfind-*.tar.gz
+.vs/slnx.sqlite

rdfind.1

@@ -80,6 +80,9 @@ is true.
 What type of checksum to be used: md5, sha1 or sha256. The default is
 sha1 since version 1.4.0.
 .TP
+.BR \-skip " " \fIfirstbytes\fR|\fIlastbytes\fR
+Skip some steps during the candidates pruning. This argument can be repeated.
+.TP
 .BR \-deterministic " " \fItrue\fR|\fIfalse\fR
 If set (the default), sort files of equal rank in an unspecified but
 deterministic order. This makes the behaviour independent of in which

rdfind.cc

@@ -62,6 +62,8 @@ usage()
        "device and inode\n"
     << " -checksum           md5 |(sha1)| sha256\n"
     << "                                  checksum type\n"
+    << " -skip           firstbytes | lastbytes\n"
+    << "                                  skip some check\n"
     << " -deterministic    (true)| false  makes results independent of order\n"
     << "                                  from listing the filesystem\n"
     << " -makesymlinks      true |(false) replace duplicate files with "
@@ -102,6 +104,8 @@ struct Options
   bool followsymlinks = false;        // follow symlinks
   bool dryrun = false;                // only dryrun, don't destroy anything
   bool remove_identical_inode = true; // remove files with identical inodes
+  bool usefirstbytes = true;       // use first bytes to check for differences
+  bool uselastbytes = true;       // use last bytes to check for differences
   bool usemd5 = false;       // use md5 checksum to check for similarity
   bool usesha1 = false;      // use sha1 checksum to check for similarity
   bool usesha256 = false;    // use sha256 checksum to check for similarity
@@ -178,6 +182,16 @@ parseOptions(Parser& parser)
                   << parser.get_parsed_string() << "\"\n";
         std::exit(EXIT_FAILURE);
       }
+    } else if (parser.try_parse_string("-skip")) {
+      if (parser.parsed_string_is("firstbytes")) {
+        o.usefirstbytes = false;
+      } else if (parser.parsed_string_is("lastbytes")) {
+        o.uselastbytes = false;
+      } else {
+        std::cerr << "expected firstbytes/lastbytes, not \""
+                  << parser.get_parsed_string() << "\"\n";
+        std::exit(EXIT_FAILURE);
+      }
     } else if (parser.try_parse_string("-sleep")) {
       const auto nextarg = std::string(parser.get_parsed_string());
       if (nextarg == "1ms") {
@@ -352,9 +366,15 @@ main(int narg, const char* argv[])
   // candidates. start looking at the contents.
   std::vector<std::pair<Fileinfo::readtobuffermode, const char*>> modes{
     { Fileinfo::readtobuffermode::NOT_DEFINED, "" },
-    { Fileinfo::readtobuffermode::READ_FIRST_BYTES, "first bytes" },
-    { Fileinfo::readtobuffermode::READ_LAST_BYTES, "last bytes" },
   };
+  if (o.usefirstbytes) {
+    modes.emplace_back(Fileinfo::readtobuffermode::READ_FIRST_BYTES,
+                       "first bytes");
+  }
+  if (o.uselastbytes) {
+    modes.emplace_back(Fileinfo::readtobuffermode::READ_LAST_BYTES,
+                       "last bytes");
+  }
   if (o.usemd5) {
     modes.emplace_back(Fileinfo::readtobuffermode::CREATE_MD5_CHECKSUM,
                        "md5 checksum");

testcases/skip_options.sh

@@ -0,0 +1,22 @@
+#!/bin/sh
+# Test that selection of skip works as expected.
+
+
+set -e
+. "$(dirname "$0")/common_funcs.sh"
+
+
+
+
+
+for skiptype in firstbytes lastbytes; do
+   reset_teststate
+   dbgecho "trying skip $skiptype"
+   echo skiptype >a
+   echo skiptype >b
+   $rdfind -skip $skiptype -deleteduplicates true a b
+   [ -e a ]
+   [ ! -e b ]
+done
+
+dbgecho "all is good in this test!"