Exclude option

.gitignore

@@ -3,3 +3,4 @@ squash
 test.tar
 test_*
 restored
+excludes

Cargo.lock

@@ -14,6 +14,7 @@ dependencies = [
  "murmurhash3 0.0.5 (registry+https://github.com/rust-lang/crates.io-index)",
  "pkg-config 0.3.9 (registry+https://github.com/rust-lang/crates.io-index)",
  "quick-error 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "rmp-serde 0.12.2 (registry+https://github.com/rust-lang/crates.io-index)",
  "rustc-serialize 0.3.23 (registry+https://github.com/rust-lang/crates.io-index)",
  "serde 0.9.11 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -23,6 +24,14 @@ dependencies = [
  "squash-sys 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
+[[package]]
+name = "aho-corasick"
+version = "0.6.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "memchr 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "ansi_term"
 version = "0.9.0"
@@ -136,6 +145,14 @@ name = "log"
 version = "0.3.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
+[[package]]
+name = "memchr"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "libc 0.2.21 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "mmap"
 version = "0.1.1"
@@ -205,6 +222,23 @@ name = "redox_syscall"
 version = "0.1.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
+[[package]]
+name = "regex"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "aho-corasick 0.6.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "memchr 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex-syntax 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "thread_local 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "utf8-ranges 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+
 [[package]]
 name = "rmp"
 version = "0.8.5"
@@ -295,6 +329,24 @@ dependencies = [
  "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
+[[package]]
+name = "thread-id"
+version = "3.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libc 0.2.21 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "thread_local"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "thread-id 3.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "unreachable 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "time"
 version = "0.1.36"
@@ -316,11 +368,29 @@ name = "unicode-width"
 version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
+[[package]]
+name = "unreachable"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "utf8-ranges"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+
 [[package]]
 name = "vec_map"
 version = "0.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
+[[package]]
+name = "void"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+
 [[package]]
 name = "winapi"
 version = "0.2.8"
@@ -340,6 +410,7 @@ dependencies = [
 ]
 
 [metadata]
+"checksum aho-corasick 0.6.3 (registry+https://github.com/rust-lang/crates.io-index)" = "500909c4f87a9e52355b26626d890833e9e1d53ac566db76c36faa984b889699"
 "checksum ansi_term 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)" = "23ac7c30002a5accbf7e8987d0632fa6de155b7c3d39d0067317a391e00a2ef6"
 "checksum atty 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "d912da0db7fa85514874458ca3651fe2cddace8d0b0505571dbdcd41ab490159"
 "checksum bitflags 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)" = "aad18937a628ec6abcd26d1489012cc0e18c21798210f491af69ded9b881106d"
@@ -356,6 +427,7 @@ dependencies = [
 "checksum libsodium-sys 0.0.14 (registry+https://github.com/rust-lang/crates.io-index)" = "cbbc6e46017815abf8698de0ed4847fad45fd8cad2909ac38ac6de79673c1ad1"
 "checksum linked-hash-map 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)" = "6d262045c5b87c0861b3f004610afd0e2c851e2908d08b6c870cbb9d5f494ecd"
 "checksum log 0.3.7 (registry+https://github.com/rust-lang/crates.io-index)" = "5141eca02775a762cc6cd564d8d2c50f67c0ea3a372cbf1c51592b3e029e10ad"
+"checksum memchr 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "1dbccc0e46f1ea47b9f17e6d67c5a96bd27030519c519c9c91327e31275a47b4"
 "checksum mmap 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "0bc85448a6006dd2ba26a385a564a8a0f1f2c7e78c70f1a70b2e0f4af286b823"
 "checksum murmurhash3 0.0.5 (registry+https://github.com/rust-lang/crates.io-index)" = "a2983372caf4480544083767bf2d27defafe32af49ab4df3a0b7fc90793a3664"
 "checksum num 0.1.37 (registry+https://github.com/rust-lang/crates.io-index)" = "98b15ba84e910ea7a1973bccd3df7b31ae282bf9d8bd2897779950c9b8303d40"
@@ -366,6 +438,8 @@ dependencies = [
 "checksum quick-error 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "0aad603e8d7fb67da22dbdf1f4b826ce8829e406124109e73cf1b2454b93a71c"
 "checksum rand 0.3.15 (registry+https://github.com/rust-lang/crates.io-index)" = "022e0636ec2519ddae48154b028864bdce4eaf7d35226ab8e65c611be97b189d"
 "checksum redox_syscall 0.1.17 (registry+https://github.com/rust-lang/crates.io-index)" = "29dbdfd4b9df8ab31dec47c6087b7b13cbf4a776f335e4de8efba8288dda075b"
+"checksum regex 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "4278c17d0f6d62dfef0ab00028feb45bd7d2102843f80763474eeb1be8a10c01"
+"checksum regex-syntax 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "2f9191b1f57603095f105d317e375d19b1c9c5c3185ea9633a99a6dcbed04457"
 "checksum rmp 0.8.5 (registry+https://github.com/rust-lang/crates.io-index)" = "333f01365885cb192edaa22acb06d7e2f196bfd19d6969419e8b61307e0710ea"
 "checksum rmp-serde 0.12.2 (registry+https://github.com/rust-lang/crates.io-index)" = "06ec4d0cdea2645de5d0e649f90c3e654205d913e14adefa452257314a24e76e"
 "checksum rustc-serialize 0.3.23 (registry+https://github.com/rust-lang/crates.io-index)" = "684ce48436d6465300c9ea783b6b14c4361d6b8dcbb1375b486a69cc19e2dfb0"
@@ -377,10 +451,15 @@ dependencies = [
 "checksum strsim 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "b4d15c810519a91cf877e7e36e63fe068815c678181439f2f29e2562147c3694"
 "checksum tempdir 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "87974a6f5c1dfb344d733055601650059a3363de2a6104819293baff662132d6"
 "checksum term_size 0.2.3 (registry+https://github.com/rust-lang/crates.io-index)" = "07b6c1ac5b3fffd75073276bca1ceed01f67a28537097a2a9539e116e50fb21a"
+"checksum thread-id 3.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "4437c97558c70d129e40629a5b385b3fb1ffac301e63941335e4d354081ec14a"
+"checksum thread_local 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)" = "c85048c6260d17cf486ceae3282d9fb6b90be220bf5b28c400f5485ffc29f0c7"
 "checksum time 0.1.36 (registry+https://github.com/rust-lang/crates.io-index)" = "211b63c112206356ef1ff9b19355f43740fc3f85960c598a93d3a3d3ba7beade"
 "checksum unicode-segmentation 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "18127285758f0e2c6cf325bb3f3d138a12fee27de4f23e146cd6a179f26c2cf3"
 "checksum unicode-width 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "bf3a113775714a22dcb774d8ea3655c53a32debae63a063acc00a91cc586245f"
+"checksum unreachable 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "1f2ae5ddb18e1c92664717616dd9549dde73f539f01bd7b77c2edb2446bdff91"
+"checksum utf8-ranges 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "662fab6525a98beff2921d7f61a39e7d59e0b425ebc7d0d9e66d316e55124122"
 "checksum vec_map 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)" = "f8cdc8b93bd0198ed872357fb2e667f7125646b1762f16d60b2c96350d361897"
+"checksum void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)" = "6a02e4885ed3bc0f2de90ea6dd45ebcbb66dacffe03547fadbb0eeae2770887d"
 "checksum winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)" = "167dc9d6949a9b857f3451275e911c3f44255842c1f7a76f33c55103a909087a"
 "checksum winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "2d315eee3b34aca4797b2da6b13ed88266e6d612562a0c46390af8299fc699bc"
 "checksum yaml-rust 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "e66366e18dc58b46801afbf2ca7661a9f59cc8c5962c29892b6039b4f86fa992"

Cargo.toml

@@ -22,6 +22,7 @@ byteorder = "1.0"
 ansi_term = "0.9"
 sodiumoxide = "0.0.14"
 filetime = "0.1"
+regex = "0.2"
 libc = "*"
 
 [build-dependencies]

src/cli/args.rs

@@ -20,7 +20,9 @@ pub enum Arguments {
         src_path: String,
         full: bool,
         reference: Option<String>,
-        same_device: bool
+        same_device: bool,
+        excludes: Vec<String>,
+        excludes_from: Option<String>
     },
     Restore {
         repo_path: String,
@@ -196,6 +198,8 @@ pub fn parse() -> Arguments {
             (@arg full: --full "create a full backup")
             (@arg reference: --ref +takes_value "the reference backup to use for partial backup")
             (@arg same_device: --xdev -x "do not cross filesystem boundaries")
+            (@arg exclude: --exclude -e ... +takes_value "exclude this path or file")
+            (@arg excludes_from: --excludesfrom +takes_value "read the list of exludes from this file")
             (@arg SRC: +required "source path to backup")
             (@arg BACKUP: +required "repository::backup path")
         )
@@ -308,6 +312,8 @@ pub fn parse() -> Arguments {
             backup_name: backup.unwrap().to_string(),
             full: args.is_present("full"),
             same_device: args.is_present("same_device"),
+            excludes: args.values_of("exclude").map(|v| v.map(|k| k.to_string()).collect()).unwrap_or_else(|| vec![]),
+            excludes_from: args.value_of("excludes_from").map(|v| v.to_string()),
             src_path: args.value_of("SRC").unwrap().to_string(),
             reference: args.value_of("reference").map(|v| v.to_string())
         }

src/cli/mod.rs

@@ -5,9 +5,13 @@ mod algotest;
 use ::prelude::*;
 
 use chrono::prelude::*;
+use regex::{self, RegexSet};
+
 use std::process::exit;
 use std::collections::HashMap;
 use std::fmt::Display;
+use std::io::{BufReader, BufRead};
+use std::fs::File;
 
 use self::args::Arguments;
 
@@ -174,7 +178,7 @@ pub fn run() {
             }
             print_config(&repo.config);
         },
-        Arguments::Backup{repo_path, backup_name, src_path, full, reference, same_device} => {
+        Arguments::Backup{repo_path, backup_name, src_path, full, reference, same_device, mut excludes, excludes_from} => {
             let mut repo = open_repository(&repo_path);
             let mut reference_backup = None;
             if !full {
@@ -188,8 +192,27 @@ pub fn run() {
                     info!("No reference backup found, doing a full scan instead");
                 }
             }
+            if let Some(excludes_from) = excludes_from {
+                for line in BufReader::new(checked(File::open(excludes_from), "open excludes file")).lines() {
+                    excludes.push(checked(line, "read excludes file"));
+                }
+            }
+            let excludes: Vec<String> = excludes.into_iter().map(|mut exclude| {
+                exclude = regex::escape(&exclude).replace('?', ".").replace(r"\*\", ".*").replace(r"\*", "[^/]*");
+                if exclude.starts_with('/') {
+                    format!(r"^{}($|/)", exclude)
+                } else {
+                    format!(r"/{}($|/)", exclude)
+                }
+            }).collect();
+            let excludes = if excludes.is_empty() {
+                None
+            } else {
+                Some(checked(RegexSet::new(excludes), "parse exclude patterns"))
+            };
             let options = BackupOptions {
-                same_device: same_device
+                same_device: same_device,
+                excludes: excludes
             };
             let backup = match repo.create_backup_recursively(&src_path, reference_backup.as_ref(), &options) {
                 Ok(backup) => backup,

src/main.rs

@@ -16,6 +16,7 @@ extern crate byteorder;
 extern crate sodiumoxide;
 extern crate ansi_term;
 extern crate filetime;
+extern crate regex;
 extern crate libc;
 
 pub mod util;

src/repository/backup.rs

@@ -6,6 +6,7 @@ use std::collections::{HashMap, BTreeMap, VecDeque};
 use std::os::linux::fs::MetadataExt;
 
 use chrono::prelude::*;
+use regex::RegexSet;
 
 
 quick_error!{
@@ -25,7 +26,8 @@ quick_error!{
 
 
 pub struct BackupOptions {
-    pub same_device: bool
+    pub same_device: bool,
+    pub excludes: Option<RegexSet>
 }
 
 
@@ -171,18 +173,24 @@ impl Repository {
             let parent_dev = try!(path.metadata()).st_dev();
             for ch in try!(fs::read_dir(path)) {
                 let child = try!(ch);
+                let child_path = child.path();
                 if options.same_device {
                     let child_dev = try!(child.metadata()).st_dev();
                     if child_dev != parent_dev {
                         continue
                     }
                 }
+                if let Some(ref excludes) = options.excludes {
+                    let child_path_str = child_path.to_string_lossy();
+                    if excludes.is_match(&child_path_str) {
+                        continue
+                    }
+                }
                 let name = child.file_name().to_string_lossy().to_string();
                 let ref_child = reference.as_ref()
                     .and_then(|inode| inode.children.as_ref())
                     .and_then(|map| map.get(&name))
                     .and_then(|chunks| self.get_inode(chunks).ok());
-                let child_path = child.path();
                 let chunks = match self.create_backup_recurse(&child_path, ref_child.as_ref(), options, backup, failed_paths) {
                     Ok(chunks) => chunks,
                     Err(_) => {