15 Commits
0.1.2 ... 0.1.3

Author SHA1 Message Date
sreedev
1a573fbf3f release github flow changes 2023-01-23 03:24:12 -05:00
sreedev
b12b662ba1 workflow updates 2023-01-23 03:19:57 -05:00
sreedev
96337b6b48 artifacts upload options added to github workflows 2023-01-23 02:39:37 -05:00
sreedev
02f70aa55f updated release github actions 2023-01-23 02:28:57 -05:00
Sreedev Kodichath
4a18b3fd91 Merge pull request #33 from sreedevk/development
Version 0.1.3
2023-01-23 02:20:59 -05:00
sreedev
876fdc69ac added version changes 2023-01-23 02:20:42 -05:00
sreedev
8acc3f8c8f added github workflow for building binaries with releases 2023-01-23 02:18:23 -05:00
Sreedev Kodichath
38f6c48452 Merge pull request #35 from sreedevk/dependabot/cargo/tokio-1.23.1
build(deps): bump tokio from 1.23.0 to 1.23.1
2023-01-21 19:31:49 -05:00
sreedev
3e1e9c0e2e Merge branch 'main' into development 2023-01-20 21:01:50 -05:00
Sreedev Kodichath
0674ab55bf Update README.md 2023-01-20 21:01:47 -05:00
dependabot[bot]
ccff95dfc5 build(deps): bump tokio from 1.23.0 to 1.23.1
Bumps [tokio](https://github.com/tokio-rs/tokio) from 1.23.0 to 1.23.1.
- [Release notes](https://github.com/tokio-rs/tokio/releases)
- [Commits](https://github.com/tokio-rs/tokio/compare/tokio-1.23.0...tokio-1.23.1)

---
updated-dependencies:
- dependency-name: tokio
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
2023-01-21 01:57:43 +00:00
Sreedev Kodichath
53d28abb62 Merge pull request #32 from sreedevk/performance/glob_improvements
Performance/glob improvements
2023-01-20 19:17:39 -05:00
Sreedev Kodichath
ea6cf94952 Update README.md 2023-01-19 23:37:13 -05:00
sreedev
7b275d40d0 min opts 2 2023-01-19 23:27:40 -05:00
sreedev
30398c3ca9 min opts 2023-01-19 23:25:16 -05:00
8 changed files with 85 additions and 89 deletions

41
.github/workflows/release.yml vendored Normal file
View File

@@ -0,0 +1,41 @@
name: Release
env:
PROJECT_NAME: deduplicator
PROJECT_DESC: "Filter, Sort & Delete Duplicate Files Recursively"
PROJECT_AUTH: "sreedevk"
on:
push:
tags:
- "[0-9]+.*"
jobs:
create-release:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: taiki-e/create-gh-release-action@v1.6.2
with:
changelog: CHANGELOG.md
token: ${{ secrets.GITHUB_TOKEN }}
branch: main
upload-assets:
strategy:
matrix:
os:
- ubuntu-latest
- macos-latest
- windows-latest
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v3
- uses: taiki-e/upload-rust-binary-action@v1
with:
bin: deduplicator
tar: unix
zip: windows
token: ${{ secrets.GITHUB_TOKEN }}
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

View File

@@ -1,20 +0,0 @@
name: Rust
on:
push:
branches: [ "main" ]
pull_request:
branches: [ "main" ]
env:
CARGO_TERM_COLOR: always
jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Build
run: cargo build --verbose
- name: Run tests
run: cargo test --verbose

6
Cargo.lock generated
View File

@@ -305,7 +305,7 @@ dependencies = [
[[package]]
name = "deduplicator"
version = "0.1.2"
version = "0.1.3"
dependencies = [
"anyhow",
"bytesize",
@@ -932,9 +932,9 @@ dependencies = [
[[package]]
name = "tokio"
version = "1.23.0"
version = "1.23.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "eab6d665857cc6ca78d6e80303a02cea7a7851e85dfbd77cbdc09bd129f1ef46"
checksum = "38a54aca0c15d014013256222ba0ebed095673f89345dd79119d912eb561b7a8"
dependencies = [
"autocfg",
"bytes",

View File

@@ -1,6 +1,6 @@
[package]
name = "deduplicator"
version = "0.1.2"
version = "0.1.3"
edition = "2021"
description = "find,filter,delete Duplicates"
license = "MIT"
@@ -23,5 +23,5 @@ memmap2 = "0.5.8"
prettytable-rs = "0.10.0"
rayon = "1.6.1"
thiserror = "1.0.38"
tokio = { version = "1.23.0", features = ["full"] }
tokio = { version = "1.23.1", features = ["full"] }
unicode-segmentation = "1.10.0"

View File

@@ -45,7 +45,4 @@ cargo install deduplicator
<h2 align="center">Screenshots</h2>
<img src="https://user-images.githubusercontent.com/36154121/211948081-63c12b94-6251-487b-a49f-ac5418169d5a.gif" />
<img src="https://user-images.githubusercontent.com/36154121/211458077-90092aa3-496c-492f-a061-618059890d5f.png" />
<img src="https://user-images.githubusercontent.com/36154121/213618143-e5182e39-731e-4817-87dd-1a6a0f38a449.gif" />

View File

@@ -40,10 +40,6 @@ fn modified_time(path: &String) -> Result<String> {
Ok(modified_time.format("%Y-%m-%d %H:%M:%S").to_string())
}
fn print_meta_info() {
println!("Deduplicator v{}", std::env!("CARGO_PKG_VERSION"));
}
fn scan_group_instruction() -> Result<String> {
println!("\nEnter the indices of the files you want to delete.");
println!("You can enter multiple files using commas to seperate file indices.");
@@ -115,8 +111,6 @@ fn process_group_action(duplicates: &Vec<File>, dup_index: usize, dup_size: usiz
}
pub fn interactive(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
print_meta_info();
if duplicates.is_empty() {
println!(
"\n{}",
@@ -128,8 +122,8 @@ pub fn interactive(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
duplicates
.clone()
.into_iter()
.sorted_unstable_by_key(|f| {
-(f.1.first().and_then(|ff| ff.size).unwrap_or_default() as i64)
.sorted_unstable_by_key(|(_, f)| {
-(f.first().and_then(|ff| ff.size).unwrap_or_default() as i64)
}) // sort by descending file size in interactive mode
.enumerate()
.for_each(|(gindex, (_, group))| {
@@ -140,7 +134,7 @@ pub fn interactive(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
itable.add_row(row![
index,
format_path(&file.path, opts).unwrap_or_default().blue(),
file_size(&file).unwrap_or_default().red(),
file_size(file).unwrap_or_default().red(),
modified_time(&file.path).unwrap_or_default().yellow()
]);
});
@@ -150,8 +144,6 @@ pub fn interactive(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
}
pub fn print(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
print_meta_info();
if duplicates.is_empty() {
println!(
"\n{}",
@@ -164,14 +156,14 @@ pub fn print(duplicates: DashMap<String, Vec<File>>, opts: &Params) {
output_table.set_titles(row!["hash", "duplicates"]);
duplicates
.into_iter()
.sorted_unstable_by_key(|f| f.1.first().and_then(|ff| ff.size).unwrap_or_default()) // sort by ascending size
.sorted_unstable_by_key(|(_, f)| f.first().and_then(|ff| ff.size).unwrap_or_default())
.for_each(|(hash, group)| {
let mut inner_table = Table::new();
inner_table.set_format(*format::consts::FORMAT_NO_BORDER_LINE_SEPARATOR);
group.iter().for_each(|file| {
inner_table.add_row(row![
format_path(&file.path, opts).unwrap_or_default().blue(),
file_size(&file).unwrap_or_default().red(),
file_size(file).unwrap_or_default().red(),
modified_time(&file.path).unwrap_or_default().yellow()
]);
});

View File

@@ -33,8 +33,8 @@ impl Params {
pub fn get_directory(&self) -> Result<String> {
let dir_pathbuf: PathBuf = self
.dir
.clone()
.unwrap_or(std::env::current_dir()?)
.as_ref()
.unwrap_or(&std::env::current_dir()?)
.as_os_str()
.into();
@@ -47,17 +47,18 @@ impl Params {
Ok(dir)
}
pub fn get_glob_patterns(&self) -> Vec<PathBuf> {
self.types
.clone()
.unwrap_or_else(|| String::from("*"))
.split(',')
.map(|filetype| format!("*.{}", filetype))
.map(|filetype| {
vec![self.get_directory().unwrap(), String::from("**"), filetype]
.iter()
.collect()
})
.collect()
pub fn get_glob_patterns(&self) -> PathBuf {
match self.types.as_ref() {
Some(filetypes) => vec![
self.get_directory().unwrap(),
String::from("**"),
format!("{{{}}}", filetypes),
]
.iter()
.collect::<PathBuf>(),
None => vec![self.get_directory().unwrap().as_str(), "**", "*"]
.iter()
.collect::<PathBuf>(),
}
}
}

View File

@@ -40,27 +40,22 @@ pub fn duplicates(app_opts: &Params) -> Result<DashMap<String, Vec<File>>> {
}
fn scan(app_opts: &Params) -> Result<Vec<File>> {
let glob_patterns: Vec<PathBuf> = app_opts.get_glob_patterns();
let files: Vec<File> = glob_patterns
.par_iter()
let glob_patterns = app_opts.get_glob_patterns().display().to_string();
let glob_iter = glob(&glob_patterns)?;
let files = glob_iter
.filter(Result::is_ok)
.map(|file| file.unwrap())
.filter(|fpath| fpath.is_file())
.collect::<Vec<PathBuf>>()
.into_par_iter()
.progress_with_style(ProgressStyle::with_template(
"{spinner:.green} [scanning files] [{wide_bar:.cyan/blue}] {pos}/{len} files",
"{spinner:.green} [processing scan results] [{wide_bar:.cyan/blue}] {pos}/{len} files",
)?)
.filter_map(|glob_pattern| glob(glob_pattern.as_os_str().to_str()?).ok())
.flat_map(|file_vec| {
file_vec
.filter_map(|x| Some(x.ok()?.as_os_str().to_str()?.to_string()))
.filter(|glob_result| {
fs::metadata(glob_result)
.map(|f| f.is_file())
.unwrap_or(false)
})
.collect::<Vec<String>>()
})
.map(|file_path| File {
path: file_path.clone(),
.map(|fpath| fpath.display().to_string())
.map(|fpath| File {
path: fpath.clone(),
hash: None,
size: Some(fs::metadata(file_path).unwrap().len()),
size: Some(fs::metadata(fpath).unwrap().len()),
})
.filter(|file| filters::is_file_gt_minsize(app_opts, file))
.collect();
@@ -68,16 +63,8 @@ fn scan(app_opts: &Params) -> Result<Vec<File>> {
Ok(files)
}
fn process_file_hash_index(file: &File) -> Result<File> {
Ok(File {
path: file.path.clone(),
size: file.size,
hash: Some(hash_file(&file.path).unwrap_or_default()),
})
}
fn process_file_index(
file: File,
mut file: File,
store: &DashMap<String, Vec<File>>,
index_criteria: IndexCritera,
) {
@@ -89,13 +76,11 @@ fn process_file_index(
.or_insert_with(|| vec![file]);
}
IndexCritera::Hash => {
let processed_file = process_file_hash_index(&file).unwrap();
let indexhash = processed_file.clone().hash.unwrap_or_default();
file.hash = Some(hash_file(&file.path).unwrap_or_default());
store
.entry(indexhash)
.and_modify(|fileset| fileset.push(processed_file.clone()))
.or_insert_with(|| vec![processed_file]);
.entry(file.clone().hash.unwrap())
.and_modify(|fileset| fileset.push(file.clone()))
.or_insert_with(|| vec![file]);
}
}
}