Restructure ExtractIntent

Allow the user to control the log level via the conventional `-v`
command line option. Multiplying this option-as per convention-increases
the verbosity.

This is an additional way to set the log level, in addition to the
existing:
- environment variable
- build type (debug vs release)

Given it is the most intentional and direct way to set the log level it
has precedence over all other options.

.drone.yml

@@ -6,6 +6,7 @@ steps:
 - name: build-shared
   image: arminfriedl/xwim-build:shared
   commands:
+    - meson wrap install gtest || true
     - meson target/shared
     - ninja -C target/shared
     - mv target/shared/src/xwim xwim-x86_64-glibc-linux-shared
@@ -13,6 +14,7 @@ steps:
 - name: build-static
   image: arminfriedl/xwim-build:static
   commands:
+    - meson wrap install gtest || true
     - meson --default-library=static target/static
     - ninja -C target/static
     - mv target/static/src/xwim xwim-x86_64-musl-linux-static
@@ -48,6 +50,7 @@ steps:
 - name: build-shared
   image: arminfriedl/xwim-build:shared
   commands:
+    - meson wrap install gtest || true
     - meson --buildtype=release target/shared
     - ninja -C target/shared
     - strip target/shared/src/xwim
@@ -57,6 +60,7 @@ steps:
 - name: build-static
   image: arminfriedl/xwim-build:static
   commands:
+    - meson wrap install gtest || true
     - meson --buildtype=release --default-library=static target/static
     - ninja -C target/static
     - strip target/static/src/xwim

.gitignore

@@ -2,11 +2,11 @@
 build/
 target/
 compile_commands.json
-.vscode
 .ccls-cache
+.idea/codeStyles/**
 
-# Created by https://www.gitignore.io/api/vim,c++,emacs,ninja
+# Created by https://www.toptal.com/developers/gitignore/api/c++,vim,emacs,linux,macos,ninja,windows,jetbrains+all,clion+all,visualstudiocode
-# Edit at https://www.gitignore.io/?templates=vim,c++,emacs,ninja
+# Edit at https://www.toptal.com/developers/gitignore?templates=c++,vim,emacs,linux,macos,ninja,windows,jetbrains+all,clion+all,visualstudiocode
 
 ### C++ ###
 # Prerequisites
@@ -42,6 +42,94 @@ compile_commands.json
 *.out
 *.app
 
+### CLion+all ###
+# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio, WebStorm and Rider
+# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839
+
+# User-specific stuff
+.idea/**/workspace.xml
+.idea/**/tasks.xml
+.idea/**/usage.statistics.xml
+.idea/**/dictionaries
+.idea/**/shelf
+
+# AWS User-specific
+.idea/**/aws.xml
+
+# Generated files
+.idea/**/contentModel.xml
+
+# Sensitive or high-churn files
+.idea/**/dataSources/
+.idea/**/dataSources.ids
+.idea/**/dataSources.local.xml
+.idea/**/sqlDataSources.xml
+.idea/**/dynamic.xml
+.idea/**/uiDesigner.xml
+.idea/**/dbnavigator.xml
+
+# Gradle
+.idea/**/gradle.xml
+.idea/**/libraries
+
+# Gradle and Maven with auto-import
+# When using Gradle or Maven with auto-import, you should exclude module files,
+# since they will be recreated, and may cause churn.  Uncomment if using
+# auto-import.
+# .idea/artifacts
+# .idea/compiler.xml
+# .idea/jarRepositories.xml
+# .idea/modules.xml
+# .idea/*.iml
+# .idea/modules
+# *.iml
+# *.ipr
+
+# CMake
+cmake-build-*/
+
+# Mongo Explorer plugin
+.idea/**/mongoSettings.xml
+
+# File-based project format
+*.iws
+
+# IntelliJ
+out/
+
+# mpeltonen/sbt-idea plugin
+.idea_modules/
+
+# JIRA plugin
+atlassian-ide-plugin.xml
+
+# Cursive Clojure plugin
+.idea/replstate.xml
+
+# SonarLint plugin
+.idea/sonarlint/
+
+# Crashlytics plugin (for Android Studio and IntelliJ)
+com_crashlytics_export_strings.xml
+crashlytics.properties
+crashlytics-build.properties
+fabric.properties
+
+# Editor-based Rest Client
+.idea/httpRequests
+
+# Android studio 3.1+ serialized cache file
+.idea/caches/build_file_checksums.ser
+
+### CLion+all Patch ###
+# Ignore everything but code style settings and run configurations
+# that are supposed to be shared within teams.
+
+.idea/*
+
+!.idea/codeStyles
+!.idea/runConfigurations
+
 ### Emacs ###
 # -*- mode: gitignore; -*-
 *~
@@ -93,6 +181,108 @@ flycheck_*.el
 /network-security.data
 
 
+### JetBrains+all ###
+# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio, WebStorm and Rider
+# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839
+
+# User-specific stuff
+
+# AWS User-specific
+
+# Generated files
+
+# Sensitive or high-churn files
+
+# Gradle
+
+# Gradle and Maven with auto-import
+# When using Gradle or Maven with auto-import, you should exclude module files,
+# since they will be recreated, and may cause churn.  Uncomment if using
+# auto-import.
+# .idea/artifacts
+# .idea/compiler.xml
+# .idea/jarRepositories.xml
+# .idea/modules.xml
+# .idea/*.iml
+# .idea/modules
+# *.iml
+# *.ipr
+
+# CMake
+
+# Mongo Explorer plugin
+
+# File-based project format
+
+# IntelliJ
+
+# mpeltonen/sbt-idea plugin
+
+# JIRA plugin
+
+# Cursive Clojure plugin
+
+# SonarLint plugin
+
+# Crashlytics plugin (for Android Studio and IntelliJ)
+
+# Editor-based Rest Client
+
+# Android studio 3.1+ serialized cache file
+
+### JetBrains+all Patch ###
+# Ignore everything but code style settings and run configurations
+# that are supposed to be shared within teams.
+
+
+
+### Linux ###
+
+# temporary files which can be created if a process still has a handle open of a deleted file
+.fuse_hidden*
+
+# KDE directory preferences
+.directory
+
+# Linux trash folder which might appear on any partition or disk
+.Trash-*
+
+# .nfs files are created when an open file is removed but is still being accessed
+.nfs*
+
+### macOS ###
+# General
+.DS_Store
+.AppleDouble
+.LSOverride
+
+# Icon must end with two \r
+Icon
+
+
+# Thumbnails
+._*
+
+# Files that might appear in the root of a volume
+.DocumentRevisions-V100
+.fseventsd
+.Spotlight-V100
+.TemporaryItems
+.Trashes
+.VolumeIcon.icns
+.com.apple.timemachine.donotpresent
+
+# Directories potentially created on remote AFP share
+.AppleDB
+.AppleDesktop
+Network Trash Folder
+Temporary Items
+.apdisk
+
+### macOS Patch ###
+# iCloud generated files
+*.icloud
+
 ### Ninja ###
 .ninja_deps
 .ninja_log
@@ -100,6 +290,7 @@ flycheck_*.el
 ### Vim ###
 # Swap
 [._]*.s[a-v][a-z]
+!*.svg  # comment out if you don't need vector files
 [._]*.sw[a-p]
 [._]s[a-rt-v][a-z]
 [._]ss[a-gi-z]
@@ -111,14 +302,54 @@ Sessionx.vim
 
 # Temporary
 .netrwhist
-
 # Auto-generated tag files
 tags
-
 # Persistent undo
 [._]*.un~
 
-# Coc configuration directory
+### VisualStudioCode ###
-.vim
+.vscode/*
+!.vscode/settings.json
+!.vscode/tasks.json
+!.vscode/launch.json
+!.vscode/extensions.json
+!.vscode/*.code-snippets
 
-# End of https://www.gitignore.io/api/vim,c++,emacs,ninja
+# Local History for Visual Studio Code
+.history/
+
+# Built Visual Studio Code Extensions
+*.vsix
+
+### VisualStudioCode Patch ###
+# Ignore all local history of files
+.history
+.ionide
+
+### Windows ###
+# Windows thumbnail cache files
+Thumbs.db
+Thumbs.db:encryptable
+ehthumbs.db
+ehthumbs_vista.db
+
+# Dump file
+*.stackdump
+
+# Folder config file
+[Dd]esktop.ini
+
+# Recycle Bin used on file shares
+$RECYCLE.BIN/
+
+# Windows Installer files
+*.cab
+*.msi
+*.msix
+*.msm
+*.msp
+
+# Windows shortcuts
+*.lnk
+
+# End of https://www.toptal.com/developers/gitignore/api/c++,vim,emacs,linux,macos,ninja,windows,jetbrains+all,clion+all,visualstudiocode

README.md

@@ -22,6 +22,22 @@ archiving binaries (if installed). In contrast `xwim` is a compiled binary based
 directly on archiving libraries, which some may appreciate. It can optionally be
 statically linked if you want it entirely self-contained.
 
+# Install
+`xwim` currently released for Linux only. There are two flavers: statically
+linked and dynamically linked. The releases can be downloaded from
+https://git.friedl.net/incubator/xwim/releases and should run on most 64-bit
+GNU/Linux distributions. 
+
+For the dynamically linked version, the following dependencies have to be
+installed:
+- [spdlog](https://github.com/gabime/spdlog)
+- [fmt](https://github.com/fmtlib/fmt)
+- [libarchive](https://github.com/libarchive/libarchive)
+
+Windows support is planned for the first stable release. Packaging for various
+distributions is also planned once `xwim` stabilizes. Please reach out if you
+can help.
+
 # Usage
 Invoking `xwim` is as simple as:
 
@@ -90,22 +106,6 @@ Take a look `Archiver.hpp` if you want to help and have some time for testing.
 Most formats can readily be added if they are supported by libarchive. For other
 formats you have to add an `Archiver` implementation.
 
-# Install
-`xwim` currently released for Linux only. There are two flavers: statically
-linked and dynamically linked. The releases can be downloaded from
-https://git.friedl.net/incubator/xwim/releases and should run on most 64-bit
-GNU/Linux distributions. 
-
-For the dynamically linked version, the following dependencies have to be
-installed:
-- [spdlog](https://github.com/gabime/spdlog)
-- [fmt](https://github.com/fmtlib/fmt)
-- [libarchive](https://github.com/libarchive/libarchive)
-
-Windows support is planned for the first stable release. Packaging for various
-distributions is also planned once `xwim` stabilizes. Please reach out if you
-can help.
-
 # Build
 xwim is built with [meson](https://mesonbuild.com/). To compile xwim from source
 you need:

meson.build

@@ -5,7 +5,9 @@ project('xwim', 'cpp',
                           'b_ndebug=if-release'])
 
 add_global_arguments('-DVERSION='+meson.version(), language: 'cpp')
+add_global_arguments('-DSPDLOG_FMT_EXTERNAL', language: 'cpp')
+add_global_arguments('-DFMT_HEADER_ONLY', language: 'cpp')
 
 subdir('src')
 subdir('doc')
-# subdir('test')
+subdir('test')

src/Archiver.cpp

@@ -1,13 +1,21 @@
 #include "Archiver.hpp"
+#include "Formats.hpp"
 
-#include <spdlog/common.h>
 #include <spdlog/spdlog.h>
 
 #include <filesystem>
 #include <map>
 #include <memory>
 
-#include "Common.hpp"
+#include "util/Common.hpp"
+
+#if defined(unix) || defined(__unix__) || defined(__unix)
+std::string default_extension = ".tar.gz";
+#elif defined(_win32) || defined(__win32__) || defined(__windows__)
+std::string default_extension = ".zip";
+#else
+    std::string default_extension = ".zip";
+#endif
 
 namespace xwim {
 using namespace std;
@@ -30,9 +38,9 @@ fs::path archive_extension(const fs::path& path) {
 
   while (tmp_path.has_extension()) {
     tmp_ext = tmp_path.extension() += tmp_ext;
-    auto search = extensions_format.find(tmp_ext);
+    Format format = find_extension_format(tmp_ext);
 
-    if (search != extensions_format.end()) {
+    if (format != Format::UNKNOWN) {
       // (Combined) extension known. Remember as `ext` and keep
       // looking for even longer extensions.
       ext = tmp_ext;
@@ -69,9 +77,9 @@ fs::path strip_archive_extension(const fs::path& path) {
     tmp_ext = tmp_path.extension() += tmp_ext;
     spdlog::debug("Looking for {} in known extensions", tmp_ext);
 
-    auto search = extensions_format.find(tmp_ext);
+    Format format = find_extension_format(tmp_ext);
     tmp_longest_ext++;
-    if (search != extensions_format.end()) {
+    if (format != Format::UNKNOWN) {
       // (Combined) extension known. Remember as `longest_ext` and keep
       // looking for even longer extensions.
       longest_ext = tmp_longest_ext;
@@ -90,7 +98,14 @@ fs::path strip_archive_extension(const fs::path& path) {
   return tmp_path;
 }
 
-bool can_extract(const fs::path& path) {
+std::filesystem::path default_archive(const std::filesystem::path& base) {
+    string base_s = base.string();
+    string ext_s = default_extension;
+
+    return fs::path{fmt::format("{}{}", base_s, ext_s)};
+}
+
+bool can_handle_archive(const fs::path& path) {
   fs::path ext = archive_extension(path);
   if (format_extensions.find(ext.string()) != format_extensions.end()) {
     spdlog::debug("Found {} in known formats", ext);
@@ -105,17 +120,20 @@ Format parse_format(const fs::path& path) {
   spdlog::debug("Looking for path {}", path);
   fs::path ext = archive_extension(path);
   spdlog::debug("Looking for ext {}", ext);
-  auto search = extensions_format.find(ext);
+  Format format = find_extension_format(ext);
-  if (search == extensions_format.end()) {
+
+  if (format == Format::UNKNOWN) {
     throw XwimError{"No known archiver for {}", path};
   }
 
-  return search->second;
+  return format;
 }
 
 unique_ptr<Archiver> make_archiver(const string& archive_name) {
   switch (parse_format(archive_name)) {
-    case Format::TAR_GZ:
+      case Format::TAR_GZIP:      case Format::TAR_BZIP2:
+      case Format::TAR_COMPRESS:  case Format::TAR_LZIP:
+      case Format::TAR_XZ:        case Format::TAR_ZSTD:
       case Format::ZIP:
           return make_unique<LibArchiver>();
     default:
@@ -125,4 +143,5 @@ unique_ptr<Archiver> make_archiver(const string& archive_name) {
           archive_name};
   };
 }
+
 }  // namespace xwim

src/Archiver.hpp

@@ -7,18 +7,11 @@
 #include <memory>
 #include <set>
 
-#include "Common.hpp"
+#include "util/Common.hpp"
+#include "Formats.hpp"
 
 namespace xwim {
 
-// Invariant:
-// `extensions_format` defines a surjection from `format_extensions`
-// to `Formats`
-const std::set<std::string> format_extensions{".tar.gz", ".zip"};
-enum class Format { TAR_GZ, ZIP };
-const std::map<std::string, Format> extensions_format{
-    {".tar.gz", Format::TAR_GZ}, {".zip", Format::ZIP}};
-
 class Archiver {
  public:
   virtual void compress(std::set<std::filesystem::path> ins,
@@ -40,8 +33,10 @@ class LibArchiver : public Archiver {
 
 std::filesystem::path archive_extension(const std::filesystem::path& path);
 std::filesystem::path strip_archive_extension(const std::filesystem::path& path);
+std::filesystem::path default_archive(const std::filesystem::path& base);
+
 Format parse_format(const std::filesystem::path& path);
-bool can_extract(const std::filesystem::path& path);
+bool can_handle_archive(const std::filesystem::path& path);
 
 std::unique_ptr<Archiver> make_archiver(const std::string& archive_name);
 

src/Formats.hpp

@@ -0,0 +1,49 @@
+#pragma once
+
+namespace xwim {
+    using namespace std;
+
+    // Invariant:
+    // `extensions_format` defines a surjection from `format_extensions`
+    // to `Formats`
+    enum class Format {
+        UNKNOWN,
+        TAR_BZIP2, TAR_GZIP, TAR_LZIP, TAR_XZ, TAR_COMPRESS, TAR_ZSTD,
+        ZIP
+    };
+
+    const set<string> format_extensions{
+            // tar formats see: https://en.wikipedia.org/wiki/Tar_(computing)#Suffixes_for_compressed_files
+
+            /* bzip2 */    ".tar.bz2", ".tb2", ".tbz", ".tbz2", ".tz2",
+            /* gzip */     ".tar.gz", ".taz", ".tgz",
+            /* lzip */     ".tar.lz",
+            /* xz */       ".tar.xz", ".txz",
+            /* compress */ ".tar.Z", ".tZ", ".taZ",
+            /* zstd */     ".tar.zst", ".tzst",
+
+            /* zip */      ".zip"
+    };
+
+    const map<set<string>, Format> extensions_format{
+        {{".tar.bz2", ".tb2", ".tbz", ".tbz2", ".tz2"}, Format::TAR_BZIP2},
+        {{".tar.gz", ".taz", ".tgz"}, Format::TAR_GZIP},
+        {{".tar.lz"}, Format::TAR_LZIP},
+        {{".tar.xz", ".txz"}, Format::TAR_XZ},
+        {{".tar.Z", ".tZ", ".taZ"}, Format::TAR_COMPRESS},
+        {{".tar.zst", ".tzst"}, Format::TAR_ZSTD},
+
+        {{".zip"}, Format::ZIP}
+    };
+
+    inline Format find_extension_format(const string& ext) {
+        for(auto ef: extensions_format) {
+            auto f = ef.first.find(ext);
+            if(f != ef.first.end()) {
+                return ef.second;
+            }
+        }
+
+        return Format::UNKNOWN;
+    }
+}

src/UserIntent.cpp

@@ -0,0 +1,226 @@
+#include "UserIntent.hpp"
+
+#include <spdlog/spdlog.h>
+
+#include <algorithm>
+#include <filesystem>
+
+#include "Archiver.hpp"
+
+namespace xwim {
+unique_ptr<UserIntent> make_compress_intent(const UserOpt &userOpt) {
+  if (userOpt.paths.size() == 1) {
+    return make_unique<CompressSingleIntent>(
+        CompressSingleIntent{*userOpt.paths.begin(), userOpt.out});
+  }
+
+  if (!userOpt.out.has_value()) {
+    throw XwimError("Cannot guess output for multiple targets");
+  }
+
+  return make_unique<CompressManyIntent>(
+      CompressManyIntent{userOpt.paths, userOpt.out.value()});
+}
+
+unique_ptr<UserIntent> make_extract_intent(const UserOpt &userOpt) {
+  for (const path &p : userOpt.paths) {
+    if (!can_handle_archive(p)) {
+      throw XwimError("Cannot extract path {}", p);
+    }
+  }
+
+  return make_unique<ExtractIntent>(ExtractIntent{userOpt.paths, userOpt.out});
+}
+
+unique_ptr<UserIntent> try_infer_compress_intent(const UserOpt &userOpt) {
+  if (!userOpt.out.has_value()) {
+    spdlog::debug("No <out> provided");
+    if (userOpt.paths.size() != 1) {
+      spdlog::debug(
+          "Not a single-path compression. Cannot guess <out> for many-path "
+          "compression");
+      return nullptr;
+    }
+
+    spdlog::debug("Only one <path> provided. Assume single-path compression.");
+    return make_unique<CompressSingleIntent>(
+        CompressSingleIntent{*userOpt.paths.begin(), userOpt.out});
+  }
+
+  spdlog::debug("<out> provided: {}", userOpt.out.value());
+  if (can_handle_archive(userOpt.out.value())) {
+    spdlog::debug("{} given and a known archive format, assume compression",
+                  userOpt.out.value());
+    return make_compress_intent(userOpt);
+  }
+
+  spdlog::debug(
+      "Cannot compress multiple paths without a user-provided output archive");
+  return nullptr;
+}
+
+unique_ptr<UserIntent> try_infer_extract_intent(const UserOpt &userOpt) {
+  bool can_extract_all =
+      std::all_of(userOpt.paths.begin(), userOpt.paths.end(),
+                  [](const path &path) { return can_handle_archive(path); });
+
+  if (!can_extract_all) {
+    spdlog::debug(
+        "Cannot extract all provided <paths>. Assume this is not an "
+        "extraction.");
+    for (const path &p : userOpt.paths) {
+      if (!can_handle_archive(p)) {
+        spdlog::debug("Cannot handle {}", p);
+      }
+    }
+
+    return nullptr;
+  }
+
+  if (userOpt.out.has_value() && can_handle_archive(userOpt.out.value())) {
+    spdlog::debug(
+        "Could extract all provided <paths>. But also {} looks like an "
+        "archive. Ambiguous intent. Assume this is not an extraction.",
+        userOpt.out.value());
+    return nullptr;
+  }
+
+  spdlog::debug(
+      "Could extract all provided <paths>. But also <out> looks like an "
+      "archive. Ambiguous intent. Assume this is not an extraction.");
+  return make_extract_intent(userOpt);
+}
+
+unique_ptr<UserIntent> make_intent(const UserOpt &userOpt) {
+  if (userOpt.wants_compress() && userOpt.wants_extract()) {
+    throw XwimError("Cannot compress and extract simultaneously");
+  }
+  if (userOpt.paths.empty()) {
+    throw XwimError("No input given...");
+  }
+
+  // explicitly specified intent
+  if (userOpt.wants_compress()) return make_compress_intent(userOpt);
+  if (userOpt.wants_extract()) return make_extract_intent(userOpt);
+
+  spdlog::info("Intent not explicitly provided, trying to infer intent");
+
+  if (auto intent = try_infer_extract_intent(userOpt)) {
+    spdlog::info("Extraction intent inferred");
+    return intent;
+  }
+  spdlog::info("Cannot infer extraction intent");
+
+  if (auto intent = try_infer_compress_intent(userOpt)) {
+    spdlog::info("Compression intent inferred");
+    return intent;
+  }
+  spdlog::info("Cannot infer compression intent");
+
+  throw XwimError("Cannot guess intent");
+}
+
+void ExtractIntent::dwim_reparent(const path &out) {
+  // move extraction if extraction resulted in only one entry and that entries
+  // name is already the stripped archive name, i.e. reduce unnecessary nesting
+  auto dit = std::filesystem::directory_iterator(out);
+  auto dit_path = dit->path();
+
+  if (dit == std::filesystem::directory_iterator()) {
+    spdlog::debug(
+        "Cannot flatten extraction folder: extraction folder is empty");
+    return;
+  }
+
+  if (!is_directory(dit_path)) {
+    spdlog::debug("Cannot flatten extraction folder: {} is not a directory",
+                  dit_path);
+    return;
+  }
+
+  if (next(dit) != std::filesystem::directory_iterator()) {
+    spdlog::debug("Cannot flatten extraction folder: multiple items extracted");
+    return;
+  }
+
+  if (!std::filesystem::equivalent(dit_path.filename(), out.filename())) {
+    spdlog::debug(
+        "Cannot flatten extraction folder: archive entry differs from archive "
+        "name [extraction folder: {}, archive entry: {}]",
+        out.filename(), dit_path.filename());
+    return;
+  }
+
+  spdlog::debug("Output folder [{}] is equivalent to archive entry [{}]", out,
+                dit_path);
+  spdlog::info("Flattening extraction folder");
+
+  int i = rand_int(0, 100000);
+  path tmp_out = path{out};
+  tmp_out.concat(fmt::format(".xwim{}", i));
+  spdlog::debug("Move {} to {}", dit_path, tmp_out);
+  std::filesystem::rename(dit_path, tmp_out);
+  spdlog::debug("Remove parent path {}", out);
+  std::filesystem::remove(out);
+  spdlog::debug("Moving {} to {}", tmp_out, out);
+  std::filesystem::rename(tmp_out, out);
+}
+
+path ExtractIntent::out_path(const path &p) {
+  if (!this->out.has_value()) {
+    // not out path given, create from archive name
+    path out = std::filesystem::current_path() / strip_archive_extension(p);
+    create_directories(out);
+    return out;
+  }
+
+  if (this->archives.size() == 1) {
+    // out given and only one archive to extract, just extract into `out`
+    create_directories(this->out.value());
+    return this->out.value();
+  }
+
+  // out given and multiple archives to extract, create subfolder
+  // for each archive
+  create_directories(this->out.value());
+  path out = this->out.value() / strip_archive_extension(p);
+  return out;
+}
+
+void ExtractIntent::execute() {
+  for (const path &p : this->archives) {
+    std::unique_ptr<Archiver> archiver = make_archiver(p);
+    path out = this->out_path(p);
+    archiver->extract(p, out);
+    this->dwim_reparent(out);
+  }
+}
+
+path CompressSingleIntent::out_path() {
+  if (this->out.has_value()) {
+    if (!can_handle_archive(this->out.value())) {
+      throw XwimError("Unknown archive format {}", this->out.value());
+    }
+
+    return this->out.value();
+  }
+
+  return default_archive(strip_archive_extension(this->in).stem());
+}
+
+void CompressSingleIntent::execute() {
+  path out = this->out_path();
+  unique_ptr<Archiver> archiver = make_archiver(out);
+  set<path> ins{this->in};
+  archiver->compress(ins, out);
+};
+
+void CompressManyIntent::execute() {
+  if (!can_handle_archive(this->out)) {
+    throw XwimError("Unknown archive format {}", this->out);
+  }
+
+  unique_ptr<Archiver> archiver = make_archiver(this->out);
+  archiver->compress(this->in_paths, this->out);
+}
+}  // namespace xwim

src/UserIntent.hpp

@@ -0,0 +1,93 @@
+#pragma once
+
+#include <optional>
+#include <set>
+
+#include "util/Common.hpp"
+#include "UserOpt.hpp"
+
+namespace xwim {
+using namespace std;
+using std::filesystem::path;
+
+class UserIntent {
+public:
+    virtual void execute() = 0;
+    virtual ~UserIntent() = default;
+};
+
+/* Factory method to construct a UserIntent which implements `execute()` */
+unique_ptr<UserIntent> make_intent(const UserOpt& userOpt);
+
+/**
+* Extraction intent
+*
+* Extracts one or multiple archives. Optionally extracts them to given `out` folder. Otherwise extracts them to the
+* current working directory.
+*/
+class ExtractIntent: public UserIntent {
+private:
+    set<path> archives;
+    optional<path> out;
+
+    void dwim_reparent(const path& out);
+    path out_path(const path& p);
+
+   public:
+    ExtractIntent(set<path> archives, optional<path> out): archives(archives), out(out) {};
+    ~ExtractIntent() override = default;
+
+    void execute() override;
+};
+
+/**
+* Compress intent for a single file or folder.
+*
+* Compresses a single path which may be a file or a folder.
+*
+* No `out` path given:
+* - derives the archive name from the input path
+* - uses the default archive format for the platform
+*
+* `out` path given:
+* - `out` path must be a path with a valid archive name (including extension)
+* - tries to compress the input to the out archive
+* - if the `out` base name is different from the input base name, puts the input into a new folder
+*   with base name inside the archive (archive base name is always the name of the archive content)
+*/
+class CompressSingleIntent : public UserIntent {
+private:
+    path in;
+    optional<path> out;
+
+    path out_path();
+
+public:
+    CompressSingleIntent(path in, optional<path> out) : UserIntent(), in(in), out(out) {};
+    ~CompressSingleIntent() override = default;
+
+    void execute() override;
+};
+
+/**
+ * Compress intent for multiple files and/or folders.
+ *
+ * Compresses multiple files and/or folders to a single archive as given by the `out` path. Since `out` cannot be
+ * guessed from the input in this case it is mandatory.
+ *
+ * A new, single root folder with base name equal to base name of the `out` archive is created inside the archive. All
+ * input files are put into this root folder.
+ */
+class CompressManyIntent: public UserIntent {
+private:
+    set<path> in_paths;
+    path out;
+
+public:
+    CompressManyIntent(set<path> in_paths, path out): UserIntent(), in_paths(in_paths), out(out) {};
+    ~CompressManyIntent() override = default;
+
+    void execute() override;
+};
+
+}  // namespace xwim

src/UserOpt.cpp

@@ -0,0 +1,52 @@
+#include "UserOpt.hpp"
+
+#include <tclap/CmdLine.h>
+
+template <>
+struct TCLAP::ArgTraits<std::filesystem::path> {
+  // We use `operator=` here for path construction
+  // because `operator>>` (`ValueLike`) causes a split at
+  // whitespace
+  typedef StringLike ValueCategory;
+};
+
+namespace xwim {
+UserOpt::UserOpt(int argc, char** argv) {
+  // clang-format off
+  TCLAP::CmdLine cmd
+    {"xwim - Do What I Mean Extractor", ' ', "0.3.0"};
+
+  TCLAP::SwitchArg arg_compress
+    {"c", "compress", "Compress <files>", cmd, false};
+
+  TCLAP::SwitchArg arg_extract
+    {"x", "extract", "Extract <file>", cmd, false};
+
+  TCLAP::SwitchArg arg_noninteractive
+    {"i", "non-interactive", "Non-interactive, fail on ambiguity", cmd, false};
+
+  TCLAP::ValueArg<fs::path> arg_outfile
+    {"o", "out", "Out <file-or-path>", false, fs::path{}, "A path on the filesystem", cmd};
+
+  TCLAP::MultiSwitchArg arg_verbose
+    {"v", "verbose", "Verbosity level", cmd, 0};
+
+  TCLAP::UnlabeledMultiArg<fs::path> arg_paths
+    {"files", "Archive(s) to extract or file(s) to compress", true, "A path on the filesystem", cmd};
+  // clang-format on
+
+  cmd.parse(argc, argv);
+
+  if (arg_compress.isSet()) this->compress = arg_compress.getValue();
+  if (arg_extract.isSet()) this->extract = arg_extract.getValue();
+  if (arg_outfile.isSet()) this->out = arg_outfile.getValue();
+
+  this->verbosity = arg_verbose.getValue();
+  this->interactive = !arg_noninteractive.getValue();
+
+  if (arg_paths.isSet()) {
+    this->paths =
+        set<fs::path>{arg_paths.getValue().begin(), arg_paths.getValue().end()};
+  }
+}
+}  // namespace xwim

src/UserOpt.hpp

@@ -0,0 +1,31 @@
+#pragma once
+
+#include <optional>
+#include <set>
+
+#include "util/Common.hpp"
+
+namespace xwim {
+using namespace std;
+namespace fs = std::filesystem;
+
+struct UserOpt {
+  optional<bool> compress;
+  optional<bool> extract;
+  bool interactive;
+  int verbosity;
+  std::optional<fs::path> out;
+  std::set<fs::path> paths;
+
+  UserOpt(int argc, char** argv);
+
+  bool wants_compress() const {
+      return this->compress.has_value() && this->compress.value();
+  }
+
+  bool wants_extract() const {
+      return this->extract.has_value() && this->extract.value();
+  }
+};
+
+}  // namespace xwim

src/Xwim.cpp

@@ -1,165 +0,0 @@
-#include "Xwim.hpp"
-
-#include <spdlog/spdlog.h>
-
-#include <algorithm>
-#include <cstdlib>
-#include <filesystem>
-#include <ios>
-#include <iostream>
-#include <iterator>
-#include <random>
-#include <string>
-
-#include "Archiver.hpp"
-#include "Common.hpp"
-
-namespace xwim {
-using namespace std;
-namespace fs = std::filesystem;
-
-#if defined(unix) || defined(__unix__) || defined(__unix)
-std::string default_extension = ".tar.gz";
-#elif defined(_win32) || defined(__win32__) || defined(__windows__)
-std::string default_extension = ".zip";
-#else
-std::string default_extension = ".zip";
-#endif
-
-Xwim::Xwim() : action{Action::UNKNOWN} {}
-
-void Xwim::try_infer() {
-  infer_action();
-  infer_output();
-
-  if (action == Action::COMPRESS) {
-    archiver = make_archiver(out.string());
-  } else if (action == Action::EXTRACT) {
-    // we can only handle one archive for extraction at a time.
-    // Checked in `infer_extraction_output`
-    archiver = make_archiver(ins.begin()->string());
-  }
-}
-void Xwim::dwim() {
-  switch (action) {
-    case Action::COMPRESS:
-      this->archiver->compress(ins, out);
-      break;
-    case Action::EXTRACT:
-      this->archiver->extract(*ins.begin(), out);
-      sanitize_output();
-      break;
-    default:
-      spdlog::error("Unknown action");
-  }
-}
-
-void Xwim::sanitize_output() {
-  fs::path in_stripped = xwim::strip_archive_extension(*ins.begin());
-
-  int count = 0;
-  fs::directory_entry first_entry;
-  for(auto& e: fs::directory_iterator(out)) {
-    count++;
-    if(first_entry.path().empty()) {
-      first_entry = e;
-    }
-  }
-
-  if (count >= 2) {
-    spdlog::debug("Found multiple entries in extraction directory. Moving {} to {}", out, in_stripped);
-    fs::rename(out, in_stripped);
-  } else {
-    if(first_entry.is_directory()) {
-      spdlog::debug("Found single directory in extraction directory. Moving {} to {}",
-          first_entry.path(), in_stripped);
-      fs::rename(first_entry, in_stripped);
-      fs::remove(out);
-    } else {
-      spdlog::debug(
-          "Found single file in extraction directory. Moving {} to {}", out, in_stripped);
-      fs::rename(out, in_stripped);
-    }
-  }
-}
-
-void Xwim::infer_action() {
-  if (action != Action::UNKNOWN) return;
-
-  if (ins.size() == 1 && can_extract(*ins.begin())) {
-    action = Action::EXTRACT;
-  } else {
-    action = Action::COMPRESS;
-  }
-  spdlog::debug("Inferred action: {}", action);
-}
-
-void Xwim::infer_output() {
-  if (!out.empty()) return;
-
-  switch (action) {
-    case Action::COMPRESS:
-      infer_compression_output();
-      break;
-    case Action::EXTRACT:
-      infer_extraction_output();
-      break;
-    default:
-      throw XwimError{"Cannot infer output, action is unknown"};
-  }
-
-  spdlog::debug("Inferred out: {}", out.string());
-}
-
-void Xwim::infer_compression_output() {
-  if (ins.size() == 1) {
-    // archive name is just the name of the input with default archive
-    // extension
-    fs::path archive_stem = xwim::strip_archive_extension(*ins.begin());
-    archive_stem += default_extension;
-    out = archive_stem;
-  } else {
-    // We cannot guess the name of the output archive
-
-    // TODO use readline/lineoise/editline for path completion
-    cout << "Archive name: ";
-    cin >> out;
-    out = fs::path(out);
-  }
-}
-void Xwim::infer_extraction_output() {
-  if (ins.size() > 1) {
-    throw XwimError{"Cannot extract more than one archive at a time"};
-  }
-
-  // create a temporary path for extraction
-  fs::path archive_stem = xwim::strip_archive_extension(*ins.begin());
-
-  // note: we use here what is considered an `extensions` by `fs::path` so that
-  // we can strip it again easily later on
-  archive_stem += ".";
-  archive_stem += to_string(rand_int(999, 99999));
-  archive_stem += ".tmp";
-  this->out = archive_stem;
-}
-
-void Xwim::setCompress() {
-  this->action = Action::COMPRESS;
-  spdlog::debug("Set action to {}", this->action);
-}
-void Xwim::setExtract() {
-  this->action = Action::EXTRACT;
-  spdlog::debug("Set action to {}", this->action);
-}
-void Xwim::setOut(fs::path path) {
-  this->out = path;
-  spdlog::debug("Set out to {}", this->out);
-}
-void Xwim::setIns(vector<fs::path> ins) {
-  this->ins.insert(ins.begin(), ins.end());
-  if (this->ins.size() != ins.size()) {
-    spdlog::warn("Duplicate input files found. Removed {} duplicate(s).",
-                 (ins.size() - this->ins.size()));
-  }
-}
-}  // namespace xwim

src/Xwim.hpp

@@ -1,63 +0,0 @@
-#pragma once
-
-#include <fmt/core.h>
-#include <fmt/format.h>
-
-#include <exception>
-#include <memory>
-#include <set>
-#include <stdexcept>
-
-#include "Common.hpp"
-#include "Archiver.hpp"
-
-namespace xwim {
-using namespace std;
-namespace fs = std::filesystem;
-
-enum class Action { UNKNOWN, EXTRACT, COMPRESS };
-
-class Xwim {
- private:
-  Action action;
-  fs::path out;
-  set<fs::path> ins;
-  unique_ptr<Archiver> archiver;
-
-  void infer_action();
-  void infer_output();
-  void infer_compression_output();
-  void infer_extraction_output();
-  void sanitize_output();
-
- public:
-  Xwim();
-
-  void try_infer();
-  void dwim();
-
-  void setCompress();
-  void setExtract();
-  void setOut(fs::path);
-  void setIns(vector<fs::path> ins);
-};
-
-}  // namespace xwim
-
-template <>
-struct fmt::formatter<xwim::Action> {
-  constexpr auto parse(format_parse_context& ctx) { return ctx.begin(); }
-
-  template <typename FormatContext>
-  auto format(const xwim::Action& action, FormatContext& ctx) {
-    switch (action) {
-      case xwim::Action::UNKNOWN:
-        return format_to(ctx.out(), "UNKNOWN");
-      case xwim::Action::EXTRACT:
-        return format_to(ctx.out(), "EXTRACT");
-      case xwim::Action::COMPRESS:
-        return format_to(ctx.out(), "COMPRESS");
-    };
-    return format_to(ctx.out(), "");
-  }
-};

src/archiver/LibArchiver.cpp

@@ -9,8 +9,8 @@
 #include <iostream>
 #include <memory>
 
-#include "Archiver.hpp"
+#include "../Archiver.hpp"
-#include "Common.hpp"
+#include "../util/Common.hpp"
 
 namespace xwim {
 using namespace std;
@@ -27,8 +27,9 @@ void LibArchiver::compress(set<fs::path> ins, fs::path archive_out) {
   // complete type. `archive` is forward declared only.
   shared_ptr<archive> writer;
   writer = shared_ptr<archive>(archive_write_new(), archive_write_free);
-  archive_write_add_filter_gzip(writer.get());
+//  archive_write_add_filter_gzip(writer.get());
-  archive_write_set_format_pax_restricted(writer.get());
+//  archive_write_set_format_pax_restricted(writer.get());
+  archive_write_set_format_filter_by_ext(writer.get(), archive_out.c_str());
   archive_write_open_filename(writer.get(), archive_out.c_str());
 
   shared_ptr<archive> reader;

src/main.cpp

@@ -1,74 +1,25 @@
 #include <spdlog/common.h>
-#include <spdlog/logger.h>
 #include <spdlog/spdlog.h>
-#include <tclap/ArgException.h>
-#include <tclap/CmdLine.h>
-#include <tclap/StdOutput.h>
-#include <tclap/SwitchArg.h>
-#include <tclap/UnlabeledMultiArg.h>
-#include <tclap/ValueArg.h>
 
 #include <cstdlib>
 #include <filesystem>
 
-#include "Common.hpp"
+#include "UserIntent.hpp"
-#include "Log.hpp"
+#include "UserOpt.hpp"
-#include "Xwim.hpp"
+#include "util/Common.hpp"
+#include "util/Log.hpp"
 
 using namespace xwim;
 using namespace std;
-namespace fs = std::filesystem;
-
-template <>
-struct TCLAP::ArgTraits<std::filesystem::path> {
-  typedef ValueLike ValueCategory;
-};
 
 int main(int argc, char** argv) {
   log::init();
-
+  UserOpt user_opt = UserOpt{argc, argv};
-  TCLAP::CmdLine cmd{"xwim - Do What I Mean Extractor", ' ', "0.3.0"};
+  log::init(user_opt.verbosity);
-
-  TCLAP::SwitchArg arg_compress{"c", "compress", "Compress <files>", cmd,
-                                false};
-  TCLAP::SwitchArg arg_extract{"x", "extract", "Extract <file>", cmd, false};
-
-  TCLAP::ValueArg<fs::path> arg_outfile{
-      "o",   "out",      "Out <file-or-path>",
-      false, fs::path{}, "A path on the filesystem",
-      cmd};
-  TCLAP::UnlabeledMultiArg<fs::path> arg_infiles{
-      "Files", "Archive to extract or files to compress", true,
-      "A path on the filesystem", cmd};
-
-  Xwim xwim;
-
-  cmd.parse(argc, argv);
-
-  if (arg_extract.isSet() && arg_compress.isSet()) {
-    // This is a bit ugly but `none-or-xor` only available in
-    // tclap-1.4 which is not well supported in current
-    // distributions
-    auto out = TCLAP::StdOutput{};
-    TCLAP::ArgException e{
-        "Cannot compress `-c` and extract `-x` simultaneously"};
-    try {
-      out.failure(cmd, e);
-    } catch (TCLAP::ExitException& e) {
-      exit(e.getExitStatus());
-    }
-  }
-
-  // `none-or-xor` ensured already
-  if (arg_extract.isSet()) xwim.setExtract();
-  if (arg_compress.isSet()) xwim.setCompress();
-
-  if (arg_outfile.isSet()) xwim.setOut(arg_outfile.getValue());
-  if (arg_infiles.isSet()) xwim.setIns(arg_infiles.getValue());
 
   try {
-    xwim.try_infer();
+    unique_ptr<UserIntent> user_intent = make_intent(user_opt);
-    xwim.dwim();
+    user_intent->execute();
   } catch (XwimError& e) {
     spdlog::error(e.what());
   }

src/meson.build

@@ -1,11 +1,12 @@
-xwim_src = ['main.cpp', 'Xwim.cpp', 'Archiver.cpp']
+xwim_src = ['main.cpp', 'Archiver.cpp', 'UserOpt.cpp', 'UserIntent.cpp']
+
 xwim_archiver = ['archiver/LibArchiver.cpp']
 
 is_static = get_option('default_library')=='static'
 
 xwim_libs = [dependency('libarchive', required: true, static: is_static),
-             dependency('fmt', required: true, static: is_static),
              dependency('spdlog', required: true, static: is_static),
+             dependency('fmt', required: true, static: is_static),
              dependency('tclap', required: true, static: is_static)]
 
 executable('xwim', xwim_src+xwim_archiver, dependencies: xwim_libs)

src/util/Log.hpp

@@ -1,6 +1,7 @@
 #pragma once
 #include <spdlog/common.h>
 #include <spdlog/spdlog.h>
+
 #include <cstdlib>
 #ifdef NDEBUG
 #define XWIM_LOGLEVEL SPDLOG_LEVEL_ERROR
@@ -57,7 +58,27 @@ spdlog::level::level_enum _init_from_compile() {
  * The determined level is then set for the default logger via
  * `spdlog::set_level`.
  */
-void init(spdlog::level::level_enum level = spdlog::level::level_enum::off) {
+void init(int verbosity = -1,
+          spdlog::level::level_enum level = spdlog::level::level_enum::off) {
+  if (verbosity != -1) {
+    switch (verbosity) {
+      case 0:
+        spdlog::set_level(spdlog::level::off);
+        break;
+      case 1:
+        spdlog::set_level(spdlog::level::info);
+        break;
+      case 2:
+        spdlog::set_level(spdlog::level::debug);
+        break;
+      case 3:
+      default:
+        spdlog::set_level(spdlog::level::trace);
+        break;
+    }
+    return;
+  }
+
   if (spdlog::level::level_enum::off != level) {
     spdlog::set_level(level);
     return;

test/meson.build

@@ -2,22 +2,10 @@
 gtest_proj = subproject('gtest')
 gtest_dep = gtest_proj.get_variable('gtest_main_dep')
 
-xwim_src = ['../src/archive.cpp',
+# subdir('archives')
-            '../src/archive_sys.cpp']
+user_opt_test_exe = executable('user_opt_test_exe',
-
+                              sources: ['user_opt_test.cpp', '../src/UserOpt.cpp'],
-subdir('archives')
-
-archive_test_exe = executable('archive_test_exe',
-                              sources: ['archive_test.cpp', xwim_src],
                               include_directories: ['../src'],
-                              dependencies: [gtest_dep, xwim_libs])
+                              dependencies: [gtest_dep])
 
-
+test('user opt parsing test', user_opt_test_exe)
-test('archive test', archive_test_exe)
-
-fileformats_test_exe = executable('fileformats_test_exe',
-                              sources: ['fileformats_test.cpp', xwim_src],
-                              include_directories: ['../src'],
-                              dependencies: [gtest_dep, xwim_libs])
-
-test('fileformats test', fileformats_test_exe)

test/user_opt_test.cpp

@@ -0,0 +1,90 @@
+#include <gtest/gtest-death-test.h>
+#include "gtest/gtest.h"
+#include <filesystem>
+#include <string>
+
+#include "UserOpt.hpp"
+
+TEST(UserOpt, compress) {
+  using namespace xwim;
+
+  // clang-format off
+  char* args[] = {
+    const_cast<char*>("xwim"),
+    const_cast<char*>("-c"),
+    const_cast<char*>("mandator_paths"),
+    nullptr};
+  // clang-format on
+
+  UserOpt uo = UserOpt{3, args};
+  ASSERT_TRUE(uo.compress);
+  ASSERT_FALSE(uo.extract);
+}
+
+TEST(UserOpt, exclusive_actions) {
+  using namespace xwim;
+
+  // clang-format off
+  char* args[] = {
+    const_cast<char*>("xwim"),
+    const_cast<char*>("-c"),
+    const_cast<char*>("-x"),
+    const_cast<char*>("mandatory_paths"),
+    nullptr};
+  // clang-format on
+
+  UserOpt uo = UserOpt{4, args};
+  ASSERT_TRUE(uo.compress);
+  ASSERT_TRUE(uo.extract);
+}
+
+TEST(UserOpt, whitespace_in_path) {
+  using namespace xwim;
+
+  // clang-format off
+  char* args[] = {
+    const_cast<char*>("xwim"),
+    const_cast<char*>("-c"),
+    const_cast<char*>("/foo/bar baz/a file"),
+    nullptr};
+  // clang-format on
+
+  UserOpt uo = UserOpt{3, args};
+  ASSERT_TRUE(uo.paths.find(std::filesystem::path("/foo/bar baz/a file")) !=
+              uo.paths.end());
+}
+
+TEST(UserOpt, mixed_output_and_paths) {
+  using namespace xwim;
+
+  // clang-format off
+  char* args[] = {
+    const_cast<char*>("xwim"),
+    const_cast<char*>("-o"),
+    const_cast<char*>("/foo/bar baz/output"),
+    const_cast<char*>("/foo/bar baz/a path"),
+    const_cast<char*>("/foo/bar baz/another path"),
+    nullptr};
+  // clang-format on
+
+  UserOpt uo = UserOpt{5, args};
+  ASSERT_TRUE(uo.paths.find(std::filesystem::path("/foo/bar baz/a path")) !=
+              uo.paths.end());
+  ASSERT_TRUE(uo.paths.find(std::filesystem::path("/foo/bar baz/another path")) !=
+              uo.paths.end());
+  ASSERT_TRUE(uo.out == std::filesystem::path("/foo/bar baz/output"));
+}
+
+TEST(UserOpt, output_defaults_to_nullopt) {
+  using namespace xwim;
+
+  // clang-format off
+  char* args[] = {
+    const_cast<char*>("xwim"),
+    const_cast<char*>("/foo/bar"),
+    nullptr};
+  // clang-format on
+
+  UserOpt uo = UserOpt{2, args};
+  ASSERT_FALSE(uo.out);
+}