test nouvelles fonctions

.circleci/config.yml

@@ -5,4 +5,5 @@ jobs:
       - image: purplekarrot/gcc-8
     steps:
       - checkout
+      - run: apt-get update && apt-get install -y doxygen doxygen-docs doxygen-latex doxygen-gui graphviz
       - run: cd build && cmake -DCMAKE_BUILD_TYPE=Release .. && make -j

.gitignore

@@ -1,13 +1,15 @@
 *~
 
 gmon\.out
+/cmake-build-debug/Makefile
 
+cmake-build-debug/
+
+\.idea/
 *.lzw
 
-\.scannerwork/
+docs/Doxyfile
 
-bw-output/
+docs/html/
 
-bin/
-build/
-debug/
+docs/latex/

.gitlab-ci.yml

@@ -1,7 +0,0 @@
-image: rikorose/gcc-cmake:latest
-stages:
-  - build
-build:
-  stage: build
-  script:
-    - mkdir -p build bin && cd build && cmake -DCMAKE_BUILD_TYPE=Release .. && make -j

.travis.yml

@@ -4,48 +4,66 @@ compiler: clang
 os: linux
 dist: trusty
 addons:
-    sonarcloud:
-        organization: "phundrak-github"
-        token:
-            secure: ${SONAR_TOKEN}
     apt:
-        config:
-            retries: true
         sources:
             - ubuntu-toolchain-r-test
             - llvm-toolchain-trusty-5.0
         packages:
             - g++-7
             - clang-5.0
-cache:
-    apt: true
+            - doxygen
+            - doxygen-doc
+            - doxygen-latex
+            - doxygen-gui
+            - graphviz
+
 matrix:
   include:
     # Linux Clang C++17
     - env:
-        BUILD_TYPE=debug
+        BUILD_TYPE=Debug
         BIN_DIR=debug
         CC=clang-5.0
         CXX=clang++-5.0
     - env:
-        BUILD_TYPE=release
+        BUILD_TYPE=Release
         BIN_DIR=bin
         CC=clang-5.0
         CXX=clang++-5.0
 
     # Linux GCC C++17
     - env:
-        - MATRIX_EVAL="BUILD_TYPE=debug && BIN_DIR=debug && CC=gcc-7 && CXX=g++-7"
+        - MATRIX_EVAL="BUILD_TYPE=Debug && BIN_DIR=debug && CC=gcc-7 && CXX=g++-7"
       compiler: gcc
       before_install:
         - eval "${MATRIX_EVAL}"
     - env:
-        - MATRIX_EVAL="BUILD_TYPE=release && BIN_DIR=bin && CC=gcc-7 && CXX=g++-7"
+        - MATRIX_EVAL="BUILD_TYPE=Release && BIN_DIR=bin && CC=gcc-7 && CXX=g++-7"
       compiler: gcc
       before_install:
         - eval "${MATRIX_EVAL}"
 
+    # OSX Clang
+    # - os: osx
+    #   osx_image: xcode9.3
+    #   env:
+    #     BUILD_TYPE=Release
+    #     BIN_DIR=bin
+    #   before_install:
+    #     - brew update
+    #     - brew install doxygen graphviz
+    #   install:
+    #     brew upgrade cmake
+    # - os: osx
+    #   osx_image: xcode9.3
+    #   env:
+    #     BUILD_TYPE=Debug
+    #     BIN_DIR=debug
+    #   before_install:
+    #     brew update
+    #   install:
+    #     brew upgrade cmake
 script:
-    - make $BUILD_TYPE
-    - build-wrapper-linux-x86-64 --out-dir bw-output make clean all
-    - sonar-scanner -Dsonar.projectKey=Phundrak_lzw-assignment -Dsonar.sources=./src/ -Dsonar.cfamily.build-wrapper-output=bw-output -Dsonar.host.url=https://sonarcloud.io
+    - cd build
+    - cmake -DCMAKE_BUILD_TYPE=$BUILD_TYPE ..
+    - make -j

CMakeLists.txt

@@ -7,7 +7,7 @@ set(TGT "projet_lzw")
 set(${TGT}_VERSION_MAJOR 0)
 set(${TGT}_VERSION_MINOR 1)
 
-set(CXX_COVERAGE_COMPILE_FLAGS "-pedantic -Wall -Wextra -Wold-style-cast -Woverloaded-virtual -Wfloat-equal -Wwrite-strings -Wpointer-arith -Wcast-qual -Wcast-align -Wconversion -Wsign-conversion -Wshadow -Weffc++ -Wredundant-decls -Wdouble-promotion -Winit-self -Wswitch-default -Wswitch-enum -Wundef -Winline -Wunused -Wnon-virtual-dtor -pthread")
+set(CXX_COVERAGE_COMPILE_FLAGS "-pedantic -Wall -Wextra -Wold-style-cast -Woverloaded-virtual -Wfloat-equal -Wwrite-strings -Wpointer-arith -Wcast-qual -Wcast-align -Wshadow -Weffc++ -Wredundant-decls -Wdouble-promotion -Winit-self -Wswitch-default -Wswitch-enum -Wundef -Winline -Wunused -Wnon-virtual-dtor -Wno-conversion -pthread")
 set(CMAKE_CXX_FLAGS_DEBUG "${CXX_COVERAGE_COMPILE_FLAGS} -DDebug -g -pg")
 set(CMAKE_CXX_FLAGS_RELEASE "${CXX_COVERAGE_COMPILE_FLAGS} -O3")
 
@@ -34,6 +34,26 @@ set(CMAKE_RUNTIME_OUTPUT_DIRECTORY_DEBUG "../debug/")
 set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${CXX_COVERAGE_COMPILE_FLAGS}")
 set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${CXX_COVERAGE_COMPILE_FLAGS}")
 
+# indicate the doc build as an option, ON by default
+if(CMAKE_BUILD_TYPE MATCHES "^[Rr]elease")
+    option(BUILD_DOC "Build documentation" ON)
+    find_package(Doxygen
+                 REQUIRED dot)
+    if(DOXYGEN_FOUND)
+        set(DOXYGEN_IN ${CMAKE_CURRENT_SOURCE_DIR}/docs/Doxyfile.in)
+        set(DOXYGEN_OUT ${CMAKE_CURRENT_SOURCE_DIR}/docs/Doxyfile)
+        configure_file(${DOXYGEN_IN} ${DOXYGEN_OUT} @ONLY)
+        message("Doxygen build started")
+        add_custom_target(docs ALL
+            COMMAND ${DOXYGEN_EXECUTABLE} ${DOXYGEN_OUT}
+            WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}
+            COMMENT "Generating API documentation with Doxygen"
+            VERBATIM )
+    else()
+        message("Doxygen needs to be installed to generate the doxygen documentation.")
+    endif()
+endif()
+
 include_directories(includes)
 file(GLOB SOURCES "src/*.cc")
 add_executable(${TGT} ${SOURCES})

Makefile

@@ -1,16 +0,0 @@
-all:
-	@mkdir -p build
-	@cd build && pwd && cmake -DCMAKE_BUILD_TYPE=Release .. && make
-	@strip bin/project_lzw
-
-release:
-	@mkdir -p build
-	@cd build && pwd && cmake -DCMAKE_BUILD_TYPE=Release .. && make
-	@strip bin/projet_lzw
-
-debug:
-	@mkdir -p build
-	@cd build && pwd && cmake -DCMAKE_BUILD_TYPE=Debug .. && make
-
-clean:
-	rm -rf bin/* build/* debug/*

README.md

@@ -2,7 +2,6 @@
 [![Travis Badge](https://travis-ci.org/Phundrak/lzw-assignment.svg?branch=master)](https://travis-ci.org/Phundrak/lzw-assignment)
 [![Codacy Badge](https://api.codacy.com/project/badge/Grade/80cf9a0514554f368effaf78d8e4ae15)](https://www.codacy.com/app/Phundrak/lzw-assignment?utm_source=github.com&utm_medium=referral&utm_content=Phundrak/lzw-assignment&utm_campaign=Badge_Grade)
 [![CodeFactor](https://www.codefactor.io/repository/github/phundrak/lzw-assignment/badge)](https://www.codefactor.io/repository/github/phundrak/lzw-assignment)
-[![SonarCloud Badge](https://sonarcloud.io/api/project_badges/measure?project=Phundrak_lzw-assignment&metric=alert_status)](https://sonarcloud.io/dashboard?id=Phundrak_lzw-assignment)
 [![Spacemacs Badge](https://cdn.rawgit.com/syl20bnr/spacemacs/442d025779da2f62fc86c2082703697714db6514/assets/spacemacs-badge.svg)](http://spacemacs.org)
 
 # LZW Compressing tool
@@ -10,16 +9,3 @@
 This is a university assignment for which I aim to create an LZW algorithm implementation to create a small tool similar to `gzip` and `gunzip` that can compress and uncompress files in a lossless fashion.
 
 This project is written is C++17, compiled with clang under a UNIX environment. Other compilers and environments will not be tested.
-
-## How to use it
-
-Currently, five different options are available to the user:
-- `-h` or `--help` will show how to use `projet_lzw`
-- `-c` or `--compress` tells `projet_lzw` to compress the input file
-- `-u` or `--uncompress` tells `projet_lzw` to uncompress the input file
-- `-i <file>` or `--input <file path>` specifies the input file to be compressed or uncompressed (**MANDATORY**)
-- `-o <file>` or `--output <file path>` specifies the name of the output file. If not used, the default output name for compression is `output.lzw`, and the default name for uncompressing is `<filename>_uncompressed`.
-
-By default, `projet_lzw` will uncompress the (mandatory) input file.
-
-It is planned to add in the future a sixth option, `-p` or `--passes` that will allow to compress multiple times the input file.

bin/.gitignore

@@ -0,0 +1,2 @@
+*
+!.gitignore

build/.gitignore

@@ -0,0 +1,2 @@
+*
+!.gitignore

debug/.gitignore

@@ -0,0 +1,2 @@
+*
+!.gitignore

docs/Doxyfile.in

@@ -0,0 +1,33 @@
+OUTPUT_DIRECTORY       = @CMAKE_CURRENT_SOURCE_DIR@/docs/
+INPUT                  = @CMAKE_CURRENT_SOURCE_DIR@/src/ @CMAKE_CURRENT_SOURCE_DIR@/docs
+DOXYFILE ENCODING      = UTF-8
+PROJECT_NAME           = "Compression LZW"
+PROJECT_NUMBER         = 0.3
+PROJECT_BRIEF          = "Utilitaire de compression/décompression de fichiers via l’algorithme LZW"
+ALLOW_UNICODE_NAMES    = YES
+OUTPUT_LANGUAGE        = French
+FULL_PATH_NAMES        = NO
+TAB_SIZE               = 2
+EXTRACT_ALL            = YES
+CASE_SENSE_NAMES       = YES
+INPUT_ENCODING         = UTF-8
+FILE_PATTERNS          = *.cc \
+                          *.hh
+RECURSIVE              = YES
+EXAMPLE_PATTERNS       = *
+SOURCE_BROWSER         = YES
+INLINE_SOURCES         = YES
+REFERENCED_BY_RELATION = YES
+REFERENCES_RELATION    = YES
+USE_HTAGS              = NO
+HTML_TIMESTAMP         = YES
+GENERATE_DOCSET        = NO
+GENERATE_HTMLHELP      = NO
+LATEX_SOURCE_CODE      = YES
+LATEX_TIMESTAMP        = YES
+ENABLE_PREPROCESSING   = NO
+HAVE_DOT               = YES
+UML_LOOK               = YES
+CALL_GRAPH             = YES
+CALLER_GRAPH           = YES
+INTERACTIVE_SVG        = YES

src/bitpack.hh

@@ -1,3 +1,8 @@
+/**
+ *   \file bitpack.hh
+ *   \brief Header for bit-packing functions
+ */
+
 #ifndef LZW_SRC_BITPACK_H_
 #define LZW_SRC_BITPACK_H_
 
@@ -5,28 +10,84 @@
 #include <string>
 #include <vector>
 
-[[nodiscard]] std::vector<unsigned char>
-pack(const std::vector<std::uint16_t> &);
+///////////////////////////////////////////////////////////////////////////////
+//                                  packing                                  //
+///////////////////////////////////////////////////////////////////////////////
 
-[[nodiscard]] std::vector<unsigned char>
-pack_n(std::vector<std::uint16_t>::const_iterator,
-       std::vector<std::uint16_t>::const_iterator, std::vector<unsigned char> &,
-       int);
+std::vector<unsigned char> pack(const std::vector<std::uint16_t> &);
 
-[[nodiscard]] std::vector<unsigned char>
-pack_16(std::vector<std::uint16_t>::const_iterator,
-        std::vector<std::uint16_t>::const_iterator,
-        std::vector<unsigned char> &);
+std::vector<unsigned char> pack_9(std::vector<std::uint16_t>::const_iterator,
+                                  std::vector<std::uint16_t>::const_iterator);
 
-[[nodiscard]] std::vector<std::uint16_t>
-unpack(std::basic_string<unsigned char> &&);
+std::vector<unsigned char> pack_10(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
 
-[[nodiscard]] std::vector<std::uint16_t>
-unpack_n(std::basic_string<unsigned char>::const_iterator t_begin,
-         std::basic_string<unsigned char>::const_iterator t_end,
-         std::vector<std::uint16_t> &, int t_n);
+std::vector<unsigned char> pack_11(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
 
-[[nodiscard]] std::vector<std::uint16_t>
+std::vector<unsigned char> pack_12(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
+
+std::vector<unsigned char> pack_13(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
+
+std::vector<unsigned char> pack_14(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
+
+std::vector<unsigned char> pack_15(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
+
+std::vector<unsigned char> pack_16(std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<std::uint16_t>::const_iterator,
+                                   std::vector<unsigned char> &);
+
+///////////////////////////////////////////////////////////////////////////////
+//                                   unpack                                  //
+///////////////////////////////////////////////////////////////////////////////
+
+std::vector<std::uint16_t> unpack(std::basic_string<unsigned char> &&);
+
+std::vector<std::uint16_t>
+unpack_9(std::basic_string<unsigned char>::const_iterator,
+         std::basic_string<unsigned char>::const_iterator);
+
+std::vector<std::uint16_t>
+unpack_10(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
+unpack_11(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
+unpack_12(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
+unpack_13(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
+unpack_14(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
+unpack_15(std::basic_string<unsigned char>::const_iterator,
+          std::basic_string<unsigned char>::const_iterator,
+          std::vector<std::uint16_t> &);
+
+std::vector<std::uint16_t>
 unpack_16(std::basic_string<unsigned char>::const_iterator,
           std::basic_string<unsigned char>::const_iterator,
           std::vector<std::uint16_t> &);

src/common.cc

@@ -1,3 +1,8 @@
+/**
+ *   \file common.cc
+ *   \brief Implementation for functions in common
+ */
+
 #include "common.hh"
 
 using std::uint16_t;
@@ -6,7 +11,7 @@ using dic_comp_t = std::map<std::pair<uint16_t, uint8_t>, uint16_t>;
 using ustring = std::basic_string<unsigned char>;
 using p_ustring = std::shared_ptr<ustring>;
 
-[[nodiscard]] int ipow(int base, int exp) {
+int ipow(int base, int exp) {
   int result = 1;
   for (;;) {
     if (exp & 1) {
@@ -21,21 +26,43 @@ using p_ustring = std::shared_ptr<ustring>;
   return result;
 }
 
-[[nodiscard]] std::pair<bool, uint16_t>
-dico(dic_comp_t &t_dictionary, const uint16_t t_nr_chaine, const uint8_t t_c) {
+/**
+ *  Cette fonction a pour double usage la recherche d’une chaine de caractères
+ *  dans le dictionnaire, ou bien l’ajout d’une nouvelle chaîne si celle-ci
+ * n’est pas déjà présente. Une chaine de caractères est représentée par un
+ * couple numéro de chaine / caractère, le numéro de chaine renvoyant au
+ * caractère précédent (soit son code ASCII, soit son indice dans le
+ * dictionnaire) et le caractère se référant au dernier caractère de la chaine
+ * courante. Si le numéro de chaine est -1, alors il s’agit du premier caractère
+ * de la chaine, et la valeur renvoyée sera la valeur ASCII du caractère. La
+ * fonction renvoie une paire bool/uint16_t, la valeur booléene indiquant si une
+ * nouvelle fut ajoutée dans le dictionnaire ou non, et le uint16_t indiquant la
+ * valeur numérique de la chaîne dans le dictionnaire.
+ *
+ *  \param t_dictionary Dictionnaire
+ *  \param t_nr_chaine Numéro de la chaine précédant le caractères \p t_c dans \p t_dictionary
+ *  \param t_c Caractère suivant la chaine de caractères \p t_nr_chaine
+ *  \return const std::pair<bool, uint16_t>
+ */
+std::pair<bool, uint16_t> dico(dic_comp_t &t_dictionary,
+                               const uint16_t t_nr_chaine, const uint8_t t_c) {
   if (t_nr_chaine == 0xFFFF) {
     return std::make_pair(true, t_c);
   }
   auto &e = t_dictionary[std::make_pair(t_nr_chaine, t_c)];
-  if (e != 0)
-    return std::make_pair(true, e);
-  e = static_cast<uint16_t>(t_dictionary.size() + 255);
-  return std::make_pair(false, e);
+  return (e != 0) ? std::make_pair(true, e)
+                  : std::make_pair(false, (e = static_cast<uint16_t>(
+                                               t_dictionary.size() + 255)));
 }
 
-[[nodiscard]] ustring dico_uncompress(std::map<uint16_t, ustring> &t_dict,
-                                      const uint16_t t_code,
-                                      const uint16_t t_old) {
+/**
+ *  Detailed description
+ *
+ *  \param t_dict Dictionnaire
+ *  \return Retourne une chaîne de caractères non signés
+ */
+ustring dico_uncompress(std::map<uint16_t, ustring> &t_dict,
+                        const uint16_t t_code, const uint16_t t_old) {
   // le code existe dans le dictionnaire s’il est < 256
   if (t_code < 256) {
     ustring e{static_cast<unsigned char>(t_code)};

src/common.hh

@@ -1,3 +1,8 @@
+/**
+ *   \file common.hh
+ *   \brief Header for functions in common
+ */
+
 #ifndef LZW_SRC_COMMON_H_
 #define LZW_SRC_COMMON_H_
 
@@ -5,14 +10,15 @@
 #include <map>
 #include <memory>
 
-[[nodiscard]] int ipow(int, int);
+int ipow(int, int);
 
-[[nodiscard]] std::pair<bool, std::uint16_t>
+/// \brief Recherche ou ajout de chaine dans le dictionnaire
+std::pair<bool, std::uint16_t>
 dico(std::map<std::pair<std::uint16_t, std::uint8_t>, std::uint16_t> &,
-     const std::uint16_t, const std::uint8_t);
+     std::uint16_t, std::uint8_t);
 
-[[nodiscard]] std::basic_string<unsigned char>
+std::basic_string<unsigned char>
 dico_uncompress(std::map<std::uint16_t, std::basic_string<unsigned char>> &,
-                const std::uint16_t, const std::uint16_t);
+                std::uint16_t, std::uint16_t);
 
 #endif /* LZW_SRC_COMMON_H_ */

src/compress.cc

@@ -1,11 +1,14 @@
+/**
+ *   \file compress.cc
+ *   \brief Implementation of compression
+ */
+
 #include "compress.hh"
-#include "common.hh"
 #include "io.hh"
+#include "common.hh"
 #include <cassert>
 #include <cstdlib>
 #include <fstream>
-#include <iterator>
-using std::ios;
 using std::string;
 using std::uint16_t;
 using std::uint8_t;
@@ -17,39 +20,33 @@ using ustring = std::basic_string<unsigned char>;
 using dict_t = std::map<std::pair<uint16_t, uint8_t>, uint16_t>;
 using std::printf;
 
-[[nodiscard]] ustring read_file(const string &filename) {
-  std::ifstream file{filename, ios::binary};
-  assert(file);
-  file.unsetf(ios::skipws);
-  file.seekg(0, ios::end);
-  const auto file_size = file.tellg();
-  file.seekg(0, ios::beg);
-  ustring res{};
-  res.reserve(file_size);
-  res.insert(res.begin(), std::istream_iterator<unsigned char>(file),
-             std::istream_iterator<unsigned char>());
-  file.close();
-  return res;
-}
+const size_t DICT_MAX = static_cast<size_t>(ipow(2, 17) - 256); /* 16 bits */
 
-[[nodiscard]] vvuint16 lzw_compress(ustring &&t_text) {
-  vvuint16 res{};
-  const auto DICT_MAX = static_cast<size_t>(ipow(2, 14) - 256); /* 16 bits */
+/**
+ *  La chaîne de caractères \p t_text est lue caractère par caractère, et est
+ * selon la valeur de retour de la fonction \ref dico (permettant dans le même
+ * temps la création du dictionnaire), on rajoute ou non un nouveau caractère
+ * encodé sur 12bits dans le chunk courant. Dès que le dictionnaire est plein
+ * (2^12 caractères), le chunk est sauvegardé et vidé, et le dictionnaire est
+ * réinitialisé.
+ *
+ *  \param t_text Chaîne de caractères uint8_t représentant le fichier d'entrée
+ *  \return Vecteur de chunks (vecteurs de uint16_t)
+ */
+vvuint16 lzw_compress(ustring &&t_text) {
+  std::puts("Compressing...");
   uint16_t w = 0xFFFF;
   vuint16 chunk{};
+  vvuint16 res{};
   dict_t dict{};
   for (const auto c : t_text) {
     if (dict.size() >= DICT_MAX) {
-      if (w != 0xFFFF) {
-        chunk.push_back(w);
-      }
-      res.push_back(chunk);
+      res.push_back(std::move(chunk));
+      chunk = vuint16{};
+      dict = dict_t{};
       w = 0xFFFF;
-      chunk.clear();
-      dict.clear();
     }
-    if (const auto &[exists, pos] = dico(dict, w, static_cast<uint8_t>(c));
-        exists) {
+    if (const auto &[yes, pos] = dico(dict, w, static_cast<uint8_t>(c)); yes) {
       w = pos;
     } else {
       chunk.push_back(w);
@@ -63,15 +60,30 @@ using std::printf;
   return res;
 }
 
+/**
+ *  Wrapper de la fonction \ref lzw_compress gérant l'ouverture, la lecture,
+ *  l'écriture et la fermeture des fichiers d’entrée et de sortie. Si \p
+ *  t_out_file est nul (chemin non spécifié), il prendra alors la valeur de
+ *  \p t_in_file à laquelle sera annexé l’extension `.lzw`.
+ *
+ *  \param[in] t_in_file Chemin vers le fichier d’entrée
+ *  \param[in] t_out_file Chemin vers le fichier de sortie
+ */
 void compress(const std::string &t_in_file, const char *t_out_file) {
-  std::ofstream out{(t_out_file != nullptr) ? t_out_file : "output.lzw",
-                    ios::out | ios::binary};
-  if (!out.is_open()) {
-    std::cerr << "Error at " << __FILE__ << ":" << __LINE__ - 4
-              << ": could not open output file. Aborting...\n";
-    exit(1);
-  }
-  const auto compressed_text(lzw_compress(read_file(t_in_file)));
+  FILE *const input_file = fopen(t_in_file.c_str(), "rb");
+  assert(input_file);
+  FILE *const out = (t_out_file != nullptr) ? fopen(t_out_file, "wb")
+                                            : fopen("output.lzw", "wb");
+  assert(out);
+
+  std::fseek(input_file, 0L, SEEK_END);
+  const auto file_size = static_cast<size_t>(ftell(input_file));
+  std::rewind(input_file);
+
+  auto raw_text = std::make_unique<unsigned char[]>(file_size);
+  std::fread(raw_text.get(), sizeof(unsigned char), file_size, input_file);
+  const auto compressed_text(lzw_compress(ustring{raw_text.get(), &raw_text[file_size]}));
   write_file(out, compressed_text);
-  out.close();
+  fclose(out);
+  fclose(input_file);
 }

src/compress.hh

@@ -1,3 +1,8 @@
+/**
+ *   \file compress.hh
+ *   \brief Header for compression functions
+ */
+
 #ifndef LZW_SRC_COMPRESS_H_
 #define LZW_SRC_COMPRESS_H_
 
@@ -6,9 +11,11 @@
 #include <thread>
 #include <vector>
 
-[[nodiscard]] std::vector<std::vector<std::uint16_t>>
+/// \brief Compression d'une chaine de caractères
+std::vector<std::vector<std::uint16_t>>
 lzw_compress(std::basic_string<unsigned char> &&);
 
+/// \brief Wrapper de \ref lzw_compress
 void compress(const std::string &, const char *);
 
 #endif /* LZW_SRC_COMPRESS_H_ */

src/io.cc

@@ -1,27 +1,65 @@
+/**
+ *   \file io.cc
+ *   \brief Body for file reading and writing
+ */
+
 #include "io.hh"
 #include "bitpack.hh"
 #include <array>
+#include <algorithm>
+
+#ifdef Debug
+constexpr bool debug_mode = true;
+#include <algorithm>
+#else
+constexpr bool debug_mode = false;
+#endif
 
-using std::uint16_t;
 using std::vector;
+using std::uint16_t;
 using vuint16 = vector<uint16_t>;
 using vvuint16 = vector<vuint16>;
 
-void write_file(std::ofstream &t_out, const vvuint16 &t_chunks) {
-  const auto nr_chunks = static_cast<uint16_t>(t_chunks.size());
-#ifdef Debug
-  std::printf("Number of chunks: %u\n", nr_chunks);
-#endif
-  t_out.write(reinterpret_cast<const char *>(&nr_chunks), sizeof(nr_chunks));
-  for (const auto &chunk : t_chunks) {
+/**
+ *  Écrit dans le fichier \p t_out les chunks passés en paramètre. Le fichier de
+ *  sortie est composé des éléments suivants :\n
+ *  - Sur deux octets sont écrit un `uint16_t` déterminant le nombre de chunk
+ *    composant le fichier\n
+ *  - Sont ensuite écrits les chunks sur un nombre variable d’octets suivant la
+ *    taille des chunks\n
+ *  \n
+ *  Un chunk est composé de la manière qui suit :\n
+ *  - Sur deux octets sont écrit un `uint32_t` déterminant le nombre d’octets
+ *    composant le chunk\n
+ *  - Sur le nombre d’octets précisés par le header du chunk se trouvent les
+ *    données compressées par l’algorithme lzw puis via bit-packing.\n
+ *
+ *  \param[out] t_out Fichier de sortie
+ *  \param[in] t_text Collection ordonnée des chunks à écrire dans \p t_out
+ */
+void write_file(FILE *const t_out, const vvuint16 &t_text) {
+  const auto size = static_cast<uint16_t>(t_text.size());
+  if constexpr (debug_mode) {
+    std::printf("Number of chunks: %u\n", size);
+  }
+  fwrite(&size, sizeof(size), 1, t_out);
+  for (const auto &chunk : t_text) {
+    if constexpr (debug_mode)
+      std::printf("Chunk!\n");
     write_chunk(t_out, chunk);
   }
 }
 
-void write_chunk(std::ofstream &t_out, const vuint16 &t_chunk) {
+/**
+ *  Écrit dans le fichier \p t_out le chunk unique \p t_chunk. Se référer à la
+ *  documentation de \ref write_file pour plus de détails.
+ *
+ *  \param t_out Output file
+ *  \param t_chunk Chunk to be written to \p t_out
+ */
+void write_chunk(FILE *const t_out, const vuint16 &t_chunk) {
   const auto output = pack(t_chunk);
   const auto chunk_size = static_cast<uint32_t>(output.size());
-  t_out.write(reinterpret_cast<const char *>(&chunk_size), sizeof(chunk_size));
-  t_out.write(reinterpret_cast<const char *>(output.data()),
-              sizeof(output[0]) * output.size());
+  fwrite(&chunk_size, sizeof(chunk_size), 1, t_out);
+  fwrite(output.data(), sizeof(output[0]), output.size(), t_out);
 }

src/io.hh

@@ -1,15 +1,35 @@
+/**
+ *   \file io.hh
+ *   \brief Header for file reading and writing
+ */
+
 #ifndef LZW_SRC_IO_H_
 #define LZW_SRC_IO_H_
 
 #include <cstdint>
 #include <cstdio>
-#include <fstream>
 #include <iostream>
 #include <vector>
 
-void write_file(std::ofstream &,
-                const std::vector<std::vector<std::uint16_t>> &);
+/*
+ * Un fichier compressé se compose ainsi :
+ * char_size : taille d'un caractère en bits (1B)
+ * nb_chunk : nombre de chunks (4B)
+ * chunks* : chunks
+ *
+ * Un chunk se compose ainsi :
+ * nb_char_chunk : nombre de caractères du chunk (2B)
+ * text* : caractères de taille char_size (ceil((char_size * nb_char_chunk) /
+ * 8))
+ *
+ * Si le dernier caractère ne termine pas le dernier octet du chunk, les
+ * derniers bits sont mit à zéro
+ */
 
-void write_chunk(std::ofstream &, const std::vector<std::uint16_t> &);
+/// \brief Écrit dans le fichier le texte compressé
+void write_file(FILE *, const std::vector<std::vector<std::uint16_t>> &);
+
+/// \brief Écrit un chunk dans le fichier de sortie
+void write_chunk(FILE *, const std::vector<std::uint16_t> &);
 
 #endif /* LZW_SRC_IO_H_ */

src/main.cc

@@ -1,8 +1,16 @@
+/**
+ *   \file main.cc
+ *   \brief Main file
+ *
+ *
+ *
+ */
+
+#include <getopt.h>
+#include <cassert>
+#include <tuple>
 #include "compress.hh"
 #include "uncompress.hh"
-#include <cassert>
-#include <getopt.h>
-#include <tuple>
 
 using std::printf;
 using std::puts;
@@ -11,35 +19,41 @@ using std::tuple;
 
 // custom types ///////////////////////////////////////////////////////////////
 
+/*
+  Dictionnaire :
+  <
+      <
+          numéro chaine précédente,
+          caractère ASCII
+      >
+      numéro chaine courante
+  >
+ */
 using dic_t = std::map<std::pair<uint32_t, uint8_t>, uint32_t>;
 using ustring = std::basic_string<uint8_t>; // chaine non encodée
 using uvec = std::vector<uint32_t>;         // chaine encodée
 
+/**
+ *  \brief Affichage d’aide
+ */
 void help() {
-  puts("Usage:\n\
-lzw  [-options] [-i path] [-o path]\n\n\
-The default action is to compress the input file to a .lzw file\n\
-in which the directory in which the software is executed.\n\
-Options available:\n\
--h --help\n\
-\tdisplay the current message\n\
--i --input\n\
-\tpath to the input file (MANDATORY)\n\
--o --output\n\
-\tpath to the output file (if the file already exists, it will be\n\n\
-\toverwritten). Default: input path + \".lzw\\n\
--c --compress\n\
-\tcompress the input file\n\
--u --uncompress\n\
-\tuncompresses the input file to the output file. If no output path\n\
-\thas not been entered and if the input file ends with \".lzw\",\n\
-\tthe extension \".lzw\" will be removed; otherwise, the extension\n\
-\t\"_uncompresed\" will be added");
+  puts("Usage:");
+  puts("lzw  [-options] [-i path] [-o path]");
+  puts("\tThe default action is to compress the input file to a .lzw file");
+  puts("\tin which the directory in which the software is executed.");
+  puts("\tOptions available:");
+  puts("\t-i\tpath to the input file (mandatory)");
+  puts("\t-o\tpath to the output file (if the file already exists, it will");
+  puts("\t\tbe overwritten). Default: input path + \".lzw\"");
+  puts("\t-c\tcompress the input file");
+  puts("\t-u\tuncompresses the input file to the output file. If no output");
+  puts("\t\tpath has not been entered and if the input file ends with ");
+  puts("\t\t\".lzw\", the extension \".lzw\" will be removed; otherwise, the ");
+  puts("\t\textension \".uncompresed\" will be added");
 }
 
-[[nodiscard]] std::tuple<string, string, bool> process_args(int t_argc,
-                                                            char *t_argv[]) {
-  auto ret = std::make_tuple(string{}, string{}, true);
+std::tuple<string, string, bool> process_args(int t_argc, char *t_argv[]) {
+  auto ret = std::make_tuple(string{}, string{}, false);
   while (true) {
     int option_index = 0;
     static struct option long_options[] = {
@@ -50,8 +64,7 @@ Options available:\n\
         {"uncompress", no_argument, nullptr, 'u'},
         {nullptr, 0, nullptr, 0}};
     int c = getopt_long(t_argc, t_argv, "hi:o:cu", long_options, &option_index);
-    if (c == -1)
-      break;
+    if (c == -1) break;
     switch (c) {
     case 0:
       break;
@@ -81,12 +94,10 @@ Options available:\n\
   return ret;
 }
 
+/* TODO: compression multiple : nombre de compressions puis fichier compressé */
 int main(int argc, char *argv[]) {
   const auto [input_path, output_path, compressing] = process_args(argc, argv);
-  if (input_path.empty()) {
-    help();
-    return 0;
-  }
+  assert(!input_path.empty());
   if (compressing) {
     compress(input_path, (output_path.empty()) ? nullptr : output_path.c_str());
   } else {

src/uncompress.cc

@@ -4,8 +4,6 @@
 #include <cassert>
 #include <cstdio>
 #include <cstdlib>
-#include <fstream>
-#include <iterator>
 
 using std::fclose;
 using std::fopen;
@@ -16,16 +14,18 @@ using std::vector;
 using ustring = std::basic_string<unsigned char>;
 using vuint16 = vector<uint16_t>;
 
-[[nodiscard]] ustring lzw_uncompress(vuint16 &&t_compressed) {
+ustring lzw_uncompress(vuint16 &&t_compressed) {
   ustring ret{};
   uint16_t old = 0;
   std::map<uint16_t, ustring> dict{};
-  ret.append({static_cast<unsigned char>(t_compressed[0])});
-  old = t_compressed[0];
+  uint16_t v = t_compressed[0];
+  ret.append({static_cast<unsigned char>(v)});
+  old = v;
   for (auto it = t_compressed.begin() + 1; it != t_compressed.end(); ++it) {
-    const auto uncompressed{dico_uncompress(dict, *it, old)};
+    v = *it;
+    const auto uncompressed{dico_uncompress(dict, v, old)};
     ret.insert(ret.end(), uncompressed.begin(), uncompressed.end());
-    old = *it;
+    old = v;
   }
 
   return ret;
@@ -34,27 +34,31 @@ using vuint16 = vector<uint16_t>;
 void uncompress(const string &t_input_name, const char *t_output_name) {
   FILE *const input = std::fopen(t_input_name.c_str(), "rb");
   assert(input);
-  std::ofstream output{(t_output_name != nullptr)
-                           ? t_output_name
-                           : t_input_name + "_uncompressed",
-                       std::ios::out | std::ios::binary};
-  assert(output.is_open());
+
+  FILE *const output =
+      (t_output_name != nullptr)
+          ? std::fopen(t_output_name, "wb")
+          : std::fopen((t_input_name + "_uncompressed").c_str(), "wb");
+  assert(output);
+
   uint16_t nb_chunks = 0;
+  std::fseek(input, 0, SEEK_SET);
   std::fread(&nb_chunks, sizeof(nb_chunks), 1, input);
+
   for (uint16_t i = 0; i < nb_chunks; ++i) {
-    uncompress_chunk(input, output);
+    uint32_t size_chunk = 0;
+    fread(&size_chunk, sizeof(size_chunk), 1, input);
+    auto *chunk = static_cast<unsigned char *>(
+        std::malloc(sizeof(unsigned char) * size_chunk));
+    fread(chunk, sizeof(unsigned char), size_chunk, input);
+
+    auto unpacked = unpack(ustring{chunk, chunk + size_chunk});
+    const auto uncompressed_chunk = lzw_uncompress(std::move(unpacked));
+    // sometimes will add null char
+    std::fwrite(uncompressed_chunk.data(), sizeof(uncompressed_chunk[0]),
+                uncompressed_chunk.size(), output);
   }
-  output.close();
+
+  std::fclose(output);
   std::fclose(input);
 }
-
-void uncompress_chunk(FILE *const t_input, std::ofstream &t_output) {
-  uint32_t size_chunk = 0;
-  fread(&size_chunk, sizeof(size_chunk), 1, t_input);
-  auto chunk = std::make_unique<unsigned char[]>(size_chunk);
-  fread(chunk.get(), sizeof(unsigned char), size_chunk, t_input);
-  auto unpacked = unpack(ustring{chunk.get(), chunk.get() + size_chunk});
-  auto uncompressed_chunk = lzw_uncompress(std::move(unpacked));
-  t_output.write(reinterpret_cast<const char *>(uncompressed_chunk.data()),
-                 sizeof(uncompressed_chunk[0]) * uncompressed_chunk.size());
-}

src/uncompress.hh

@@ -1,16 +1,12 @@
 #ifndef LZW_SRC_UNCOMPRESS_H_
 #define LZW_SRC_UNCOMPRESS_H_
 
-#include <fstream>
 #include <memory>
 #include <string>
 #include <vector>
 
-[[nodiscard]] std::basic_string<unsigned char>
-lzw_uncompress(std::vector<std::uint16_t> &&);
+std::basic_string<unsigned char> lzw_uncompress(std::vector<std::uint16_t> &&);
 
 void uncompress(const std::string &, const char *);
 
-void uncompress_chunk(FILE *, std::ofstream &);
-
 #endif /* LZW_SRC_UNCOMPRESS_H_ */