support for arguments + moved compression lines

This commit is contained in:
Phuntsok Drak-pa 2018-04-10 10:39:41 +02:00
parent 2345c3642a
commit d48012eb1d
6 changed files with 185 additions and 59 deletions

2
.gitignore vendored
View File

@ -1 +1,3 @@
*~ *~
gmon\.out

View File

@ -27,7 +27,7 @@ using dic_t = std::map<std::pair<uint32_t, uint8_t>, uint32_t>;
* \param c Caractère suivant la chaine de caractères \p nr_chaine * \param c Caractère suivant la chaine de caractères \p nr_chaine
* \return std::pair<bool, uint16_t> * \return std::pair<bool, uint16_t>
*/ */
std::pair<bool, uint32_t> const std::pair<bool, uint32_t>
dico(std::map<std::pair<uint32_t, uint8_t>, uint32_t> &t_dictionary, dico(std::map<std::pair<uint32_t, uint8_t>, uint32_t> &t_dictionary,
uint32_t t_nr_chaine, uint8_t t_c) { uint32_t t_nr_chaine, uint8_t t_c) {
if (t_nr_chaine == 0xFFFFFFFF) if (t_nr_chaine == 0xFFFFFFFF)

View File

@ -10,7 +10,7 @@
#include <map> #include <map>
/// \brief Recherche ou ajout de chaine dans le dictionnaire /// \brief Recherche ou ajout de chaine dans le dictionnaire
std::pair<bool, std::uint32_t> const std::pair<bool, std::uint32_t>
dico(std::map<std::pair<std::uint32_t, std::uint8_t>, std::uint32_t> dico(std::map<std::pair<std::uint32_t, std::uint8_t>, std::uint32_t>
&t_dictionary, &t_dictionary,
std::uint32_t t_nr_chaine, std::uint8_t t_c); std::uint32_t t_nr_chaine, std::uint8_t t_c);

View File

@ -4,11 +4,14 @@
*/ */
#include "compress.hh" #include "compress.hh"
#include "utf8.hh"
#include <cstdlib> #include <cstdlib>
#include <fstream>
using dict_t = std::map<std::pair<uint32_t, uint8_t>, uint32_t>; using dict_t = std::map<std::pair<uint32_t, uint8_t>, uint32_t>;
using ustring = std::basic_string<uint8_t>; // chaine non encodée using ustring = std::basic_string<uint8_t>; // chaine non encodée
using uvec = std::vector<uint32_t>; // chaine encodée using uvec = std::vector<uint32_t>; // chaine encodée
using std::printf;
/** /**
* La chaine de caractère \p text est lue caractère par caractère, et est et * La chaine de caractère \p text est lue caractère par caractère, et est et
@ -21,20 +24,26 @@ using uvec = std::vector<uint32_t>; // chaine encodée
* \param t_dictionary Dictionnaire de compression * \param t_dictionary Dictionnaire de compression
* \return std::vector<uint16_t> * \return std::vector<uint16_t>
*/ */
const uvec compress(const ustring &t_text, dict_t &t_dictionary) { const uvec lzw_compress(const ustring &t_text, dict_t &t_dictionary) {
std::puts("Compressing..."); std::puts("Compressing...");
uvec res{}; uvec res{};
uint32_t w = 0xFFFFFFFF; uint32_t w = 0xFFFFFFFF;
uint32_t len = 0; uint32_t len = 0;
constexpr size_t DICT_MAX = 7936; /* 12 bits */
#ifdef Debug #ifdef Debug
size_t progress = 0; size_t progress = 0;
#endif #endif
for (const auto &c : t_text) { for (const auto &c : t_text) {
++len; ++len;
#ifdef Debug #ifdef Debug
printf("\rprogress: %zu / %zu", progress++, t_text.size()); printf("\rprogress: %zu / %zu", ++progress, t_text.size());
#endif #endif
if (len > 9) {
if (/* len > LENGTH_MAX || */ t_dictionary.size() >= DICT_MAX ) {
res.push_back(static_cast<uint32_t>(w)); res.push_back(static_cast<uint32_t>(w));
w = c; w = c;
len = 0; len = 0;
@ -50,3 +59,61 @@ const uvec compress(const ustring &t_text, dict_t &t_dictionary) {
printf("\n"); printf("\n");
return res; return res;
} }
/**
* \brief function description
*
* Wrapper de la fonction \ref lzw_compress gérant l'ouverture, la lecture,
* l'écriture et la fermeture des fichiers dentrée 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 lextension `.lzw`.
*
* \param t_in_file Chemin vers le fichier dentrée
* \param t_out_file Chemin vers le fichier de sortie
*/
void compress(const std::string &t_in_file, const char *t_out_file) {
// Fichier dentrée
std::ifstream input_file{t_in_file};
// Fichier de sortie
FILE *out = nullptr;
if (t_out_file) {
out = fopen(t_out_file, "wb");
} else {
std::string out_name{t_in_file};
out_name.append(".lzw");
out = fopen(out_name.c_str(), "wb");
}
input_file.seekg(0, std::ios::end);
// string contenant le fichier dentrée
ustring str(static_cast<unsigned long>(input_file.tellg()),
static_cast<unsigned char>(0));
input_file.seekg(0, std::ios::beg);
// assignation du contenu du fichier à str
str.assign((std::istreambuf_iterator<char>(input_file)),
std::istreambuf_iterator<char>());
printf("Size of input file: %zu\n", str.size());
dict_t dictionary{};
const auto comp_str{lzw_compress(str, dictionary)};
printf("\n############################################\n");
printf(" Compressed!\n");
printf("############################################\n\n");
printf("Size of compressed string: %zu\n", comp_str.size());
printf("Size of dictionary: %zu\n", dictionary.size());
printf("Compression ratio: %.10f\n",
static_cast<double>(str.size() / comp_str.size()));
for(const auto c : comp_str)
write_utf8(out, c);
fclose(out);
input_file.close();
return;
}

View File

@ -11,8 +11,11 @@
/// \brief Compression d'une chaine de caractères /// \brief Compression d'une chaine de caractères
const std::vector<std::uint32_t> const std::vector<std::uint32_t>
compress(const std::basic_string<uint8_t> &t_text, lzw_compress(const std::basic_string<uint8_t> &t_text,
std::map<std::pair<std::uint32_t, std::uint8_t>, std::uint32_t> std::map<std::pair<std::uint32_t, std::uint8_t>, std::uint32_t>
&t_dictionary); &t_dictionary);
/// \brief Wrapper de \ref lzw_compress
void compress(const std::string &t_in_file, const char *t_out_file);
#endif /* LZW_SRC_COMPRESS_H_ */ #endif /* LZW_SRC_COMPRESS_H_ */

View File

@ -6,11 +6,8 @@
* *
*/ */
#include "utf8.hh"
#include "compress.hh" #include "compress.hh"
#include <fstream> #include "getopt.h"
#include <iostream>
#include <streambuf>
using std::printf; using std::printf;
using std::puts; using std::puts;
@ -31,53 +28,110 @@ using dic_t = std::map<std::pair<uint32_t, uint8_t>, uint32_t>;
using ustring = std::basic_string<uint8_t>; // chaine non encodée using ustring = std::basic_string<uint8_t>; // chaine non encodée
using uvec = std::vector<uint32_t>; // chaine encodée using uvec = std::vector<uint32_t>; // chaine encodée
void help() {
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-d\tdecompresses 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");
}
int main(int argc, char *argv[]) { int main(int argc, char *argv[]) {
if (argc < 2) { #ifdef Debug
puts("Usage: lzw <filename>");
return 1;
} else {
for (int i = 0; i < argc; ++i) for (int i = 0; i < argc; ++i)
printf("argv[%d] = %s\n", i, argv[i]); printf("argv[%d] = %s\n", i, argv[i]);
#endif
std::string input_path{};
std::string output_path{};
bool compressing = true;
while (1) {
int option_index = 0;
static struct option long_options[] = {
{"help", no_argument, nullptr, 'h'},
{"input", required_argument, nullptr, 'i'},
{"output", required_argument, nullptr, 'o'},
{"compress", no_argument, nullptr, 'c'},
{"uncompress", no_argument, nullptr, 'u'},
{0, 0, 0, 0}};
int c = getopt_long(argc, argv, "hi:o:cu", long_options, &option_index);
if (c == -1)
break;
switch (c) {
case 0: {
#ifdef Debug
printf("\noption %s", long_options[option_index].name);
if (optarg) {
printf(" with arg %s\n", optarg);
}
#endif
break;
}
case 'h': {
#ifdef Debug
printf("From main - option --help passed\n");
#endif
help();
return 0;
}
case 'i': {
#ifdef Debug
printf("From main - option --input with value '%s'\n", optarg);
#endif
input_path = optarg;
break;
}
case 'o': {
#ifdef Debug
printf("From main - option --output with value '%s'\n", optarg);
#endif
output_path = optarg;
break;
}
case 'c': {
#ifdef Debug
printf("From main - option --compress\n");
#endif
compressing = true;
break;
}
case 'u': {
#ifdef Debug
printf("From main - option --uncompress\n");
#endif
compressing = false;
break;
}
case '?': {
puts("Error: unknown parameter.");
#ifdef Debug
printf("From main - option -?\n");
#endif
help();
return 1;
}
}
} }
// fichier d'entrée if (input_path == "") {
std::ifstream t{argv[1]}; puts("Error: no input file specified");
return 2;
}
// fichier de sortie if (compressing) {
FILE *out = fopen("output.lzw", "wb"); compress(input_path, output_path.c_str());
} else {
// string contenant le fichier d'entrée puts("Not yet implemented :(");
ustring str{}; }
// réserver la mémoire dans le string pour le fichier à compresser
t.seekg(0, std::ios::end);
str.reserve(static_cast<unsigned long>(t.tellg()));
t.seekg(0, std::ios::beg);
// assigner au string les caractères du fichier
str.assign((std::istreambuf_iterator<char>(t)),
std::istreambuf_iterator<char>());
// test pour voir si tout est bien chargé
printf("Size of str: %zu\n", str.size());
// le dictionnaire, initialement vide.
dic_t dictionary{};
// compresssion du texte
const auto comp_str{compress(str, dictionary)};
printf("Compressed, size of compressed string: %zu\n", comp_str.size());
printf("Compression ratio: %.10f\n",
static_cast<double>(str.size()) / static_cast<double>(comp_str.size()));
printf("Number of custom words in the dictionary: %zu\n", dictionary.size());
for(const auto c : comp_str)
write_utf8(out, c);
fclose(out);
t.close();
return 0; return 0;
} }