int main(int argc, char *argv[]) {
DIR *fromdir;
char fullFromName[MAXPATHLEN];
char fullDestName[MAXPATHLEN];
char baseSrcPath[MAXPATHLEN];
char baseFromPath[MAXPATHLEN];
char baseToPath[MAXPATHLEN];
struct stat st;
int allowAppend = 0;
int allowLongerDest = 0;
char archiveExtension[MAXPATHLEN];
baseSrcPath[0] = '\0';
archiveExtension[0] = '\0';
argv++; argc--;
while (argc && argv[0][0] == '-') {
if (strlen(argv[0]) != 2) printUsageAndDie();
switch (argv[0][1]) {
case 's':
argv++; argc--;
if (!argc || argv[0][0] == '-') printUsageAndDie();
strcpy(baseSrcPath,argv[0]);
printf("Source directory set to %s\n",baseSrcPath);
argv++; argc--;
break;
case 'a':
printf("Allowing append\n");
allowAppend = 1;
argv++; argc--;
break;
case 'l':
printf("Allowing destination to be longer than source\n");
allowLongerDest = 1;
argv++; argc--;
break;
case 'r':
argv++; argc--;
if (!argc || argv[0][0] == '-') printUsageAndDie();
sprintf(archiveExtension,".%s",argv[0]);
printf("Archive extension set to %s\n",archiveExtension);
argv++; argc--;
break;
default:
fprintf(stderr,"Error: Unknown option %s\n",argv[0]);
printUsageAndDie();
break;
}
}
if (argc != 2) {
fprintf(stderr,"Error: Didn't end up with two dirs\n");
printUsageAndDie();
}
if ((fromdir = opendir(argv[0])) == NULL) {
fprintf(stderr,"Error: Couldn't open source perl directory %s\n",argv[0]);
exit(1);
}
if ((opendir(argv[1])) != NULL) {
if (!allowAppend) {
fprintf(stderr,"Error: Destination perl directory already exists\n");
exit(1);
}
} else if (allowAppend) {
fprintf(stderr,"Error: -a (append) specified but destination dir doesn't exist\n");
}
/*
if (!realpath(argv[0],fullFromName)) {
fprintf(stderr,"Error: Couldn't get realpath to fromdir\n");
exit(1);
}
if (!realpath(argv[1],fullDestName)) {
fprintf(stderr,"Error: Couldn't get realpath to destdir\n");
exit(1);
}
*/
strcpy(fullFromName,argv[0]);
strcpy(fullDestName,argv[1]);
if (!allowAppend) {
stat(fullFromName,&st);
if (mkdir(fullDestName,st.st_mode)) {
fprintf(stderr, "Error: Failed making directory %s\n",fullDestName);
exit(1);
}
}
strcpy(baseFromPath,fullFromName);
strcpy(baseToPath,fullDestName);
if (baseSrcPath[0]=='\0') {
strcpy(baseSrcPath,fullFromName);
}
if (strlen(baseSrcPath) < strlen(baseToPath) && !allowLongerDest) {
fprintf(stderr,"Error: Length of to path can't be longer than from path\n");
exit(1);
}
descendAndDo(fromdir,fullFromName,fullDestName,
baseSrcPath,baseFromPath,baseToPath,
allowAppend,archiveExtension,doFuncInp);
}
int main(int argc, char **argv)
try {
if (argc != 3) {
printUsageAndDie(argv[0]);
}
std::ifstream origFile(argv[1]);
if (! origFile) {
throw std::runtime_error("Cannot open " + std::string(argv[1]));
}
std::ifstream splitFile(argv[2]);
if (! splitFile) {
throw std::runtime_error("Cannot open " + std::string(argv[2]));
}
unsigned startPos = 0;
int origPos = 0;
std::string line;
unsigned lineNo = 0;
while (std::getline(splitFile, line)) {
++lineNo;
unsigned endPos = startPos + line.size();
unsigned origStartPos;
if (line.size() == 0) { /// empty line
origStartPos = origPos;
}
else {
/// collect heading spaces
std::string headingWs;
for (std::string::const_iterator ch = line.begin();
ch != line.end(); ++ch) {
if (std::isspace(*ch)) {
headingWs += *ch;
}
else {
break;
}
}
// if (line[0] == ' ') {
// throw std::runtime_error(
// "Error: Split text contains a heading space");
// }
/// skip newlines in the original file
char origChar;
while (origFile.get(origChar)) {
++origPos;
if (origChar != '\n') {
origFile.unget();
--origPos;
break;
}
}
/// collect spaces in the original file
std::string origWs;
while (origFile.get(origChar)) {
++origPos;
if (origChar == ' ' || origChar == '\t') {
origWs += origChar;
}
else {
origFile.unget();
--origPos;
break;
}
}
#if 0
/// skip spaces and newlines in the original file
char origChar;
do {
if (! origFile.get(origChar)) {
throw std::runtime_error(
"Split/original text do not match");
}
++origPos;
} while (origChar == ' ' || origChar == '\n');
origFile.unget();
--origPos;
#endif
/// check if the heading space part has a corresponding region
/// in the original file
if (origWs.size() < headingWs.size()) {
throw std::runtime_error("Split/original text do not match");
}
unsigned skipLen = origWs.size() - headingWs.size();
if (! headingWs.empty() && headingWs != origWs.substr(skipLen)) {
throw std::runtime_error("Split/original text do not match");
}
origStartPos = origPos - headingWs.size();
/// Check the identity for the rest of the line
for (unsigned i = headingWs.size(); i < line.size(); ++i) {
if (! origFile.get(origChar) || origChar != line[i]) {
std::string msg = errorMsgLong(
origPos, origChar, lineNo, i, line[i]);
throw std::runtime_error(msg);
}
++origPos;
}
}
std::cout << startPos << '\t' << endPos << '\t' << origStartPos
<< std::endl;
/// +1 to skip the new line character
startPos = endPos + 1;
}
return 0;
}
catch (std::runtime_error &e)
{
std::cerr << e.what() << std::endl;
exit(1);
}
catch (...) {
std::cerr << "Unknown exception" << std::endl;
exit(1);
}
int main(int argc, char** argv) {
if(argc < 4) {
printUsageAndDie();
}
FileHashGrouper fileHashGrouper;
bool enableHtmlOutput = false;
char* targetFolder;
char* algoArg;
char* thresholdArg;
float threshold = -1.0f;
HashAlgorithm selectedAlgorithm = DCT;
#ifdef DEBUG
std::cout << "Running in debug mode" << std::endl;
#endif
if(*argv[1] == '-') {
if(argc < 5)
printUsageAndDie();
targetFolder = argv[4];
algoArg = argv[2];
thresholdArg = argv[3];
if(argv[1][1] == 'a') {
#ifndef DEBUG
fileHashGrouper.thresholdRequiredForAll = false;
std::cout << "Threshold will be required to be satisfied for only ONE hash." << std::endl;
#else
std::cout << "Threshold behaviour argument ignored because of debug mode." << std::endl;
#endif
}
} else {
targetFolder = argv[3];
algoArg = argv[1];
thresholdArg = argv[2];
}
enableHtmlOutput = true;
std::cout << "HTML Output enabled." << std::endl;
std::string tmpSelectedAlgorithm(algoArg);
boost::algorithm::to_lower(tmpSelectedAlgorithm);
if(boost::iequals(tmpSelectedAlgorithm, "dct"))
selectedAlgorithm = DCT;
if(boost::iequals(tmpSelectedAlgorithm, "mh"))
selectedAlgorithm = MH;
if(boost::iequals(tmpSelectedAlgorithm, "radish"))
selectedAlgorithm = RADISH;
if(boost::iequals(tmpSelectedAlgorithm, "all"))
selectedAlgorithm = ALL;
fileHashGrouper.algorithm = selectedAlgorithm;
try {
threshold = boost::lexical_cast<int>(thresholdArg);
if(threshold > 100.0f || threshold < 1.0f)
throw 33; //No specific reason for 33, just felt like it
} catch (std::exception& e) {
std::cout << "error : threshold must be between 1 and 100\n\n";
printUsageAndDie();
}
std::cout << "Algorithm : " << tmpSelectedAlgorithm;
std::cout << ", Threshold : " << threshold << " %" << std::endl;
fileHashGrouper.threshold = 1-threshold/100;
std::cout << "Scanning target folder : " << targetFolder << std::endl;
std::vector<FileHash> fileList;
listImages(bf::path(targetFolder), fileList);
std::cout << "Done. " << fileList.size() << " file(s) to scan.\nScanning ...\n";
int scanned = 0;
for(int i = 0; i < fileList.size(); i++) {
fileList[i].compute(selectedAlgorithm);
printProgressBar(std::floor(((i+1)/static_cast<float>(fileList.size()))*100));
}
std::cout << std::endl;
ofstream outputHtml;
if(enableHtmlOutput) {
outputHtml.open("results.html");
if(!outputHtml.is_open()) {
std::cout << "Error : Could not open 'result.html' for writing, disabling HTML output.";
std::cout << std::endl;
enableHtmlOutput = false;
}
outputHtml << "<!DOCTYPE html>\n";
outputHtml << "<html>\n\t<head>\n\t\t<title>Image comparaison result</title>\n";
outputHtml << "\t\t<link href='style.css' rel='stylesheet' type='text/css'>\n\t</head>\n";
outputHtml << "\t<body>\n";
outputHtml << "\t\t<h1>Image comparaison result for " << targetFolder << "</h1>\n";
outputHtml << "\t\t<h2>" << fileList.size() << " image(s)</h2>\n";
outputHtml << "\t\t<table>\n";
}
std::vector<FileHashGroup> groups = fileHashGrouper.computeGroups(fileList);
for(int i = 0; i < groups.size(); i++) {
if(enableHtmlOutput) {
outputHtml << "\t\t\t<tr>\n";
outputHtml << "\t\t\t\t<td>\n";
groups[i][0].file.printHtml(outputHtml, 4);
outputHtml << "\t\t\t\t</td>\n";
}
#ifdef DEBUG
std::cout << groups[i][0].file.path() << " ";
#endif
for(int j = 1; j < groups[i].elements().size(); j++) {
if(enableHtmlOutput) {
outputHtml << "\t\t\t\t<td>\n";
groups[i][j].file.printHtml(outputHtml, 5);
outputHtml << "\t\t\t\t<br/><span class='similarity'>";
outputHtml << groups[i][j].similarity.percentage();
outputHtml << " %</span>\n";
outputHtml << "\t\t\t\t</td>\n";
}
#ifdef DEBUG
std::cout << groups[i][j].file.path() << " ";
std::cout << "dct : " << groups[i][j].similarity.dct << " | ";
std::cout << "mh : " << groups[i][j].similarity.mh << " | ";
std::cout << "radish : " << groups[i][j].similarity.radish << " | ";
std::cout << std::endl;
#endif
}
if(enableHtmlOutput)
outputHtml << "\t\t\t</tr>\n";
}
if(enableHtmlOutput) {
outputHtml << "\t\t</table>\n\t</body>\n</html>";
outputHtml.close();
std::cout << "Results saved to 'results.html'" << std::endl;
}
std::cout << "Done." << std::endl;
return 0;
}
