//--------------------------------------------------------------
bool ofxEditorSyntax::loadFile(const std::string& xmlFile) {
std::string path = ofToDataPath(xmlFile);
ofXml xml;
if(!xml.load(path)) {
ofLogError("ofxEditorSyntax") << "couldn't load \""
<< ofFilePath::getFileName(xmlFile) << "\"";
return false;
}
auto root = xml.getChild("syntax");
if(!root) {
ofLogWarning("ofxEditorSyntax") << "root xml tag not \"syntax\", ignoring";
return false;
}
clear();
for(auto & child : root.getChildren()) {
if(child.getName() == "lang") {setLang(child.getValue());}
else if(child.getName() == "files") {
for(auto & file : child.getChildren()) {
if(file.getName() == "ext") {addFileExt(file.getValue());}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown files xml tag \"" << file.getName() << "\"";
}
}
}
else if(child.getName() == "singlecomment") {singleLineComment = string_to_wstring(child.getValue());}
else if(child.getName() == "multicomment") {
auto begin = child.getChild("begin");
auto end = child.getChild("end");
if(begin) {multiLineCommentBegin = string_to_wstring(begin.getValue());}
if(end) {multiLineCommentBegin = string_to_wstring(end.getValue());}
}
else if(child.getName() == "stringliteral") {
auto begin = child.getChild("begin");
auto end = child.getChild("end");
if(begin) {stringLiteralBegin = string_to_wstring(begin.getValue());}
if(end) {stringLiteralEnd = string_to_wstring(end.getValue());}
}
else if(child.getName() == "preprocessor") {preprocessor = string_to_wstring(child.getValue());}
else if(child.getName() == "hexliteral") {
std::string b = child.getValue();
if(b == "true") {setHexLiteral(true);}
else if(b == "false") {setHexLiteral(false);}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown xml bool string \"" << b << "\"";
}
}
else if(child.getName() == "operator") {operatorChars = string_to_wstring(child.getValue());}
else if(child.getName() == "punctuation") {punctuationChars = string_to_wstring(child.getValue());}
else if(child.getName() == "words") {
for(auto &word : child.getChildren()) {
if(word.getName() == "keyword") {setWord(word.getValue(), KEYWORD);}
else if(word.getName() == "typename") {setWord(word.getValue(), TYPENAME);}
else if(word.getName() == "function") {setWord(word.getValue(), FUNCTION);}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown words xml tag \"" << word.getName() << "\"";
}
}
}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown xml tag \"" << child.getName() << "\"";
}
}
return true;
}
//--------------------------------------------------------------
bool ofxEditorSyntax::loadFile(const string& xmlFile) {
string path = ofToDataPath(xmlFile);
ofXml xml;
if(!xml.load(path)) {
ofLogError("ofxEditorSyntax") << "couldn't load \""
<< ofFilePath::getFileName(xmlFile) << "\"";
return false;
}
xml.setToParent();
if(!xml.exists("syntax")) {
ofLogWarning("ofxEditorSyntax") << "root xml tag not \"syntax\", ignoring";
return false;
}
xml.setTo("syntax");
int numTags = xml.getNumChildren();
clear();
for(int i = 0; i < numTags; ++i) {
xml.setToChild(i);
if(xml.getName() == "lang") {setLang(xml.getValue());}
else if(xml.getName() == "files") {
int numExts = xml.getNumChildren();
for(int e = 0; e < numExts; ++e) {
xml.setToChild(e);
if(xml.getName() == "ext") {addFileExt(xml.getValue());}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown files xml tag \"" << xml.getName() << "\"";
}
xml.setToParent();
}
}
else if(xml.getName() == "singlecomment") {singleLineComment = string_to_wstring(xml.getValue());}
else if(xml.getName() == "multicomment") {
if(xml.exists("begin")) {multiLineCommentBegin = string_to_wstring(xml.getValue("begin"));}
if(xml.exists("end")) {multiLineCommentBegin = string_to_wstring(xml.getValue("end"));}
}
else if(xml.getName() == "preprocessor") {preprocessor = string_to_wstring(xml.getValue());}
else if(xml.getName() == "hexliteral") {
string b = xml.getValue();
if(b == "true") {setHexLiteral(true);}
else if(b == "false") {setHexLiteral(false);}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown xml bool string \"" << b << "\"";
}
}
else if(xml.getName() == "operator") {operatorChars = string_to_wstring(xml.getValue());}
else if(xml.getName() == "punctuation") {punctuationChars = string_to_wstring(xml.getValue());}
else if(xml.getName() == "words") {
int numWords = xml.getNumChildren();
for(int w = 0; w < numWords; ++w) {
xml.setToChild(w);
if(xml.getName() == "keyword") {setWord(xml.getValue(), KEYWORD);}
else if(xml.getName() == "typename") {setWord(xml.getValue(), TYPENAME);}
else if(xml.getName() == "function") {setWord(xml.getValue(), FUNCTION);}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown words xml tag \"" << xml.getName() << "\"";
}
xml.setToParent();
}
}
else {
ofLogWarning("ofxEditorSyntax") << "ignoring unknown xml tag \"" << xml.getName() << "\"";
}
xml.setToParent();
}
xml.clear();
return true;
}
//--------------------------------------------------------------
void ofxEditorSyntax::addFileExt(const std::vector<std::string> &exts) {
for(int i = 0; i < exts.size(); ++i) {
addFileExt(exts[i]);
}
}
char *uploadfile(char *filename, char *lookForSimilar, char *lang) {
#ifndef CAN_MAGIC
o_log(ERROR, "Unable to determin the file type, aborting.");
return NULL;
#else
int width = 0, height = 0, itype = PLACE_HOLDER;
char *final_filename, *ocrText = NULL, *tmp;
#ifdef CAN_PDF
char *thumbext = NULL;
#else
#ifdef CAN_READODF
char *thumbext = NULL;
#endif /* CAN_READODF */
#endif /* CAN_PDF */
char *docid;
char *ftype;
char *datafile;
char *thumbfile = NULL;
PIX *pix;
datafile = o_printf("/tmp/%s.dat", filename);
magic_t cookie = magic_open(MAGIC_MIME_TYPE);
magic_load( cookie, NULL );
const char *t = magic_file( cookie, datafile );
ftype = o_strdup( t );
o_log( ERROR, "Uploaded file looks to be of type: %s", ftype );
magic_close( cookie );
// --------------------------------------
if( 0 == strcmp("application/pdf", ftype) ) {
itype = PDF_FILETYPE;
#ifdef CAN_PDF
thumbfile = o_printf("/tmp/%s.thumb", filename);
ocrText = parse_pdf( datafile, thumbfile ); // pdf_plug.cc [create thumbnail and return body text]
thumbext = o_strdup("jpg");
#endif /* CAN_PDF */
o_log( INFORMATION, "Processed PDF");
}
// --------------------------------------
else if( 0 == strcmp("application/vnd.oasis.opendocument.text", ftype) ) {
itype = ODF_FILETYPE;
#ifdef CAN_READODF
thumbfile = o_printf("/tmp/%s.thumb", filename);
get_odf_Thumb( datafile, thumbfile );
ocrText = get_odf_Text( datafile ); // odf_plug.c
thumbext = o_strdup("png");
#endif /* CAN_READODF */
o_log( INFORMATION, "Processed ODF doc");
}
// --------------------------------------
else if( 0 == strcmp("image/jpeg", ftype) ) {
itype = JPG_FILETYPE;
#ifdef CAN_OCR
PIX *pix_l;
if ( ( pix_l = pixRead( datafile ) ) == NULL) {
o_log(ERROR, "Could not load the image data into a PIX");
return NULL;
}
int depth;
pixGetDimensions( pix_l, &width, &height, &depth );
o_log(INFORMATION, "Convertion process: Loaded (depth: %d)", depth );
pix = pixScaleRGBToGrayFast( pix_l, 1, COLOR_GREEN );
pixDestroy( &pix_l );
if (pix == NULL ) {
o_log(ERROR,"Conversion process failed pixScaleRGBToGrayFast! skip ocr");
}
else {
o_log(INFORMATION, "Convertion process: Reduced depth to %d", pixGetDepth(pix));
ocrText = getTextFromImage(pix, 0, "eng");
}
#endif /* CAN_OCR */
o_log( INFORMATION, "Processed JPG doc");
}
// --------------------------------------
else {
free( ftype );
free( datafile );
o_log(ERROR, "unknown file type.");
return NULL;
}
free( ftype );
// Set a default OCR text string
if( ocrText == NULL ) {
ocrText = o_strdup( getString("LOCAL_ocr_default_text", lang ) );
}
// Save the record to the DB
o_log(DEBUGM, "Saving doc import record");
docid = addNewFileDoc(itype, width, height, ocrText); // ocrText get freed in this method
// Move the main datafile to the file store location
final_filename = o_printf("%s/scans/%s", BASE_DIR, docid); // none image imported docs, are stored with no "_x" postfix.
if( itype == JPG_FILETYPE ) {
conCat(&final_filename, "_1");
}
addFileExt(&final_filename, itype);
fcopy(datafile, final_filename);
o_log( DEBUGM, "Moved data file");
// The original file will be unlinked by the HTTPD process
free(datafile);
// Move any thumbnail image to the file store location
if( thumbfile ) {
free(final_filename); // This currently holds the main PDG or ODF file.
final_filename = o_printf("%s/scans/%s_thumb.%s", BASE_DIR, docid, thumbext); // any thumbnails are postfixed with "_thumb"
fcopy(thumbfile, final_filename);
o_log( DEBUGM, "Moved thumbnail file");
unlink(thumbfile);
free(thumbfile);
free(thumbext);
#ifdef CAN_PHASH
o_log( DEBUGM, "About to perform pHash on file");
unsigned long long hash = getImagePhash_fn( final_filename );
savePhash( atoi(docid), hash );
#endif /* CAN_PHASH */
}
else {
#ifdef CAN_PHASH
o_log( DEBUGM, "About to perform pHash on pix");
unsigned long long hash = getImagePhash_px( pix );
savePhash( atoi(docid), hash );
#endif /* CAN_PHASH */
pixDestroy( &pix );
}
free(final_filename);
// Should we look for a similar doc, on opening?
char *findSim = "";
#ifdef CAN_PHASH
if( lookForSimilar != (void *)NULL ) {
findSim = "&findSimilar=1";
}
#endif /* CAN_PHASH */
// Open the document for editing.
tmp = o_printf("<html><HEAD><META HTTP-EQUIV=\"refresh\" CONTENT=\"0;URL=/opendias/docDetail.html?docid=%s%s\"></HEAD><body></body></html>", docid, findSim);
free(docid);
return tmp;
#endif /* CAN_MAGIC */
}
