/**
* Try to open .ssl file and read the header line. Return false if
* the file cannot be opened or if the header is incorrect.
*/
bool SslReader::openSsl(){
Verbosity::debug("Opening ssl File.");
sslFile_.open(sslName_.c_str());
if( !sslFile_.is_open() ){
throw BlibException(true, "Could not open ssl file '%s'.",
sslName_.c_str());
}
// confirm that header looks correct
string line;
getline(sslFile_, line);
const char* header = "file\tscan\tcharge\tsequence\tmodifications";
if( line.compare(0, strlen(header), header) ){
throw BlibException(false,
"SSL header is not correct. Should be '%s'.",
header);
}
return true;
}
/**
* Converts a position string from a mqpar file into a MAXQUANT_MOD_POSITION.
*/
MaxQuantModification::MAXQUANT_MOD_POSITION MaxQuantModReader::stringToPosition(string positionString)
{
const char* positionStringChars = positionString.c_str();
if (isIElement(positionStringChars, "anywhere"))
{
return MaxQuantModification::ANYWHERE;
}
else if (isIElement(positionStringChars, "proteinNterm"))
{
return MaxQuantModification::PROTEIN_N_TERM;
}
else if (isIElement(positionStringChars, "proteinCterm"))
{
return MaxQuantModification::PROTEIN_C_TERM;
}
else if (isIElement(positionStringChars, "anyNterm"))
{
return MaxQuantModification::PROTEIN_C_TERM;
}
else if (isIElement(positionStringChars, "anyCterm"))
{
return MaxQuantModification::PROTEIN_C_TERM;
}
else if (isIElement(positionStringChars, "notNterm"))
{
return MaxQuantModification::NOT_N_TERM;
}
else if (isIElement(positionStringChars, "notCterm"))
{
return MaxQuantModification::NOT_C_TERM;
}
else
{
throw BlibException(false, "Invalid position value: %s",
positionStringChars);
}
}
bool SAXHandler::parse()
{
FILE* pfIn = fopen(m_strFileName_.data(), "r");
if (pfIn == NULL) {
throw BlibException(true, "Failed to open input file '%s'.",
m_strFileName_.c_str());
}
bool success = true;
string message;
try {
// HACK!! I have no idea why this is, but without this string
// declaration, the MSVC optimizer removes the catch block below.
// Very mysterious.
string temp;
char buffer[8192];
int readBytes = 0;
while (success && (readBytes = (int) fread(buffer, 1, sizeof(buffer), pfIn)) != 0) {
success = (XML_Parse(m_parser_, buffer, readBytes, false) != 0);
}
success = success && (XML_Parse(m_parser_, buffer, 0, true) != 0);
}
catch(string thrown_msg) { // from parsers
message = thrown_msg;
success = false;
}
catch(BlibException e) { // probably from BuildParser
if( e.hasFilename() ) {
throw e;
} else {
message = e.what();
success = false;
}
}
catch(std::exception e) { // other runtime errors (e.g. memory)
message = e.what();
success = false;
}
fclose(pfIn);
if (!success) {
XML_Error error = XML_GetErrorCode(m_parser_);
int lineNum = XML_GetCurrentLineNumber(m_parser_);
ostringstream stringBuilder(ostringstream::out);
stringBuilder << m_strFileName_
<< "(line " << lineNum
<< "): " << message << flush;
if (message.length() == 0) {
switch (error) {
case XML_ERROR_SYNTAX:
stringBuilder << "Syntax error parsing XML.";
break;
case XML_ERROR_INVALID_TOKEN:
stringBuilder << "Invalid token error parsing XML.";
break;
case XML_ERROR_UNCLOSED_TOKEN:
stringBuilder << "Unclosed token error parsing XML.";
break;
case XML_ERROR_NO_ELEMENTS:
stringBuilder << "No elements error parsing XML.";
break;
case XML_ERROR_TAG_MISMATCH:
stringBuilder << "Tag mismatch error parsing XML.";
break;
case XML_ERROR_DUPLICATE_ATTRIBUTE:
stringBuilder << "Duplicate attribute error parsing XML.";
break;
case XML_ERROR_UNKNOWN_ENCODING:
case XML_ERROR_INCORRECT_ENCODING:
stringBuilder << "Unknown or incorrect encoding XML error.";
break;
case XML_ERROR_UNCLOSED_CDATA_SECTION:
stringBuilder << "Unclosed data section error parsing XML.";
break;
default:
stringBuilder << "XML parsing error.";
break;
}
}
string er_msg = stringBuilder.str();
throw BlibException(true, er_msg.c_str());
}
return true;
}
/**
* Called by saxhandler when the closing tag of an xml element is
* reached. Change the state based on the tag and the current state.
*/
void MaxQuantModReader::endElement(const XML_Char* name)
{
// modifications.xml mode
if (modBank_ && !fixedMods_)
{
if (isIElement("modification", name))
{
if (curMod_.massDelta != 0.0)
{
modBank_->insert(curMod_);
}
state_ = ROOT_STATE;
}
else if (state_ == READING_POSITION)
{
curMod_.position = stringToPosition(charBuf_);
state_ = MODIFICATION_TAG;
}
}
// mqpar.xml mode
else if (!modBank_ && fixedMods_)
{
if (isIElement("fixedModifications", name))
{
state_ = ROOT_STATE;
}
else if (state_ == READING_FIXED_MODIFICATION)
{
fixedMods_->insert(charBuf_);
state_ = FIXED_MODIFICATIONS_TAG;
}
else if (isIElement("filePaths", name))
{
state_ = ROOT_STATE;
}
else if (state_ == READING_FILEPATH)
{
string rawBaseName(charBuf_);
size_t lastSlash = rawBaseName.find_last_of("/\\");
if (lastSlash != string::npos)
{
rawBaseName = rawBaseName.substr(lastSlash + 1);
}
size_t extensionBegin = rawBaseName.find_last_of(".");
if (extensionBegin != string::npos)
{
rawBaseName.erase(extensionBegin);
}
MaxQuantLabels newLabelingStates(rawBaseName);
labelBank_->push_back(newLabelingStates);
state_ = FILEPATHS_TAG;
}
else if (isIElement("paramGroupIndices", name))
{
// check that each raw file had a corresponding param group index
if (paramGroupIndices_.size() != labelBank_->size())
{
throw BlibException(false, "Number of raw files (%d) did not match "
"number of paramGroupIndices (%d).",
labelBank_->size(), paramGroupIndices_.size());
}
state_ = ROOT_STATE;
}
else if (state_ == READING_PARAMGROUPINDEX)
{
paramGroupIndices_.push_back(lexical_cast<int>(charBuf_));
state_ = PARAMGROUPINDICES_TAG;
}
else if (isIElement("parameterGroups", name) && !paramGroupIndices_.empty())
{
// check that each raw file had a corresponding parameterGroup
for (vector<int>::iterator i = paramGroupIndices_.begin(); i != paramGroupIndices_.end(); ++i)
{
if (*i > rawIndex_)
{
throw BlibException(false, "Parameter group index %d was outside the range of "
"of parameter groups (%d).", *i, rawIndex_ + 1);
}
}
}
else if (isIElement("GroupParams", name))
{
// check that each raw file had a corresponding groupParam
if (--groupParams_ == 0 &&
(rawIndex_ + 1 != (int)labelBank_->size()))
{
throw BlibException(false, "Number of raw files (%d) did not match "
"number of label sets (%d).", labelBank_->size(), rawIndex_ + 1);
}
}
else if (state_ == LABELS_TAG &&
(isIElement("labels", name) || isIElement("labelMods", name)))
{
state_ = ROOT_STATE;
}
else if (state_ == READING_LABEL)
{
vector<string> newLabelSubset;
// split multiple labels (e.g. "Arg6; Lys4")
split(newLabelSubset, charBuf_, is_any_of(";"));
// trim whitespace
for_each(newLabelSubset.begin(), newLabelSubset.end(),
boost::bind(&boost::trim<string>, _1, std::locale()));
if (!paramGroupIndices_.empty())
{
// Assign label to all raw files with this parameter index
for (size_t i = 0; i < paramGroupIndices_.size(); ++i)
{
if (paramGroupIndices_[i] == rawIndex_)
{
(*labelBank_)[i].addModsStrings(newLabelSubset);
Verbosity::debug("Adding to labelBank_[%d] : %s", i, charBuf_.c_str());
}
}
}
else
{
(*labelBank_)[rawIndex_].addModsStrings(newLabelSubset);
}
state_ = LABELS_TAG;
}
}
}
