public void abort()
{
_aborted = true;
_parser.abort();
_results.meta.aborted = true;
if (Papa.isFunction(_config.complete))
{
_config.complete(_results);
}
_input = "";
}
protected override Result _nextChunk()
{
if (base._finished)
{
return(null);
}
int size = base._config.chunkSize;
string chunk = size > 0 ? Papa.Substr(remaining, 0, size) : remaining;
remaining = size > 0 ? Papa.Substr(remaining, size) : "";
base._finished = String.IsNullOrEmpty(remaining);
return(base.parseChunk(chunk));
}
private void replaceConfig(Config config)
{
// Deep-copy the config so we can edit it
Config configCopy = Papa.copy(config);
//configCopy.chunkSize = parseInt(configCopy.chunkSize); // parseInt VERY important so we don't concatenate strings!
if (config.step == null && config.chunk == null)
{
configCopy.chunkSize = 0; // disable Range header if not streaming; bad values break IIS - see issue #196
}
this._handle = new ParserHandle(configCopy);
this._handle.streamer = this;
this._config = configCopy; // persist the copy to the caller
}
protected void _sendError(Error error)
{
if (Papa.isFunction(this._config.error))
{
this._config.error(error);
}
else if (Papa.IS_WORKER && this._config.error != null)
{
//global.postMessage({
// workerId: Papa.WORKER_ID,
// error: error,
// finished: false
//});
}
}
private string guessLineEndings(string input)
{
input = Papa.Substr(input, 0, 1024 * 1024); // max length 1 MB
string[] r = input.Split('\r');
if (r.Length == 1)
{
return("\n");
}
int numWithN = 0;
for (int i = 0; i < r.Length; i++)
{
if (r[i][0] == '\n')
{
numWithN++;
}
}
return(numWithN >= r.Length / 2 ? "\r\n" : "\r");
}
public Result parse(string input, int baseIndex = 0, bool ignoreLastRow = false)
{
// We don't need to compute some of these every time parse() is called,
// but having them in a more local scope seems to perform better
int inputLen = input.Length,
delimLen = delim.Length,
newlineLen = newline.Length,
commentsLen = comments.Length;
bool stepIsFunction = step != null;
string[] delimSplit = new string[] { delim },
newlineSplit = new string[] { newline };
// Returns an object with the results, errors, and meta.
Func <bool, Result> returnable = (stopped) =>
{
return(new Result()
{
data = data,
dataWithHeader = new List <Dictionary <string, string> >(),
errors = errors,
meta = new Meta()
{
delimiter = delim,
linebreak = newline,
aborted = aborted,
truncated = stopped,
cursor = lastCursor + baseIndex
}
});
};
// Executes the user's step function and resets data & errors.
Action doStep = () =>
{
step(returnable(false), null);
data = new List <List <string> >();
errors = new List <Error>();
};
Action <List <string> > pushRow = (newRow) =>
{
data.Add(newRow);
lastCursor = cursor;
};
// Appends the remaining input from cursor to the end into
// row, saves the row, calls step, and returns the results.
Func <string, Result> finish = (newValue) =>
{
if (ignoreLastRow)
{
return(returnable(false));
}
if (newValue == null)
{
newValue = Papa.Substr(input, cursor);
}
row.Add(newValue);
cursor = inputLen; // important in case parsing is paused
pushRow(row);
if (stepIsFunction)
{
doStep();
}
return(returnable(false));
};
//------------------------------------------------------------------------------------------------------------------------------------------------------
// Establish starting state
cursor = 0;
data = new List <List <string> >();
errors = new List <Error>();
row = new List <string>();
lastCursor = 0;
if (String.IsNullOrEmpty(input))
{
return(returnable(false));
}
if (fastMode == true || (fastMode != false && input.IndexOf(quoteChar) == -1))
{
string[] rows = input.Split(newlineSplit, StringSplitOptions.None);
for (int i = 0; i < rows.Length; i++)
{
string rowFast = rows[i];
cursor += rowFast.Length;
if (i != rows.Length - 1)
{
cursor += newline.Length;
}
else if (ignoreLastRow)
{
return(returnable(false));
}
if (!String.IsNullOrEmpty(comments) && Papa.Substr(rowFast, 0, commentsLen) == comments)
{
continue;
}
if (stepIsFunction)
{
data = new List <List <string> >();
pushRow(new List <string>(rowFast.Split(delimSplit, StringSplitOptions.None)));
doStep();
if (aborted)
{
return(returnable(false));
}
}
else
{
pushRow(new List <string>(rowFast.Split(delimSplit, StringSplitOptions.None)));
}
if (preview > 0 && i >= preview)
{
data = data.GetRange(0, preview);
return(returnable(true));
}
}
return(returnable(false));
}
int nextDelim = input.IndexOf(delim, cursor);
int nextNewline = input.IndexOf(newline, cursor);
// Appends the current row to the results. It sets the cursor
// to newCursor and finds the nextNewline. The caller should
// take care to execute user's step function and check for
// preview and end parsing if necessary.
Action <int> saveRow = (newCursor) =>
{
cursor = newCursor;
pushRow(row);
row = new List <string>();
nextNewline = input.IndexOf(newline, cursor);
};
// Parser loop
for (;;)
{
//[CR added so we never look behind the string]
if (input.Length <= cursor)
{
//System.Diagnostics.Debugger.Break();
break;
}
// Field has opening quote
if (input[cursor] == quoteChar)
{
// Start our search for the closing quote where the cursor is
int quoteSearch = cursor;
// Skip the opening quote
cursor++;
for (;;)
{
// Find closing quote
quoteSearch = input.IndexOf(quoteChar, quoteSearch + 1);
if (quoteSearch == -1)
{
if (!ignoreLastRow)
{
// No closing quote... what a pity
errors.Add(new Error()
{
type = "Quotes",
code = "MissingQuotes",
message = "Quoted field unterminated",
row = data.Count, // row has yet to be inserted
index = cursor
});
}
return(finish(null));
}
if (quoteSearch == inputLen - 1)
{
// Closing quote at EOF
string value = Regex.Replace(Papa.Substring(input, cursor, quoteSearch), quoteChar.ToString() + quoteChar.ToString(), quoteChar.ToString(), RegexOptions.Multiline);
return(finish(value));
}
// If this quote is escaped, it's part of the data; skip it
if (input[quoteSearch + 1] == quoteChar)
{
quoteSearch++;
continue;
}
if (input[quoteSearch + 1].ToString() == delim)
{
// Closing quote followed by delimiter
row.Add(Regex.Replace(Papa.Substring(input, cursor, quoteSearch), quoteChar.ToString() + quoteChar.ToString(), quoteChar.ToString(), RegexOptions.Multiline));
cursor = quoteSearch + 1 + delimLen;
nextDelim = input.IndexOf(delim, cursor);
nextNewline = input.IndexOf(newline, cursor);
break;
}
if (Papa.Substr(input, quoteSearch + 1, newlineLen) == newline)
{
// Closing quote followed by newline
row.Add(Regex.Replace(Papa.Substring(input, cursor, quoteSearch), quoteChar.ToString() + quoteChar.ToString(), quoteChar.ToString(), RegexOptions.Multiline));
saveRow(quoteSearch + 1 + newlineLen);
nextDelim = input.IndexOf(delim, cursor); // because we may have skipped the nextDelim in the quoted field
if (stepIsFunction)
{
doStep();
if (aborted)
{
return(returnable(false));
}
}
if (preview > 0 && data.Count >= preview)
{
return(returnable(true));
}
break;
}
}
continue;
}
// Comment found at start of new line
if (!String.IsNullOrEmpty(comments) && row.Count == 0 && Papa.Substr(input, cursor, commentsLen) == comments)
{
if (nextNewline == -1) // Comment ends at EOF
{
return(returnable(false));
}
cursor = nextNewline + newlineLen;
nextNewline = input.IndexOf(newline, cursor);
nextDelim = input.IndexOf(delim, cursor);
continue;
}
// Next delimiter comes before next newline, so we've reached end of field
if (nextDelim != -1 && (nextDelim < nextNewline || nextNewline == -1))
{
row.Add(Papa.Substring(input, cursor, nextDelim));
cursor = nextDelim + delimLen;
nextDelim = input.IndexOf(delim, cursor);
continue;
}
// End of row
if (nextNewline != -1)
{
row.Add(Papa.Substring(input, cursor, nextNewline));
saveRow(nextNewline + newlineLen);
if (stepIsFunction)
{
doStep();
if (aborted)
{
return(returnable(false));
}
}
if (preview > 0 && data.Count >= preview)
{
return(returnable(true));
}
continue;
}
break;
}
return(finish(null));
}
public Result parseChunk(string chunk)
{
// First chunk pre-processing
if (this.isFirstChunk && Papa.isFunction(this._config.beforeFirstChunk))
{
string modifiedChunk = this._config.beforeFirstChunk(chunk);
if (modifiedChunk != null)
{
chunk = modifiedChunk;
}
}
this.isFirstChunk = false;
// Rejoin the line we likely just split in two by chunking the file
string aggregate = this._partialLine + chunk;
this._partialLine = "";
Result results = this._handle.parse(aggregate, this._baseIndex, !this._finished);
if (this._handle.paused() || this._handle.aborted())
{
return(null);
}
int lastIndex = results.meta.cursor;
if (!this._finished)
{
this._partialLine = Papa.Substring(aggregate, lastIndex - this._baseIndex);
this._baseIndex = lastIndex;
}
if (results != null && results.data != null)
{
this._rowCount += results.data.Count;
}
bool finishedIncludingPreview = this._finished || (this._config.preview > 0 && this._rowCount >= this._config.preview);
if (Papa.IS_WORKER)
{
//global.postMessage({
// results: results,
// workerId: Papa.WORKER_ID,
// finished: finishedIncludingPreview
//});
}
else if (Papa.isFunction(this._config.chunk))
{
this._config.chunk(results, this._handle);
if (this._paused)
{
return(null);
}
results = null;
this._completeResults = null;
}
if (this._config.step == null && this._config.chunk == null)
{
this._completeResults.data = this._completeResults.data.Concat(results.data).ToList();
this._completeResults.dataWithHeader = this._completeResults.dataWithHeader.Concat(results.dataWithHeader).ToList();
this._completeResults.errors = this._completeResults.errors.Concat(results.errors).ToList();
this._completeResults.meta = results.meta;
}
if (finishedIncludingPreview && Papa.isFunction(this._config.complete) && (results == null || !results.meta.aborted))
{
this._config.complete(this._completeResults);
}
if (!finishedIncludingPreview && (results == null || !results.meta.paused))
{
this._nextChunk();
}
return(results);
}
public static string Substr(this string input, int startIndex)
{
return(Papa.Substring(input, startIndex));
}
// Parses input. Most users won't need, and shouldn't mess with, the baseIndex
// and ignoreLastRow parameters. They are used by streamers (wrapper functions)
// when an input comes in multiple chunks, like from a file.
public Result parse(string input, int baseIndex = 0, bool ignoreLastRow = false)
{
Func <bool> needsHeaderRow = () =>
{
return(_config.header && _fields.Count == 0);
};
Action fillHeaderFields = () =>
{
if (_results == null || _results.data.Count == 0)
{
return;
}
for (int i = 0; needsHeaderRow() && i < _results.data.Count; i++)
{
for (int j = 0; j < _results.data[i].Count; j++)
{
_fields.Add(_results.data[i][j]);
}
}
_results.data.RemoveRange(0, 1);
};
Func <Result> applyHeaderAndDynamicTyping = () =>
{
if (_results == null || (!_config.header && !_config.dynamicTyping))
{
return(_results);
}
for (int i = 0; i < _results.data.Count; i++)
{
Dictionary <string, string> rowWithHeader = new Dictionary <string, string>();
int j;
for (j = 0; j < _results.data[i].Count; j++)
{
//[TODO]
//if (_config.dynamicTyping)
//{
// var value = _results.data[i][j];
// if (value == "true" || value == "TRUE")
// _results.data[i][j] = true;
// else if (value == "false" || value == "FALSE")
// _results.data[i][j] = false;
// else
// _results.data[i][j] = tryParseFloat(value);
//}
if (_config.header)
{
if (j >= _fields.Count)
{
if (!rowWithHeader.ContainsKey("__parsed_extra"))
{
rowWithHeader.Add("__parsed_extra", "");
}
rowWithHeader["__parsed_extra"] += _results.data[i][j];
//[CR we can not simply put an Array into __parsed_extra, so juste pipe it]
if (j < _results.data[i].Count - 1)
{
rowWithHeader["__parsed_extra"] += "|";
}
}
else
{
rowWithHeader[_fields[j]] = _results.data[i][j];
}
}
}
if (_config.header)
{
_results.dataWithHeader.Add(rowWithHeader); //[CR we are not overwriting _results.data here but instead fill another List]
if (j > _fields.Count)
{
addError("FieldMismatch", "TooManyFields", "Too many fields: expected " + _fields.Count + " fields but parsed " + j, i);
}
else if (j < _fields.Count)
{
addError("FieldMismatch", "TooFewFields", "Too few fields: expected " + _fields.Count + " fields but parsed " + j, i);
}
}
}
if (_config.header && _results.meta != null)
{
_results.meta.fields = _fields;
}
return(_results);
};
Func <Result> processResults = () =>
{
if (_results != null && _delimiterError)
{
addError("Delimiter", "UndetectableDelimiter", "Unable to auto-detect delimiting character; defaulted to '" + Papa.DefaultDelimiter + "'");
_delimiterError = false;
}
if (_config.skipEmptyLines)
{
for (int i = 0; i < _results.data.Count; i++)
{
if (_results.data[i].Count == 1 && _results.data[i][0] == "")
{
_results.data.RemoveRange(i--, 1);
}
}
}
if (needsHeaderRow())
{
fillHeaderFields();
}
return(applyHeaderAndDynamicTyping());
};
//------------------------------------------------------------------------------------------------------------------------------------------------------
if (String.IsNullOrEmpty(_config.newline))
{
_config.newline = guessLineEndings(input);
}
_delimiterError = false;
if (String.IsNullOrEmpty(_config.delimiter))
{
DelimiterResult delimGuess = guessDelimiter(input);
if (delimGuess.successful)
{
_config.delimiter = delimGuess.bestDelimiter;
}
else
{
_delimiterError = true; // add error after parsing (otherwise it would be overwritten)
_config.delimiter = Papa.DefaultDelimiter;
}
_results.meta.delimiter = _config.delimiter;
}
if (_config.quoteChar == Char.MinValue)
{
if (Papa.Substr(input, 0, 1) == "'" && Papa.Substr(input, input.IndexOf(_config.delimiter, 0) - 1, 1) == "'")
{
_config.quoteChar = '\'';
}
else
{
_config.quoteChar = '"';
}
}
Config parserConfig = Papa.copy(_config);
if (_config.preview > 0 && _config.header)
{
parserConfig.preview++; // to compensate for header row
}
if (Papa.isFunction(_config.step))
{
Action <Result, ParserHandle> userStep = _config.step;
parserConfig.step = (results, parser) =>
{
_results = results;
if (needsHeaderRow())
{
processResults();
}
else // only call user's step function after header row
{
processResults();
// It's possbile that this line was empty and there's no row here after all
if (_results.data.Count == 0)
{
return;
}
_stepCounter += results.data.Count;
if (parserConfig.preview > 0 && _stepCounter > parserConfig.preview)
{
_parser.abort();
}
else
{
userStep(_results, this);
}
}
};
}
//----------------------------------------------------------------------
_input = input;
_parser = new Parser(parserConfig);
_results = _parser.parse(_input, baseIndex, ignoreLastRow);
processResults();
if (_paused)
{
return new Result()
{
meta = new Meta()
{
paused = true
}
}
}
;
else if (_results != null)
{
return(_results);
}
else
{
return new Result()
{
meta = new Meta()
{
paused = false
}
}
};
}
public void pause()
{
_paused = true;
_parser.abort();
_input = Papa.Substr(_input, _parser.getCharIndex());
}
