/// <summary>
/// The main function to run the program
/// </summary>
/// <param name="args">Passed in arguments to the program</param>
public static int Main(string[] args)
{
var asmName = typeof(Program).GetTypeInfo().Assembly.GetName();
var exeName = Path.GetFileName(Assembly.GetExecutingAssembly().Location); // Alternatively: System.AppDomain.CurrentDomain.FriendlyName
var version = MetaboliteValidatorOptions.GetAppVersion();
var parser = new CommandLineParser <MetaboliteValidatorOptions>(asmName.Name, version)
{
ProgramInfo = "This program reads metabolites in a .tsv file and pushes new information " + Environment.NewLine +
"to the git repository at https://github.com/PNNL-Comp-Mass-Spec/MetabolomicsCCS",
ContactInfo = "Program written by Ryan Wilson and Matthew Monroe for the Department of Energy (PNNL, Richland, WA) in 2017" +
Environment.NewLine + Environment.NewLine +
"E-mail: [email protected] or [email protected]" + Environment.NewLine +
"Website: https://github.com/PNNL-Comp-Mass-Spec/ or https://panomics.pnnl.gov/ or https://www.pnnl.gov/integrative-omics",
UsageExamples =
{
exeName + " NewMetabolites.tsv",
exeName + " NewMetabolites.tsv -i",
exeName + " NewMetabolites.tsv -preview",
exeName + " NewMetabolites.tsv -user MyUsername -password *Dfw3gf"
}
};
var result = parser.ParseArgs(args);
var options = result.ParsedResults;
try
{
if (!result.Success)
{
if (parser.CreateParamFileProvided)
{
return(0);
}
// Delay for 1500 msec in case the user double clicked this file from within Windows Explorer (or started the program via a shortcut)
Thread.Sleep(1500);
return(-1);
}
if (!options.ValidateArgs(out var errorMessage))
{
parser.PrintHelp();
Console.WriteLine();
ConsoleMsgUtils.ShowWarning("Validation error:");
ConsoleMsgUtils.ShowWarning(errorMessage);
Thread.Sleep(1500);
return(-1);
}
options.OutputSetOptions();
}
catch (Exception e)
{
Console.WriteLine();
Console.Write($"Error running {exeName}");
Console.WriteLine(e.Message);
Console.WriteLine($"See help with {exeName} --help");
return(-1);
}
var program = new Program();
var success = program.ProcessMetabolites(options);
Console.WriteLine();
if (success)
{
Console.WriteLine("Processing complete");
}
else
{
Console.WriteLine("Processing failed");
}
System.Threading.Thread.Sleep(1500);
return(0);
}
/// <summary>
/// Initialization function that controls the program
/// </summary>
/// <param name="options">Processing options</param>
/// <returns>True on success, false if an error</returns>
private bool ProcessMetabolites(MetaboliteValidatorOptions options)
{
try
{
if (string.IsNullOrWhiteSpace(options.InputFile))
{
Console.WriteLine();
Console.WriteLine("Error, input file not defined");
return(false);
}
var inputFile = new FileInfo(options.InputFile);
if (!inputFile.Exists)
{
Console.WriteLine();
Console.WriteLine("Error, input file not found: " + inputFile.FullName);
return(false);
}
// init github api interaction with the repo and owner
var github = new Github("MetabolomicsCCS", "PNNL-Comp-Mass-Spec", options.Preview);
if (!string.IsNullOrEmpty(options.Username))
{
github.Username = options.Username;
if (!string.IsNullOrEmpty(options.Password))
{
if (options.Password.StartsWith("*"))
{
github.Password = MetaboliteValidatorOptions.DecodePassword(options.Password.Substring(1));
}
else
{
github.Password = options.Password;
}
}
}
// get main data file from github
var dataFile = github.GetFile("data/" + MASTER_TSV_FILE);
// parse the new data to append to current data
var fileToAppend = new DelimitedFileParser();
fileToAppend.ParseFile(inputFile.FullName, '\t');
Console.WriteLine();
Console.WriteLine("Found {0} records in local file {1}", fileToAppend.Count(), inputFile.Name);
// Update column names if necessary
UpdateHeaders(fileToAppend);
// parse the main data file from github
var mainFile = new DelimitedFileParser();
if (dataFile == null)
{
mainFile.SetDelimiter('\t');
mainFile.SetHeaders(fileToAppend.GetHeaders());
}
else
{
mainFile.ParseString(dataFile, '\t');
Console.WriteLine();
Console.WriteLine("Found {0} records in file {1} retrieved from GitHub", mainFile.Count(), MASTER_TSV_FILE);
Console.WriteLine();
}
// Update column names if necessary
UpdateHeaders(mainFile);
var duplicateRowCount = 0;
if (!options.IgnoreErrors)
{
// Get ids for Kegg and PubChem
var keggIds = fileToAppend.GetColumnAt("kegg").Where(x => !string.IsNullOrEmpty(x)).ToList();
var cidIds = fileToAppend.GetColumnAt("pubchem cid").Where(x => !string.IsNullOrEmpty(x)).ToList();
var mainCasIds = mainFile.GetColumnAt("cas").Where(x => !string.IsNullOrEmpty(x)).ToList();
// generate PubChem and Kegg utils
var pub = new PubchemUtil(cidIds.ToArray());
var kegg = new KeggUtil(keggIds.ToArray());
var file = new StreamWriter("ValidationApi.txt");
var dupRows = new DelimitedFileParser();
dupRows.SetHeaders(fileToAppend.GetHeaders());
dupRows.SetDelimiter('\t');
var warningRows = new DelimitedFileParser();
warningRows.SetHeaders(fileToAppend.GetHeaders());
warningRows.SetDelimiter('\t');
var missingKegg = new DelimitedFileParser();
missingKegg.SetHeaders(fileToAppend.GetHeaders());
missingKegg.SetDelimiter('\t');
var dataMap = fileToAppend.GetMap();
// compare fileToAppend to utils
for (var i = dataMap.Count - 1; i >= 0; i--)
{
Compound p = null;
CompoundData k = null;
if (!string.IsNullOrEmpty(dataMap[i]["pubchem cid"]))
{
p = pub.PubChemMap[int.Parse(dataMap[i]["pubchem cid"])];
}
if (!string.IsNullOrEmpty(dataMap[i]["kegg"]) && kegg.CompoundsMap.ContainsKey(dataMap[i]["kegg"]))
{
k = kegg.CompoundsMap[dataMap[i]["kegg"]];
}
if (mainCasIds.Contains(dataMap[i]["cas"]))
{
dupRows.Add(dataMap[i]);
fileToAppend.Remove(dataMap[i]);
}
else
{
if (k == null && CheckRow(dataMap[i], p, null))
{
missingKegg.Add(dataMap[i]);
}
else if (!CheckRow(dataMap[i], p, k))
{
// remove from list add to warning file
WriteContentToFile(file, dataMap[i], p, k, warningRows.Count() + 2);
warningRows.Add(dataMap[i]);
fileToAppend.Remove(dataMap[i]);
}
}
}
duplicateRowCount = dupRows.Count();
file.Close();
if (fileToAppend.Count() > 0)
{
Console.WriteLine("Validating data file with GoodTables");
var goodTables = new GoodTables(fileToAppend.ToString(true), SchemaUrl);
if (!goodTables.Response.success)
{
//foreach(var result in goodTables.Response.report.results)
//{
// fileToAppend.Remove(result["0"].result_context[0]);
//}
goodTables.OutputResponse(new StreamWriter(GOOD_TABLES_WARNING_FILE));
Console.WriteLine();
Console.WriteLine("GoodTables reports errors; see " + GOOD_TABLES_WARNING_FILE);
Console.WriteLine("Note that data with N/A in columns that expect a number will be flagged as an error by GoodTables; those errors can be ignored");
}
}
streamToFile(DUPLICATE_ROWS_FILE, dupRows);
streamToFile(WARNING_ROWS_FILE, warningRows);
streamToFile(MISSING_KEGG_FILE, missingKegg);
if (warningRows.Count() > 0)
{
Console.WriteLine();
Console.WriteLine("Warnings were encountered; see file " + WARNING_ROWS_FILE);
}
if (missingKegg.Count() > 0)
{
Console.WriteLine();
Console.WriteLine("Warnings were encountered; see file " + MISSING_KEGG_FILE);
}
}
else
{
Console.WriteLine();
Console.WriteLine("Ignoring validation, skipping to file upload.");
}
if (fileToAppend.Count() == 0)
{
Console.WriteLine();
Console.WriteLine("No new compounds were found; see {0} for the {1} skipped compounds", DUPLICATE_ROWS_FILE, duplicateRowCount);
}
else
{
// this will add the new data tsv to the existing tsv downloaded from github
var success = mainFile.Concat(fileToAppend);
if (!success)
{
// Concatenation of new records failed; do not upload
return(false);
}
// Start command line process for GoodTables
//
// string userDirPath = Environment.GetEnvironmentVariable("GOODTABLES_PATH");
// string commandLine = $"schema \"{options.InputFile}\" --schema \"{SchemaUrl}\"";
// string GoodTablesPath = $"{userDirPath}\\GoodTables";
//CommandLineProcess pro = new CommandLineProcess(GoodTablesPath, commandLine);
//// if error display errors and exit
//if (pro.Status.Equals(CommandLineProcess.StatusCode.Error))
//{
// Console.WriteLine($"GoodTables Validation error\n\n{pro.StandardOut}{pro.StandardError}\nExiting program please check that the data is valid.");
// Console.ReadKey();
// Environment.Exit(1);
//}
//// if the GoodTables.exe file isn't found display message and exit
//else if (pro.Status.Equals(CommandLineProcess.StatusCode.FileNotFound))
//{
// Console.WriteLine("File not found. Please make sure you have installed python and GoodTables.\n"
// +"Check that the folder path for GoodTables.exe is added to an environment variable named GOODTABLES_PATH.\n"
// +"Press any key to continue.");
// Console.ReadKey();
// Environment.Exit(1);
//}
//else
//{
// Console.WriteLine($"GoodTables validation\n\n{pro.StandardOut}");
//
// This will send the completed tsv back to github
github.SendFileAsync(mainFile.ToString(true), "data/" + MASTER_TSV_FILE);
// send Agilent file to github
github.SendFileAsync(mainFile.PrintAgilent(), "data/metabolitedataAgilent.tsv");
//}
}
return(true);
}
catch (Exception ex)
{
Console.WriteLine();
Console.WriteLine("Error processing data: " + ex.Message);
Console.WriteLine(StackTraceFormatter.GetExceptionStackTraceMultiLine(ex));
return(false);
}
}
