/// <summary>
/// Modal edit of molecule
/// </summary>
public static MoleculeMx Edit(
MoleculeMx mol,
MoleculeRendererType editorType = MoleculeRendererType.Unknown,
string title = "")
{
MoleculeMx mol2 = null;
try
{
if (editorType == MoleculeRendererType.Unknown) // need to assign editor type?
{
if (mol.IsChemStructureFormat)
{
editorType = MoleculeRendererType.Chemistry;
}
else if (mol.IsBiopolymerFormat)
{
editorType = MoleculeRendererType.Helm;
}
else
{
editorType = MoleculeRendererType.Chemistry; // final default
}
}
bool useChemEditor = (editorType != MoleculeRendererType.Helm);
if (useChemEditor)
{
mol2 = mol.ConvertTo(MoleculeFormat.Molfile);
return(EditCdkMolMolecule(mol2, title));
}
else // use Helm editor
{
return(EditHelmMolecule(mol, title));
}
}
catch (Exception ex)
{
XtraMessageBox.Show(ex.Message, "Error editing molecule");
return(null);
}
}
/// <summary>
/// EditHelmMolecule
/// </summary>
/// <param name="title"></param>
/// <param name="mol"></param>
/// <returns></returns>
public static MoleculeMx EditHelmMolecule(
MoleculeMx mol,
string title = "")
{
MoleculeMx mol2 = mol.ConvertTo(MoleculeFormat.Helm);
string helm = mol2.HelmString;
HelmEditorDialog editor = new HelmEditorDialog();
string newHelm = editor.Edit(helm);
if (newHelm == null)
{
return(null);
}
mol2.SetPrimaryTypeAndValue(MoleculeFormat.Helm, newHelm);
return(mol2);
}
private void DelayedEditMolecule(object arg)
{
try
{
int ci = (int)arg;
string txt = Criteria.Text.Substring(ci + 16); // pick up structure name
int i1 = txt.IndexOf("]");
if (i1 < 0)
{
return;
}
string sid = txt.Substring(0, i1).Trim().ToUpper();
string chime = "";
if (LabeledCriteria.Structures.ContainsKey(sid))
{
chime = LabeledCriteria.Structures[sid];
}
MoleculeMx mol = new MoleculeMx(MoleculeFormat.Chime, chime);
MoleculeMx mol2 = MoleculeEditor.Edit(mol, MoleculeRendererType.Unknown, "Edit Structure");
if (mol2 == null)
{
return;
}
if (mol2.IsChemStructureFormat) // store chime for chem structure
{
mol2.ConvertTo(MoleculeFormat.Chime);
}
LabeledCriteria.Structures[sid] = mol2.GetPrimaryTypeAndValueString();
return;
}
catch (Exception ex)
{
return;
}
}
/// <summary>
/// Search button clicked, process input
/// </summary>
/// <returns></returns>
DialogResult ProcessInput()
{
CidList cidList = null;
StreamReader structureFileReader = null;
string qid; // query identifier, compoundId, file name or sdFile key value
QueryManager qm;
DataTableMx dt;
DataColumn dc;
DataRowMx dr;
//object[] dr; // if using Qe
DialogResult dlgRslt = DialogResult.OK;
Query q = null;
QueryTable qt = null;
QueryColumn simScoreQc, structQc; // query column containing latest query settings
MetaTable mt = null;
MetaColumn keyMc = null, structMc = null, dbSetMc = null, simScoreMc = null, mc;
MetaColumnType storageType;
string txt, tok;
if (DatabasesToSearch.Text == "")
{
MessageBoxMx.ShowError("Databases to search must be defined.");
DatabasesToSearch.Focus();
return(DialogResult.Cancel);
}
// Get list of databases
string[] dba = DatabasesToSearch.Text.Split(',');
List <MetaTable> tables = new List <MetaTable>();
foreach (string dbLabel0 in dba)
{
string dbLabel = dbLabel0.Trim();
RootTable dbInfo = RootTable.GetFromTableLabel(dbLabel);
if (dbInfo == null)
{
MessageBoxMx.ShowError("Can't find database " + DatabasesToSearch.Text);
DatabasesToSearch.Focus();
return(DialogResult.Cancel);
}
mt = MetaTableCollection.Get(dbInfo.MetaTableName);
if (mt == null)
{
MessageBoxMx.ShowError("Unable to locate parent structure table for database: " + DatabasesToSearch.Text);
DatabasesToSearch.Focus();
return(DialogResult.Cancel);
}
if (dbSetMc == null)
{
dbSetMc = mt.DatabaseListMetaColumn;
}
tables.Add(mt);
}
if (dbSetMc == null)
{
throw new Exception("\"Databases\" metacolumn not found for any of the databases to search");
}
// Validate other form values
RetrieveStructures = RetrieveMatchingStructures.Checked;
bool fromList = FromList.Checked;
int listCidsRead = 0;
int inputQueryCount = -1;
if (fromList) // using list, validate list name
{
if (SavedListUo == null)
{
MessageBoxMx.ShowError("Compound list must be defined.");
ListName.Focus();
return(DialogResult.Cancel);
}
cidList = CidListCommand.Read(SavedListUo);
if (cidList == null)
{
MessageBoxMx.ShowError("Error reading list.");
ListName.Focus();
return(DialogResult.Cancel);
}
inputQueryCount = cidList.Count;
}
else // Using SdFile, validate SdFile name
{
StructureFile = FileName.Text;
if (StructureFile == "")
{
MessageBoxMx.ShowError("File must be defined.");
FileName.Focus();
return(DialogResult.Cancel);
}
try { structureFileReader = new StreamReader(StructureFile); structureFileReader.Close(); }
catch (Exception ex)
{
MessageBoxMx.ShowError("Can't read file: " + Lex.Dq(StructureFile));
FileName.Focus();
return(DialogResult.Cancel);
}
keyField = KeyField.Text; // get key, blank to use name in 1st line
inputQueryCount = -1; // don't know how many queries unless we read the file (todo?)
}
tok = ResultsName.Text.Trim(); // name to store results under
if (tok == "")
{
MessageBoxMx.ShowError("A name for the results must be provided.");
ResultsName.Focus();
return(DialogResult.Cancel);
}
if (SubStruct.Checked)
{
Psc.SearchType = StructureSearchType.Substructure;
}
else if (Full.Checked)
{
Psc.SearchType = StructureSearchType.FullStructure;
}
else if (Similarity.Checked)
{
Psc.SearchType = StructureSearchType.MolSim;
}
else
{
throw new Exception("Unrecognized search type");
}
// Write initial log entries
SearchCount++;
string logFileName = ClientDirs.DefaultMobiusUserDocumentsFolder + @"\Multistructure Search " + SearchCount + ".txt";
if (!UIMisc.CanWriteFileToDefaultDir(logFileName))
{
return(DialogResult.Cancel);
}
LogStream = new StreamWriter(logFileName);
if (ResultsUo == null)
{
ResultsUo = new UserObject(UserObjectType.Annotation);
}
ResultsUo.Name = tok;
UserObjectTree.GetValidUserObjectTypeFolder(ResultsUo);
DateTime startTime = DateTime.Now;
WriteToLog("Multiple " + Psc.SearchType + " Search");
WriteToLog("Databases: " + DatabasesToSearch.Text);
WriteToLog("Date: " + startTime);
if (fromList)
{
WriteToLog("Input List: " + SavedListUo.Name);
}
else
{
WriteToLog("Input Structure File: " + StructureFile);
}
WriteToLog("Output List: " + ResultsUo.Name);
WriteToLog("Log File: " + logFileName);
WriteToLog("");
WriteToLog("Query, Match, Score");
int queryCount = 0;
int matchAtLeastOneCount = 0;
MoleculeMx queryStructure = null; // current structure being searched
CidList matchList = new CidList();
List <MatchData> matchData = new List <MatchData>();
if (FromFile.Checked) // open SdFile as required
{
structureFileReader = new StreamReader(StructureFile);
}
// Search of structures one at a time
while (true)
{
if (fromList) // get next structure from list
{
if (listCidsRead >= cidList.Count)
{
break;
}
qid = cidList[listCidsRead].Cid;
listCidsRead++;
if (qid.Trim() == "")
{
continue;
}
if (qid.ToLower().IndexOf(".mol") > 0 || qid.ToLower().IndexOf(".skc") > 0)
{ // file reference
if (!File.Exists(qid))
{
continue;
}
if (qid.ToLower().IndexOf(".mol") > 0)
{
queryStructure = MoleculeMx.ReadMolfile(qid);
}
else
{
queryStructure = MoleculeMx.ReadSketchFile(qid);
}
}
else
{
queryStructure = MoleculeUtil.SelectMoleculeForCid(qid);
}
if (queryStructure == null || queryStructure.AtomCount == 0)
{
continue; // oops
}
}
else // get next structure from input file
{
qid = null;
if (StructureFile.ToLower().EndsWith(".sdf"))
{
List <SdFileField> fList = SdFileDao.Read(structureFileReader);
if (fList == null) // end of sdFile
{
structureFileReader.Close();
break;
}
if (fList.Count == 0)
{
continue;
}
queryStructure = new MoleculeMx(MoleculeFormat.Molfile, fList[0].Data);
if (queryStructure == null || queryStructure.AtomCount == 0)
{
continue;
}
if (keyField != "") // key field specified?
{
qid = SdFileDao.GetDataField(fList, keyField);
}
else // get name from 1st line of molfile
{
string molFile = fList[0].Data;
int i1 = molFile.IndexOf("\n");
if (i1 == 0)
{
qid = "";
}
else
{
qid = molFile.Substring(0, i1).Trim();
}
}
if (string.IsNullOrEmpty(qid))
{
qid = SdFileDao.GetDataField(fList, "compound_id");
}
}
else // assume smiles file
{
string smiles = structureFileReader.ReadLine();
if (smiles == null) // end of sdFile
{
structureFileReader.Close();
break;
}
smiles = smiles.Trim();
if (smiles.Length == 0)
{
continue;
}
int i1 = smiles.IndexOf(","); // get any preceeding queryId
if (i1 < 0)
{
i1 = smiles.IndexOf("\t");
}
if (i1 >= 0)
{
qid = smiles.Substring(0, i1).Trim();
smiles = smiles.Substring(i1 + 1).Trim();
}
queryStructure = new MoleculeMx(MoleculeFormat.Smiles, smiles);
if (queryStructure == null || queryStructure.AtomCount == 0)
{
continue;
}
}
if (qid == null || qid.Trim() == "")
{
qid = (queryCount + 1).ToString(); // be sure we have a query id
}
}
queryCount++; // count the query
if (queryStructure == null || queryStructure.AtomCount == 0)
{
WriteToLog("Error converting specific structure " + queryCount.ToString() + ", " + qid);
continue;
}
queryStructure.RemoveStructureCaption(); // remove any Mobius-added caption
Psc.Molecule = queryStructure;
string msg =
"Searching Structure: " + queryCount.ToString();
if (inputQueryCount > 0)
{
msg += " of " + inputQueryCount.ToString();
}
msg += "\n" +
"Structures with one or more matches: " + matchAtLeastOneCount.ToString() + "\n" +
"Total Matches: " + matchList.Count.ToString();
Progress.Show(msg);
// Do the search over the list of databases
for (int ti = 0; ti < tables.Count; ti++)
{
mt = tables[ti];
q = new Query(); // build basic query
//q.SingleStepExecution = true; // do in single step (doesn't currently return sim score)
q.ShowStereoComments = false;
qt = new QueryTable(mt);
q.AddQueryTable(qt);
qt.SelectKeyOnly(); // start selecting desired cols
keyMc = mt.KeyMetaColumn;
structMc = mt.FirstStructureMetaColumn;
structQc = qt.GetQueryColumnByName(structMc.Name);
structQc.Selected = RetrieveStructures;
dbSetMc = mt.DatabaseListMetaColumn;
if (dbSetMc == null)
{
throw new Exception("\"Databases\" metacolumn not found for table: " + mt.Label);
}
QueryColumn dbSetQc = qt.GetQueryColumnByName(dbSetMc.Name);
dbSetQc.Selected = true; // select the database name
RootTable root = RootTable.GetFromTableName(mt.Name);
txt = " in (" + root.Label + ")";
dbSetQc.Criteria = dbSetMc.Name + txt;
dbSetQc.CriteriaDisplay = txt;
simScoreMc = mt.SimilarityScoreMetaColumn;
simScoreQc = null;
if (simScoreMc != null) // get sim score if it exists
{
simScoreQc = qt.GetQueryColumnByName(simScoreMc.Name);
simScoreQc.Selected = true; // return sim score
}
Psc.QueryColumn = structQc;
ParsedStructureCriteria psc2 = AdjustSearchForSmallWorldAsNeeded(Psc);
psc2.ConvertToQueryColumnCriteria(structQc); // format the QC for the structure search
DateTime t0 = DateTime.Now;
//QueryEngine qe = new QueryEngine();
//qe.NextRowsMin = 1000; // minimum number of rows to prefetch
//qe.NextRowsMax = -1; // maximum number of rows to prefetch
//qe.NextRowsMaxTime = 10000; // max time in milliseconds for next fetch
//qe.ExecuteQuery(q);
qm = new QueryManager();
try { dlgRslt = qm.ExecuteQuery(ref q); }
catch (Exception ex)
{
WriteToLog("Error searching structure: " + ex.Message + ", " + queryCount.ToString() + ", " + qid);
continue;
}
if (dlgRslt != DialogResult.OK)
{
return(dlgRslt);
}
double executeTime = TimeOfDay.Delta(ref t0);
int offset = qm.DataTableManager.KeyValueVoPos + 1;
//int offset = 0; // for QE
int keyPos = offset++;
int strPos = RetrieveStructures ? offset++ : -1;
int dbPos = offset++;
int simPos = offset++;
int fetchCnt = 0;
while (true)
{
dr = qm.NextRow();
//dr = qe.NextRow(); // for Qe
if (dr == null)
{
break;
}
fetchCnt++;
if (fetchCnt == 1)
{
matchAtLeastOneCount++; // number of queries that have at least one match
}
MatchData md = new MatchData();
md.Qno = queryCount;
md.Qid = qid;
if (RetrieveStructures)
{
md.Qstr = "Chime=" + queryStructure.GetChimeString();
}
CompoundId cid = CompoundId.ConvertTo(dr[keyPos]);
md.Mid = cid.Value;
if (RetrieveStructures)
{
MoleculeMx ms = MoleculeMx.ConvertTo(dr[strPos]);
if (!NullValue.IsNull(ms))
{
md.Mstr = "Chime=" + ms.GetChimeString();
}
}
StringMx db = StringMx.ConvertTo(dr[dbPos]);
if (!NullValue.IsNull(db))
{
md.Db = db.Value;
}
if (Psc.SearchType == StructureSearchType.MolSim)
{
NumberMx nex = NumberMx.ConvertTo(dr[simPos]);
if (!NullValue.IsNull(nex))
{
md.Score = nex.Value;
}
}
if (matchList.Contains(cid.Value)) // already have compound id as match for other query?
{
if (Psc.SearchType != StructureSearchType.MolSim)
{
continue; // if similarity search see if more similar
}
CidListElement le = matchList.Get(cid.Value); // reference current score
if (le.Tag > md.Score)
{
continue; // only replace if more similar
}
matchList.Remove(le.Cid); // remove from list
for (int mi = 0; mi < matchData.Count; mi++) // remove from data
{
if (matchData[mi].Mid == md.Mid)
{
matchData.RemoveAt(mi);
break;
}
}
}
matchList.Add(md.Mid);
matchList.Get(md.Mid).Tag = md.Score; // keep score in list
matchData.Add(md); // add to results
txt = md.Qid + ", " + md.Mid + ", " + md.Score.ToString();
WriteToLog(txt);
} // Fetch result loop
double fetchTime = TimeOfDay.Delta(ref t0);
} // DB loop
if (Progress.CancelRequested)
{
Progress.Hide();
MessageBoxMx.ShowError("Search cancelled.");
try { LogStream.Close(); } catch { }
return(DialogResult.Cancel);
}
} // key loop
CidListCommand.WriteCurrentList(matchList); // write the list of numbers
UsageDao.LogEvent("MultipleStructSearch");
txt =
"=== Multiple structure search complete ===\r\n\r\n" +
"Structures Searched: " + queryCount.ToString() + "\r\n";
txt +=
"Structures with one or more matches: " + matchAtLeastOneCount.ToString() + "\r\n" +
"Total Matches: " + matchList.Count.ToString() + "\r\n";
TimeSpan ts = DateTime.Now.Subtract(startTime);
ts = new TimeSpan(ts.Hours, ts.Minutes, ts.Seconds);
txt += "Total Time: " + ts + "\r\n\r\n";
WriteToLog("\r\n" + txt);
try { LogStream.Close(); } catch { }
if (matchList.Count == 0)
{
Progress.Hide();
MessageBoxMx.ShowError("No matches have been found.");
return(DialogResult.Cancel);
}
tok = "Matching compound ids";
if (Psc.SearchType == StructureSearchType.MolSim)
{
tok = "Similar compound ids";
}
txt += tok + " have been written to the current list: " + ResultsUo.Name + "\n" +
"Log file written to: " + logFileName + "\n\n" +
"Do you want to view the match results?";
DialogResult dRslt = MessageBoxMx.Show(txt, "Multiple Structure Search", MessageBoxButtons.YesNoCancel, MessageBoxIcon.Question);
if (dRslt == DialogResult.Cancel)
{
return(DialogResult.Cancel);
}
else if (dRslt == DialogResult.No) // show log
{
SystemUtil.StartProcess(logFileName);
return(DialogResult.Cancel);
}
// Display results
Progress.Show("Formatting results...");
mt = new MetaTable();
mt.Name = "MULTSTRUCTSEARCH_" + SearchCount;
mt.Label = "Multiple Structure Search " + SearchCount;
mt.TableMap = "Mobius.Tools.MultStructSearch"; // plugin id
mt.MetaBrokerType = MetaBrokerType.NoSql;
ColumnSelectionEnum structureColumnSelection = RetrieveStructures ? ColumnSelectionEnum.Selected : ColumnSelectionEnum.Unselected;
keyMc = keyMc.Clone();
keyMc.Name = "MatchingCid";
keyMc.Label = "Matching Compound Id";
mt.AddMetaColumn(keyMc);
structMc = structMc.Clone();
structMc.Name = "MatchingStructure";
structMc.Label = "Matching Structure";
structMc.InitialSelection = structureColumnSelection;
mt.AddMetaColumn(structMc);
dbSetMc = dbSetMc.Clone();
dbSetMc.Name = "Database";
mt.AddMetaColumn(dbSetMc);
//if (DatabasesToSearch.Text.Contains(","))
dbSetMc.InitialSelection = ColumnSelectionEnum.Selected;
mc = mt.AddMetaColumn("Molsimilarity", "Similarity Search Score", MetaColumnType.Number, ColumnSelectionEnum.Unselected, 10);
if (Psc.SearchType == StructureSearchType.MolSim)
{
mc.InitialSelection = ColumnSelectionEnum.Selected;
}
mc = mt.AddMetaColumn("QueryNo", "Query Number", MetaColumnType.Integer);
//mc = mt.AddMetaColumn("QueryMatchNo", "Query Match Number", MetaColumnType.Integer);
mc = mt.AddMetaColumn("QueryId", "Query Id", MetaColumnType.String);
mc = mt.AddMetaColumn("QueryStructure", "Query Structure", MetaColumnType.Structure);
mc.InitialSelection = structureColumnSelection;
q = ToolHelper.InitEmbeddedDataToolQuery(mt);
dt = q.ResultsDataTable as DataTableMx;
for (int mi = 0; mi < matchData.Count; mi++)
{
MatchData md = matchData[mi];
dr = dt.NewRow();
dr[qt.Alias + ".MatchingCid"] = new CompoundId(md.Mid);
if (RetrieveStructures)
{
dr[qt.Alias + ".MatchingStructure"] = new MoleculeMx(MoleculeFormat.Chime, md.Mstr);
}
dr[qt.Alias + ".Database"] = new StringMx(md.Db);
if (Psc.SearchType == StructureSearchType.MolSim)
{
dr[qt.Alias + ".Molsimilarity"] = new NumberMx(md.Score);
}
dr[qt.Alias + ".QueryNo"] = new NumberMx(md.Qno);
dr[qt.Alias + ".QueryId"] = new StringMx(md.Qid);
if (RetrieveStructures)
{
dr[qt.Alias + ".QueryStructure"] = new MoleculeMx(MoleculeFormat.Chime, md.Qstr);
}
dt.Rows.Add(dr);
}
ToolHelper.DisplayData(q, dt, true);
Progress.Hide();
return(DialogResult.OK);
}
/// <summary>
/// Sync the Mobius CorpMoltable replicate used to retrieve Smiles
/// Syntax: UpdateCorpDbMoltableMx [ ByDateRange | ByCorpIdRange | LoadMissing | <singleCorpId>]
/// </summary>
/// <returns></returns>
public static string UpdateCorpDbMoltableMx(
string args)
{
DateTime moleculeDateTime = DateTime.MinValue;
double mw;
string msg = "", sql = "", maxCorpIdSql, mf = "", chime = "", smiles = "", checkPointDate = "", helm = "", sequence = "", svg = "";
object[][] pva = null;
int pvaCount = 0, CorpId, lowCorpId = 0, highCorpId = 0, srcMaxCorpId = 0;
int SelectChunkSize = 20; // small chunks
int InsertBufferSize = 10;
//int SelectChunkSize = 100000; // big chunks
//int InsertBufferSize = 1000;
// Select data from corp_moltable by CorpId range
const string SelectByCorpIdRange = @"
SELECT
m.corp_nbr,
chime(m.ctab),
m.molformula,
m.molweight,
null molsmiles,
s.helm_txt,
s.sequence_txt,
m.molecule_date
FROM
corp_owner.corp_moltable m,
corp_owner.corp_substance s
where
m.corp_nbr > 0
and s.corp_nbr = m.corp_nbr
and (s.status_code is null or s.status_code = 'A')
ORDER BY corp_nbr";
// Select data from corp_moltable by date range comparing to corp_moltable_mx
const string SelectByDateRange = @"
select
m.corp_nbr,
chime(m.ctab),
m.molformula,
m.molweight,
null molsmiles,
s.helm_txt,
s.sequence_txt,
m.molecule_date,
m2.molecule_date
from
corp_owner.corp_moltable m,
corp_owner.corp_substance s,
corp_moltable_mx m2
where
m.molecule_date > to_date('1-jan-1900 000000','DD-MON-YYYY HH24MISS')
and s.corp_nbr = M.CORP_NBR
and (s.status_code is null or s.status_code = 'A')
and m2.corp_nbr (+) = m.corp_nbr
and m2.molecule_date (+) != m.molecule_date
order by m.molecule_date" ;
// Select for missing smiles strings, ex: Update CorpDbMoltableMx LoadMissing mx.molecule_date > '1-jan-2014'
const string SelectMissingSmilesFix = @"
select /* check for CorpIds in corp_moltable not in corp_moltable_mx */
corp_nbr,
chime(ctab),
molformula,
molweight,
null molsmiles,
helm_txt,
sequence_txt,
molecule_date
from
(
select
m.*,
s.helm_txt,
s.sequence_txt,
mx.molsmiles molsmiles_mx
from
corp_owner.corp_moltable m,
corp_owner.corp_substance s,
corp_moltable_mx mx
where
s.corp_nbr = M.CORP_NBR
and (s.status_code is null or s.status_code = 'A')
and mx.corp_nbr (+) = m.corp_nbr
and 1=1 /* condition to substitute */
) m
where molsmiles_mx is null /* extra condition */
order by corp_nbr" ;
// Insert missing helm info
const string SelectMissingHelmFix = @"
select /* check for CorpIds in corp_moltable not in corp_moltable_mx */
corp_nbr,
chime(ctab),
molformula,
molweight,
null molsmiles,
helm_txt,
sequence_txt,
molecule_date
from
(
select
m.*,
s.helm_txt,
s.sequence_txt,
mx.molsmiles molsmiles_mx
from
corp_owner.corp_moltable m,
corp_owner.corp_substance s,
corp_moltable_mx mx
where
s.corp_nbr = M.CORP_NBR
and (s.status_code is null or s.status_code = 'A')
and mx.corp_nbr (+) = m.corp_nbr
and 1=1 /* condition to substitute */
) m
where length(helm_txt) > 0 /* extra condition */
order by corp_nbr" ;
// Secondary "large" structure table (~5k mols)
const string SelectLargeMols = @"
select
corp_nbr,
to_clob(molstructure),
to_clob(molformula),
molweight,
molsmiles,
null helm_txt,
null sequence_txt,
molecule_date
from
(select
corp_srl_nbr corp_nbr,
'CompoundId=' || corp_srl_nbr molstructure,
null ctab,
mlclr_frml_txt molformula,
mlclr_wgt molweight,
null molsmiles,
null molecule_date
from rdm_owner.rdm_sbstnc
where rdw_src_cd = 'LRG'" ;
// Insert statement
const string InsertSql = @"
insert into mbs_owner.corp_moltable_mx (
corp_nbr,
molstructure,
molformula,
molweight,
molsmiles,
molecule_date)
values (:0, :1, :2, :3, :4, :5)" ;
// Build select sql
bool byDateRange = false, byCorpIdRange = false, missingFix = true, deleteExisting = true;
string missingFixCriteria = "";
if (Lex.IsUndefined(args) || Lex.Eq(args, "ByDateRange"))
{
byDateRange = true;
}
else if (Lex.Eq(args, "ByCorpIdRange"))
{
byCorpIdRange = true;
Progress.Show("Getting range of CorpIds to insert...");
maxCorpIdSql = "select max(corp_nbr) from corp_owner.corp_moltable"; // get highest CorpId in source db
srcMaxCorpId = SelectSingleValueDao.SelectInt(maxCorpIdSql);
if (srcMaxCorpId < 0)
{
srcMaxCorpId = 0;
}
maxCorpIdSql = "select max(corp_nbr) from mbs_owner.corp_moltable_mx"; // get highest CorpId in dest db
highCorpId = SelectSingleValueDao.SelectInt(maxCorpIdSql);
if (highCorpId < 0)
{
highCorpId = 0;
}
}
else if (Lex.StartsWith(args, "LoadMissing"))
{
missingFix = true;
if (args.Contains(" "))
{
missingFixCriteria = args.Substring(10).Trim();
}
}
else if (int.TryParse(args, out srcMaxCorpId)) // single CorpId
{
byCorpIdRange = true;
highCorpId = srcMaxCorpId - 1; // say 1 less is the max we have
}
else
{
return("Syntax: UpdateCorpDbMoltableMx [ ByDateRange | ByCorpIdRange | LoadMissing | <singleCorpId>]");
}
Log("UpdateCorpDbMoltableMx started: " + args);
int readCount = 0, insCount = 0, insertCount = 0, updateCount = 0, undefinedStructures = 0, smilesSuccess = 0, smilesFails = 0, helmStructures = 0;
List <string> CorpIdList = new List <string>();
for (int chunk = 1; ; chunk++) // loop over chunks
{
if (byDateRange) // single chunk
{
if (chunk > 1)
{
break; // break 2nd time through
}
checkPointDate = UserObjectDao.GetUserParameter("MOBIUS", "UpdateCorpDbMoltableMxCheckpointDate", "01-sep-2013 000000");
//UserObjectDao.SetUserParameter("MOBIUS", "UpdateCorpDbMoltableMxCheckpointDate", checkPointDate);
sql = Lex.Replace(SelectByDateRange, "1-jan-1900 000000", checkPointDate);
msg = "Reading where date >= " + checkPointDate;
}
else if (byCorpIdRange) // by CorpId range
{
if (highCorpId >= srcMaxCorpId)
{
break; // done
}
lowCorpId = highCorpId + 1; // start of next chunk
highCorpId = lowCorpId + SelectChunkSize;
if (highCorpId >= srcMaxCorpId)
{
highCorpId = srcMaxCorpId;
}
sql = Lex.Replace(SelectByCorpIdRange, "corp_nbr > 0", "corp_nbr between " + lowCorpId + " and " + highCorpId);
msg = "Reading: " + lowCorpId + " to " + highCorpId + ", Reads: " + readCount + ", Inserts: " + insertCount;
}
else if (missingFix)
{
if (chunk > 1)
{
break; // break 2nd time through
}
sql = SelectMissingHelmFix;
if (Lex.IsDefined(missingFixCriteria)) // substitute any criteria
{
sql = Lex.Replace(sql, "1=1", missingFixCriteria);
}
msg = "Fixing missing data";
}
Progress.Show(msg);
DbCommandMx readCmd = new DbCommandMx();
readCmd.MxConn = DbConnectionMx.GetConnection("prd123");
readCmd.PrepareUsingDefinedConnection(sql, null);
DbDataReader rdr = readCmd.ExecuteReader();
DbCommandMx insertCmd = new DbCommandMx();
OracleDbType[] pta = new OracleDbType[6];
pta[0] = OracleDbType.Int32; // corp_nbr
pta[1] = OracleDbType.Clob; // molstructure
pta[2] = OracleDbType.Clob; // molformula
pta[3] = OracleDbType.Double; // molweight
pta[4] = OracleDbType.Clob; // smiles
pta[5] = OracleDbType.Date; // molecule_date
insertCmd.Prepare(InsertSql, pta);
insertCmd.BeginTransaction(); // be sure we have a transaction going
pva = DbCommandMx.NewObjectArrayArray(6, InsertBufferSize); // alloc insert row array
object[] vo = new object[6];
while (true)
{
bool readOk = rdr.Read();
if (readOk)
{
rdr.GetValues(vo);
CorpId = readCmd.GetInt(0); // corp_nbr
vo[0] = CorpId;
CorpIdList.Add(CorpId.ToString());
if (!readCmd.IsNull(1)) // molstructure
{
chime = readCmd.GetClob(1);
chime = OracleMx.ClearStringIfExceedsMaxStringSize(chime);
vo[1] = chime;
}
else
{
chime = "";
}
if (!readCmd.IsNull(2)) // molformula
{
mf = readCmd.GetClob(2);
mf = OracleMx.ClearStringIfExceedsMaxStringSize(mf);
vo[2] = mf;
}
if (!readCmd.IsNull(3)) // molweight
{
mw = readCmd.GetDouble(3);
vo[3] = mw;
}
if (Lex.IsDefined(chime)) // molsmiles - calculate from chime string
{
MoleculeMx cs = new MoleculeMx(MoleculeFormat.Chime, chime);
if (cs.AtomCount > 1) // need more than one atom
{
MoleculeMx cs2 = cs.ConvertTo(MoleculeFormat.Smiles);
smiles = cs2.GetSmilesString();
if (Lex.IsDefined(smiles))
{
smilesSuccess++;
}
else
{
Log("Smiles conversion failure for CorpId: " + CorpId);
smilesFails++;
}
smiles = OracleMx.ClearStringIfExceedsMaxStringSize(smiles);
vo[4] = smiles;
}
else
{
undefinedStructures++;
}
}
else
{
undefinedStructures++;
}
if (!readCmd.IsNull(5))
{
helm = readCmd.GetClob(5);
if (Lex.IsDefined(helm))
{
svg = HelmControl.GetSvg(helm);
vo[1] = SvgUtil.CompressSvgString(svg); // store compressed svg in molstructure column for now
helmStructures++;
}
}
if (!readCmd.IsNull(6))
{
sequence = readCmd.GetClob(6);
if (Lex.IsDefined(sequence))
{
// nothing yet
}
}
moleculeDateTime = DateTime.MinValue;
if (!readCmd.IsNull(7)) // molecule_date
{
moleculeDateTime = readCmd.GetDateTime(7);
vo[5] = moleculeDateTime;
}
for (int pi = 0; pi < 6; pi++) // invert for insert
{
pva[pi][pvaCount] = vo[pi];
}
if (Debug)
{
msg = String.Format("CorpId: {0}, mf: {1}, chime: {2}, smiles: {3}", CorpId.ToString(), mf.Length, chime.Length, smiles.Length);
Log(msg);
}
pvaCount++;
}
if (pvaCount >= InsertBufferSize || (!readOk && pvaCount > 0)) // write if buffer full or at end
{
try
{
if (deleteExisting)
{
int delCount = DoDeletes(CorpIdList);
updateCount += delCount; // count deletes as updates
insertCount -= delCount; // subtract from inserts
}
CorpIdList.Clear();
insCount = insertCmd.ExecuteArrayNonReader(pva, ref pvaCount);
insertCmd.Commit();
insertCmd.BeginTransaction();
insertCount += insCount;
}
catch (Exception ex)
{
throw new Exception(ex.Message, ex);
}
if (byDateRange)
{
string checkPointDate2 = String.Format("{0:dd-MMM-yyyy HHmmss}", moleculeDateTime); // format date time that will work with oracle
UserObjectDao.SetUserParameter("MOBIUS", "UpdateCorpDbMoltableMxCheckpointDate", checkPointDate2);
msg = "Processing where date >= " + checkPointDate + ", Reads: " + readCount + ", Inserts: " + insertCount + ", Updates: " + updateCount;
}
else if (byCorpIdRange) // CorpId range
{
msg = "Processing: " + lowCorpId + " to " + highCorpId + ", Reads: " + readCount + ", Inserts: " + insertCount;
}
else if (missingFix)
{
msg = "Fixing missing smiles, Updates: " + updateCount;
}
msg += String.Format(", Undefined structures: {0} , Smiles failures: {1}, Helms: {2}", undefinedStructures, smilesFails, helmStructures);
Progress.Show(msg);
}
if (!readOk)
{
break;
}
readCount++;
}
readCmd.Dispose();
insertCmd.Dispose();
} // end for select chunk
msg = "UpdateCorpDbMoltableMx - Inserts: " + insertCount + ", Updates: " + updateCount;
msg += String.Format(", Undefined structures: {0} , Smiles failures: {1}, Helms: {2}", undefinedStructures, smilesFails, helmStructures);
Log(msg);
return(msg);
}