gCubeSystem
/
database-resource-manager
16
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

bug fixed for the smart smapling (with table col2oct2010-> scientific_names). Corrections in computeSmartSampleWithThreshold method. 

git-svn-id: https://svn.d4science.research-infrastructures.eu/gcube/trunk/data-access/DatabasesResourcesManager@99565 82a268e6-3cf1-43bd-a215-b396298e98cf
This commit is contained in:
Loredana Liccardo 2014-09-05 15:36:52 +00:00
parent 5af78b2871
commit b72e647ea8
1 changed files with 72 additions and 61 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

51
src/main/java/org/gcube/dataaccess/databases/sampler/Sampler.java
View File

@ -172,12 +172,15 @@ public class Sampler {
if (DBType.equals(POSTGRES)) {
// attribute = "CAST(" + listColumns.get(i) + " as text), ";
attribute = "CAST(" + listColumns.get(i) + " as character varying(255)), ";
attribute = "CAST(" + listColumns.get(i)
+ " as character varying(255)), ";
if (i == (listColumns.size() - 1)) {
// attribute = "CAST(" + listColumns.get(i) + " as text)";
attribute = "CAST(" + listColumns.get(i) + " as character varying(255))";
// attribute = "CAST(" + listColumns.get(i) +
// " as text)";
attribute = "CAST(" + listColumns.get(i)
+ " as character varying(255))";
}
@ -208,14 +211,17 @@ public class Sampler {
// attribute = "CONVERT(" + listColumns.get(i) +
// ", BINARY)";
attribute = "CAST(" + "`" +listColumns.get(i)+ "`"
+ " as CHAR(255) CHARACTER SET utf8)";
attribute = "CAST(" + "`" + listColumns.get(i)
+ "`" + " as CHAR(255) CHARACTER SET utf8)";
}
} else {
attribute = "CAST(CAST(" + "`" +listColumns.get(i)+ "`"
attribute = "CAST(CAST("
+ "`"
+ listColumns.get(i)
+ "`"
+ " as BINARY) as CHAR(255) CHARACTER SET utf8), ";
if (i == (listColumns.size() - 1)) {
@ -226,7 +232,10 @@ public class Sampler {
// attribute = "CONVERT(" + listColumns.get(i) +
// ", BINARY)";
attribute = "CAST(CAST(" + "`" +listColumns.get(i)+ "`"
attribute = "CAST(CAST("
+ "`"
+ listColumns.get(i)
+ "`"
+ " as BINARY) as CHAR(255) CHARACTER SET utf8)";
}
@ -560,7 +569,7 @@ public class Sampler {
// extract 200 rows randomly for each iteration
extractionRows: for (int i = 0; i < NIterations; i++) {
System.out.println("index iteration: " + i);
// System.out.println("index iteration: " + i);
AnalysisLogger.getLogger().debug(
"Sampler->executing the query: " + query);
@ -746,7 +755,8 @@ public class Sampler {
// //check rows added in the final result
// AnalysisLogger.getLogger().debug(
// "Sampler->adding row with index: " + i + " " + listRows.get(i).getRow());
// "Sampler->adding row with index: " + i + " " +
// listRows.get(i).getRow());
rows.add(listRows.get(i).getRow());
}
@ -915,7 +925,8 @@ public class Sampler {
if (DBType.equals(POSTGRES)) {
// the full name equal to "schemaname.tablename"
tablename = schemaName + "." + "\"" + tablename + "\"";
String tableName = "";
tableName = schemaName + "." + "\"" + tablename + "\"";
query = String.format(
queryForSmartSampleWithThresholdOnTablePostgres,
@ -924,9 +935,9 @@ public class Sampler {
// build the query for database mysql
if (DBType.equals(MYSQL)) {
// the full name equal to "dbname.tablename"
tablename = schemaName + "." + tablename;
String tableName = "";
tableName = schemaName + "." + tablename;
query = String.format(
queryForSmartSampleWithThresholdOnTableMysql,
@ -999,18 +1010,20 @@ public class Sampler {
// if (value == columnArray.length) {
AnalysisLogger.getLogger().debug(
"Sampler-> column array dimension: "
+ columnArray.length);
// print check
// AnalysisLogger.getLogger().debug(
// "Sampler-> column array dimension: "
// + columnArray.length);
double thresholdRank = ((columnArray.length) * 80);
thresholdRank = thresholdRank / 100;
double valCeil = Math.round(thresholdRank);
AnalysisLogger.getLogger().debug(
"Sampler-> threshold: " + thresholdRank
+ " rounded value: " + valCeil);
// print check
// AnalysisLogger.getLogger().debug(
// "Sampler-> threshold: " + thresholdRank
// + " rounded value: " + valCeil);
if (value >= (int) valCeil) {
@ -1326,8 +1339,6 @@ public class Sampler {
}
if (DBType.equals(MYSQL)) { // MySQL
// the full name equal to "dbname.tablename"