This README should allow the replication top to end of the approach to extract OO metrics and developers
It is a combination of bash commands (starting with '$') and mysql commands (starting with 'mysql>')
================================================================================================================

There are few steps to perform, each with some pre-requisite tool:
1) cloning repositories
2) storing metadata via CVSAnaly2
3) evaluating SLOCs
4) extracting OO metrics (via SciTools Understand) 
5) evaluating contributors
    5.1) removing duplicates
    5.2) extracting distinct Java files worked on by each contributor
    5.3) evaluating Top, Middle and Bottom developers
6) collating OO data and authors data

================================================================================================================

# GIT CLONE + CVSANALY2 (Requires cvsanaly2)

mysql> CREATE DATABASE DB_ALL;

$ for i in {1..30}; do wget -O $i.txt https://github.com/search?o=desc&p=$i&q=java&s=forks&type=Repositories && sleep 200; done
$ for i in *txt; do grep "\<a class\=\"v-align\-middle\" data\-hydro\-click\=" $i | awk -F';' '{print $34}' | sed s/\&quot// >> list; done
$ for i in `cat list`; do proj=`basename "$i"` && printf "*** NOW DOING $proj\n" && git clone $i && cvsanaly2 -u XXX -p YYY -d DB_ALL $proj && find ./$proj -type f ! -name '*.java' -delete; done
$ for i in `cat all`; do mysql --user=XXX --password=YYY newSamplingForks --skip-column-names -s -e "SELECT name,id FROM repositories WHERE name='$i';" >> repo_ID.csv; done

# SLOCs (Requires cloc)

$ for i in `cat list`; do basename $i >> all; done
$ IFS=$'\n'
$ for i in `cat all`; do find ./$i -name \"*.java\" > $i/all_source_files.txt; done
$ for i in `cat all`; do for j in `cat $i/all_source_files.txt`; do printf "$j," >> $i/SLOCS_$i.txt && cloc $j | grep Java | awk '{print $5}' >> $i/SLOCS_$i.txt; done; done
$ for i in `cat all`; do sed "s/\.\/$i\///g" <ALLPROJ/$i/SLOCS_$i.txt >> ALL_SLOCs.txt; done

mysql > CREATE TABLE `SLOCS` ( `filepath` text, `slocs` int(11) DEFAULT NULL ) ENGINE=InnoDB DEFAULT CHARSET=latin1;
mysql > LOAD DATA LOCAL INFILE "ALL_SLOCs.txt" INTO TABLE SLOCS FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';

# OO Metrics
$ for i in `cat all`; do ./prova-OO_METRICS.pl $i; done
$ for i in `cat all`; do mv $i/$i-CK-ClassMetrics.txt CK-Metrics; done
$ for i in `cat all`; do ./prova_CLEAN_OO.pl CK-Metrics/$i-CK-ClassMetrics.txt >> sql_oo.csv; done

mysql> LOAD DATA LOCAL INFILE "sql_oo.csv" INTO TABLE oo FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';

### AUTHORS
$ for i in `cat all`; do ./prova_AUTHOR_RECONCILIATION-FORKS.pl $i; done
$ for i in `cat all`; do ./prova_MULTIPLE_AUTHORS.pl $i; done
$ for i in `cat all`; do cat Authors/Authors_$i.txt >> sql_Authors-intermediate.csv; done
$ awk 'BEGIN { FS = OFS = ","; } FNR == NR { x[$1] = $2; next; } { if ($1 in x) { $1 = x[$1]; } } 1' repo_ID.csv sql_Authors-intermediate.csv >> sql_repoID_fileID_filename_authors.csv

mysql> LOAD DATA LOCAL INFILE "sql_repoID_fileID_filename_authors.csv" INTO TABLE projID_fileID_filepath_authors FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';

mysql> select projectID, filepath, max(aurthors) from projID_fileID_filepath_authors group by projectID, filepath into outfile '/tmp/sql_Authors.csv' FIELDS TERMINATED BY ',';

mysql> LOAD DATA LOCAL INFILE "/tmp/sql_Authors.csv" INTO TABLE authorsTable FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';

## JOIN
mysql> select projID, oo.classname, oo.fullpath , authors, IFANIN,CBO,NOC,NIM,NIV,WMC,RFC,DIT,LCOM from oo, authorsTable where oo.fullpath = authorsTable.fullpath into outfile '/var/lib/mysql-files/results_Forks.tab';

### TD, MD, BD
$ for i in `cat repoID`; do ./prova_TD_MD_BD.pl $i >> Developers_TOP_MIDDLE_BOTTOM.txt; done

### SLOCS
mysql> select results.*, SLOCS.slocs from results, SLOCS where results.fullpath = SLOCS.filepath into outfile '/tmp/results_plus_slocs.csv' FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';
mysql> LOAD DATA LOCAL INFILE "Developers_TOP_MIDDLE_BOTTOM.txt" INTO TABLE TDMDBD FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n';

$ for i in `cat repoID`; do ./prova_REPO_AUT_TYPE_FILE.pl $i >> projID_authorID_type_file_ID.csv; done

$ prova_SINGLETONS projID_authorID_type_file_ID.csv y

$ awk -F',' '{print $4}' projID_authorID_type_file_ID.csv > all_fileIDs.txt

$ for i in `cat all_fileIDs.txt`; do printf "$i," >> RESULTS_easy-short.csv && grep ",$i$" projID_authorID_type_file_ID.csv | grep -c TD  | tr -d "\n"  >> RESULTS_easy-short.csv && printf ","  >> RESULTS_easy-short.csv && grep ",$i$" projID_authorID_type_file_ID.csv | grep -c MD  | tr -d "\n"  >> RESULTS_easy-short.csv && printf ","  >> RESULTS_easy-short.csv && grep ",$i$" projID_authorID_type_file_ID.csv | grep -c BD  >> RESULTS_easy-short.csv; done
        
mysql> select projID, classname, fullpath, fileID_TD_MD_BD.fileID, authors, IFANIN,CBO,NOC,NIM,NIV,WMC,RFC,DIT,LCOM,TD,MD,BD from results, fileID_TD_MD_BD, projID_fileID_filepath_authors where results.fullpath = projID_fileID_filepath_authors.filepath AND projID_fileID_filepath_authors.fileID = fileID_TD_MD_BD.fileID;

## COLLATE OO metrics and SLOCS metrics
$ ./bash_COLLATE_OO_SLOCs.sh > results_Forks_SLOCS.csv