数据集 TPC-H、TPC-DS、IMDB的导入使用
数据集地址
一. TPC-H 数据集
- 数据集下载
TPC-H数据集
可采用gcc下载或者直接下载zip包，然后解压即可。
具体使用方法可以查看 dbgen文件目录下的：README文件 - 数据表的创建与约束
在下载的文件包下的dbgen文件夹下，有2个文件，分别是dss.ddl、dss.ri。一个用于创建表、一个用于规约。
（1）dss.ddl是创建表的语句，将里面的内容放入到dss.sql中（你自己创建的sql文件）,执行表中的语句即可创建表。
\i /pathA/dss.sql
# 该文件对应的路径
（2）dss.ri 是表之间的限制条件,也将内容复制到sql文件下,dss_ri,sql，然后执行（如果在导入数据的时候报错，先导入数据，再进行规约）
这里红色的框内需要删除或者注释掉，注意TPCD也需要删除
注意：数据约束这里有问题，需要修改下表内的内容，把外键这里的语句改成如下格式。
ADD FOREIGN KEY (L_PARTKEY,L_SUPPKEY) references PARTSUPP;
\i /pathA/dss_ri.sql
# 该文件对应的路径
- 数据的生成与导入
（1）生成dbgen、qgen
通过下面命令，生成两个文件，分别是dbgen、qgen。分别用于产生数据和查询。
make -f Makefile
（2）生成数据
生成8个tbl文件，生成以后修改八个文件的可执行权限，将这八个tbl文件放入创建好的文件目录tbl下(注意1表示1G数据量，0.1表示100M数据量)
./dbgen -s 1 -f
chmod 777 tbl/*.tbl
（3）数据导入
copy customer from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/customer.tbl' with delimiter as '|' NULL '';
copy lineitem from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/lineitem.tbl' with delimiter as '|' NULL '';
copy nation from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/nation.tbl' with delimiter as '|' NULL '';
copy orders from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/orders.tbl' with delimiter as '|' NULL '';
copy part from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/part.tbl' with delimiter as '|' NULL '';
copy partsupp from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/partsupp.tbl' with delimiter as '|' NULL '';
copy region from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/region.tbl' with delimiter as '|' NULL '';
copy supplier from '/home/ywb/Data/tpch-kit-master/dbgen/tbl/supplier.tbl' with delimiter as '|' NULL '';
或者命令行
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/region.tbl' into table region fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/nation.tbl' into table nation fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/customer.tbl' into table customer fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/supplier.tbl' into table supplier fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/part.tbl' into table part fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/orders.tbl' into table orders fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/partsupp.tbl' into table partsupp fields terminated by '|' lines terminated by '\n';
load data local infile '/home/Data/tpch-kit-master/dbgen/tbl/lineitem.tbl' into table lineitem fields terminated by '|' lines terminated by '\n';
- 查询语句生成
（1）将dists.dss、qgen复制到queries下
（2）批量生成sql语句（在queries目录下执行命令）
for i in {1..22}
do
name="d$i.sql"
echo $name
./qgen -d $i >$name
done
二. TPC-DS数据集
- 数据下载
TPC-DS数据集
采用gcc下载或者直接下载zip包，然后解压即可。
阿里云实现TPC-DS教程
具体使用方法可以查看 tools文件目录下的：How_To_Guide-DS-V2.0.0.docx - 数据表创建与约束
在下载的文件包下的tools文件夹下，有3个sql文件，分别是tpcds.sql、tocds_ri.sql、tpcds_source.sql。
（1）tpcds.sql是创建表的语句，执行表中的语句即可创建表。
\i /pathA/tpcds.sql
# 该文件对应的路径
（2）tpcds_ri.sql 是表之间的限制条件
注意这一步表的约束最好放到导入数据完以后再执行，否则主外键的的错误。
\i /pathA/tpcds_ri.sql
# 该文件对应的路径
- 数据生成与导入
（1）数据生成（在tools目录下）
dsdgen、dsqgen分别用于生成数据和查询模板
注意：这里可能make的时候没有生成dsqgen这个，原因是少了两个包，根据错误提示下载即可
sudo apt-get install byacc # 或者 apt-get install yacc
sudo apt-get install flex
make -f Makefile.suite # dsdgen、dsqgen 生成
./dsdgen -sc 1 -DIR 'data_output_path'
# ./dsdgen -sc 1 -DIR 'data/' # data是我创建的数据目录位置，在tools下
# ./dsdgen -h 查看用法
#data_output_path： 数据生成路径
# 如果没有dsdgen，则执行命令：
# make -f Makefile.suite
（2）修改文件权限
修改data/下所有.dat文件的可执行权限
chmod 777 data/*.dat
（3） 数据导入
注意：生成的数据中会多一个 ’ | ’ 字符，导致导入数据报错，所以需要先删除每行最后一个 |,直接在dat文件目录下终端执行该命令。
删除多余的 |
for i in `ls *.dat`
do
name=$i
echo $name
sed -i 's#|$##g' $name
done
导入数据
copy call_center from '/home/ywb/Data/tpcds-kit-master/tools/data/call_center.dat' with delimiter as '|' NULL '';
copy catalog_page from '/home/ywb/Data/tpcds-kit-master/tools/data/catalog_page.dat' with delimiter as '|' NULL '';
copy catalog_returns from '/home/ywb/Data/tpcds-kit-master/tools/data/catalog_returns.dat' with delimiter as '|' NULL '';
copy catalog_sales from '/home/ywb/Data/tpcds-kit-master/tools/data/catalog_sales.dat' with delimiter as '|' NULL '';
copy customer from '/home/ywb/Data/tpcds-kit-master/tools/data/customer.dat' with delimiter as '|' NULL '';
copy customer_address from '/home/ywb/Data/tpcds-kit-master/tools/data/customer_address.dat' with delimiter as '|' NULL '';
copy customer_demographics from '/home/ywb/Data/tpcds-kit-master/tools/data/customer_demographics.dat' with delimiter as '|' NULL '';
copy date_dim from '/home/ywb/Data/tpcds-kit-master/tools/data/date_dim.dat' with delimiter as '|' NULL '';
copy dbgen_version from '/home/ywb/Data/tpcds-kit-master/tools/data/dbgen_version.dat' with delimiter as '|' NULL '';
copy household_demographics from '/home/ywb/Data/tpcds-kit-master/tools/data/household_demographics.dat' with delimiter as '|' NULL '';
copy income_band from '/home/ywb/Data/tpcds-kit-master/tools/data/income_band.dat' with delimiter as '|' NULL '';
copy inventory from '/home/ywb/Data/tpcds-kit-master/tools/data/inventory.dat' with delimiter as '|' NULL '';
copy item from '/home/ywb/Data/tpcds-kit-master/tools/data/item.dat' with delimiter as '|' NULL '';
copy promotion from '/home/ywb/Data/tpcds-kit-master/tools/data/promotion.dat' with delimiter as '|' NULL '';
copy reason from '/home/ywb/Data/tpcds-kit-master/tools/data/reason.dat' with delimiter as '|' NULL '';
copy ship_mode from '/home/ywb/Data/tpcds-kit-master/tools/data/ship_mode.dat' with delimiter as '|' NULL '';
copy store from '/home/ywb/Data/tpcds-kit-master/tools/data/store.dat' with delimiter as '|' NULL '';
copy store_returns from '/home/ywb/Data/tpcds-kit-master/tools/data/store_returns.dat' with delimiter as '|' NULL '';
copy store_sales from '/home/ywb/Data/tpcds-kit-master/tools/data/store_sales.dat' with delimiter as '|' NULL '';
copy time_dim from '/home/ywb/Data/tpcds-kit-master/tools/data/time_dim.dat' with delimiter as '|' NULL '';
copy warehouse from '/home/ywb/Data/tpcds-kit-master/tools/data/warehouse.dat' with delimiter as '|' NULL '';
copy web_page from '/home/ywb/Data/tpcds-kit-master/tools/data/web_page.dat' with delimiter as '|' NULL '';
copy web_returns from '/home/ywb/Data/tpcds-kit-master/tools/data/web_returns.dat' with delimiter as '|' NULL '';
copy web_sales from '/home/ywb/Data/tpcds-kit-master/tools/data/web_sales.dat' with delimiter as '|' NULL '';
copy web_site from '/home/ywb/Data/tpcds-kit-master/tools/data/web_site.dat' with delimiter as '|' NULL '';
（4）生成查询
for i in `seq 1 99`
do
./dsqgen -DIRECTORY ../query_templates/ -TEMPLATE "query${i}.tpl" -DIALECT netezza -FILTER Y > ../sql/query${i}.sql
done
注意：因为query_templates只有99个查询模板，所以最多一次只能生成99个查询语句，但是如果生成多次，生成的99个查询语句是一样
-DIRECTORY：SQL模板的路径。 这个是查询模板的文件目录位置 …/query_templates/
-TEMPLATE：SQL模板的名称
-DIALECT：include query dialect defintions found in < s >.tpl 。-DIALECT postgresql 是哪种数据库的查询语句
注意：DIALECT支持oracle,db2,sqlserver，netezza，ansi，但是没有postgresql。可以去query_templates文件目录下看是否有对应的模板文件，推荐netezza，pg可以解析。
-FILTER：重定向到标准输出，即…/sql/query${i}.sql 是输出路径和文件类型
三. IMDB数据集
- 数据下载
JOB查询语句
IMDB数据集
第二个链接好像直接下数据集有点慢，建议在第一个链接中的REMADE中的链接进行跳转下载
- 数据集导入
（1）sql文件时创建表的文件、csv为对应的数据文件
（2）创建表
注意如果在图形界面运行命令则是i /path/xxx.sql;
\i /path/xxx.sql;
（3）导入数据
\copy info_type from '/home/ywb/Data/imdb/info_type.csv' with delimiter as ',' csv quote '"' escape as '\';
注意：这里有些csv字段中包含了分隔符逗号，所以用之前的命令会报错，注意csv quote '"' escape as '\'里面是双引号,这里真的需要小心，数据格式有点问题，搞了我好久。
\copy aka_name from '/home/ywb/Data/imdb/aka_name.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy aka_title from '/home/ywb/Data/imdb/aka_title.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy cast_info from '/home/ywb/Data/imdb/cast_info.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy char_name from '/home/ywb/Data/imdb/char_name.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy comp_cast_type from '/home/ywb/Data/imdb/comp_cast_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy company_name from '/home/ywb/Data/imdb/company_name.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy company_type from '/home/ywb/Data/imdb/company_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy complete_cast from '/home/ywb/Data/imdb/complete_cast.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy info_type from '/home/ywb/Data/imdb/info_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy keyword from '/home/ywb/Data/imdb/keyword.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy kind_type from '/home/ywb/Data/imdb/kind_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy link_type from '/home/ywb/Data/imdb/link_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy movie_companies from '/home/ywb/Data/imdb/movie_companies.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy movie_info from '/home/ywb/Data/imdb/movie_info.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy movie_info_idx from '/home/ywb/Data/imdb/movie_info_idx.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy movie_keyword from '/home/ywb/Data/imdb/movie_keyword.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy movie_link from '/home/ywb/Data/imdb/movie_link.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy name from '/home/ywb/Data/imdb/name.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy person_info from '/home/ywb/Data/imdb/person_info.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy role_type from '/home/ywb/Data/imdb/role_type.csv' with delimiter as ',' csv quote '"' escape as '\';
\copy title from '/home/ywb/Data/imdb/title.csv' with delimiter as ',' csv quote '"' escape as '\';
delimiter指定了字段之间的分隔符号位逗号
escape指定了在引号中的转义字符为反斜杠，这样即使在引号字串中存在引号本身，也可以用该字符进行转义，变为一般的引号字符，而不是字段终结
header true：指定文件中存在表头。如果没有的话，则设置为false
quote指定了以双引号作为字符串字段的引号，这样它会将双引号内的内容作为一个字段值来进行处理