• PostgreSQL | 学习笔记&语句汇总


    PostgreSQL 是世界上功能最强大的开源关系型数据库,除了标准的 SQL 的支持外,还有丰富的特性,可以通过扩展来增强数据库的能力,如uuid-ossp,pgcrypto,hstore,PostGIS等扩展。支持如 hstore、range、GIS、json 等更多的数据类型。此外,Greenplum也是基于PG来提供结构化大数据分析仓库解决方案。本文整理了postgre常用基本操作,包括:命令行操作、查询语句以及postgis空间操作等相关语句。

    1 命令行操作

    postgresql备份:
    pg_dump -U postgres -d gisportal -F c -f i:db2.dump
    postgresql还原:
    pg_restore -U postgres -d gisportal i:db2.dump
    pg_restore -U postgres -d gisportal e:postgres-20150420.backup
    备份单张表:
    pg_dump -U postgres -F c -b -v -f i:	est.bak -t sqs.test gisportal

    备份整个数据库(非模式):

    pg_dump -U postgres -F c -b -v -f i:	est.bak -t sqs.test gisportal 
    备份数据库下的模式:         
    pg_dump -U postgres -F c -b -v -f i:	est.bak -n sqs gisportal
    还原单张表:
    pg_restore -U postgres -W -d gisportal -v "d:poi.bak"

    要执行语句,先执行:cd postgresql安装目录bin中

    pg_dump -U postgres  -F c -f i:gisportal.dump gisportal(备份整个数据库,备份单个模式表为另一种写法)
    pg_dump -h localhost -U postgres(用户名) 数据库名(缺省时同用户名)  -t table(表名) >/data/dum.sql
    pg_dump -h localhost -U postgres gisportal  -t 省界_region >i:pros.sql

    2 基本操作

    创建表格

    create table schema_name."table_name"(
    id integer,
    name character varying,
    geom geometry)

    新建表格并复制某表内容

    create table table1 
    as 
    select * from table2

    插入另一张表的整列:

    insert into sqs.sqs_role(username) select username from sqs.sqs_user

    添加多个列

    alter table "table_name" 
    add "num_1" INTEGER, 
    add "num_2" INTEGER,
    add "length" FLOAT
    ...

    更新列

    UPDATE table1
       set pop1 = a.pop
      from 
    (select "OID", pop
       from "table2"
      where  pop != 0) a
     where table1."OID" = a."OID"

    列类型转换

    1)表格新建一列
    update table1
    set num1 = num1::integer
    
    2)直接修改
    alter table table_1
    alter column num1 type integer

    查询表的字段类型

    select column_name, data_type 
    from information_schema.columns 
    where table_name = 'gridbuffer_wgs'

    设置null为零值

    update schema."table_name" 
       set "column_name" = 0
     where "column_name" is null;

    插入当天日期date:

    update sqs.sqs_shopmall set version=current_date;

    截取字符:

    update sqs.sqs_poi2 set prcode=substring(prcode,1,2)  (index从1开始,1保留)
    空字符串处理:
    update property set memorial_no = btrim(memorial_no, ' ') where memorial_no like ' %'
    update property set memorial_no = 2  where memorial_no= btrim(memorial_no, ' ')
    多条件模糊查询:
    SELECT * FROM table WHERE column LIKE 'Text%' OR column LIKE 'Link%' OR column LIKE 'Hello%' OR column LIKE '%World%'

    只查询前几条数据:

    LIMIT num  OFFSET startNum //可以实现指定条数开始取几条
     替换replace:
    update cz_basedata_test2 set cnty=replace(cnty,city,'')//不可为null,否则整列将会变为null

    3 数值统计

    类别统计 (case when语句)

    select b.id, sum(case when c.category = '公交车站' then 1 end) "公交车站" 
      from schema."table_name" b left join schema.table_2 c
                                        on st_intersects(b.geom, c.geom)
    and c.cityname = '上海'
    group by b.id

    4 PostGIS空间操作

    查询投影类型
    select st_srid(geom) from table

    投影类型转换

    update table_1
       set geom_4326 = st_transform(geom, 4326)

    计算路网长度

    select b.id, sum(st_length(st_intersection(geography(b.geom),a.geom))) length
      from schema."table1" b  left join schema2.table2 a
                                     on st_intersects(a.geom, b.geom)
    and a.city = '深圳'
    group by b.id

    计算相交面积

    ST_Area(ST_INTERSECTION(b.geom_gcj, c.geom))

    计算缓冲区

    st_buffer(geography(a.geom_gcj), 3000)
    -- 计算完后需要再进行geometry转换
    geometry(st_buffer(geography(a.geom_gcj), 5000))
    计算面积(根据geometry计算面积)
    update sqs.sqs_town 
       set totarea=st_area(geography(geom))/1000000
    计算距离(包含经纬度转换地理坐标)
    select name,st_astext(geom) as geomtext, st_distance(Geography(ST_PointFromText('POINT(121.43543 31.2399)',4326)),Geography(geom)) as distance from table_1
     where ST_Intersects(st_buffer(geography(ST_PointFromText('POINT(121.43543 31.2399)',4326)),3000),geom) 
       and dlvydate> date '2010-01-01'
    交叉计算:
    select st_astext(geom) as GeomText ,"County" ,jdlon as Centerx , jdlat as Centery ,( people::DECIMAL/area::DECIMAL*1000000.0 )::INT as Density from vtown
    where ST_Intersects(st_buffer( geography( ST_PointFromText('POINT(121.492498 31.229649)',4326)),4000),geom)
     
    数组转换为geometry(对百度数组也适用):
    select  name from sqs.sqs_pipeline where ST_Intersects(ST_GeomFromText('POLYGON((121.550391 31.048365,121.845897 31.063214,121.734364 30.859081,121.580286 30.940391,121.550391 31.048365))',4326),geom)
    select  name from sqs.sqs_pipeline where ST_Within(geom,ST_PolygonFromText('POLYGON((121.550391 31.048365,121.845897 31.063214,121.734364 30.859081,121.580286 30.940391,121.550391 31.048365))',4326)) and status=1 
    根据geom计算x、y:
    update sqs.sqs_poi set lng=st_x(geom),lat=st_y(geom) 
    获取点geometry的点x、y坐标:
    select *, st_x(geom), st_y(geom) 
    from sqs.sqs_store
    where status=1
    and username='admin'
    and ST_Within(geom,ST_GeomFromText('POLYGON((121.289092 31.38394,121.333648 31.275857,121.462429 31.333862,121.459267 31.38172,121.379928 31.405393,121.289092 31.38394))',4326)) 
    点转换为geometry:
    update sqs.sqs_pipeline 
    set pipelineid='pp00000008', name='admin', geom=ST_PointFromText('POINT(121.845897 31.063214 )',4326),status=1;
    update sqs.yichang
    set geom=ST_PointFromText('POINT('||lng||' '||lat||')',4326); update sqs.sqs_pipeline_wfztest set geomta500=ST_PolygonFromText(st_astext(st_buffer(geography(ST_PointFromText('POINT('||st_x(geom)||' '||st_y(geom)||')',4326)), 500 )) ,4326); String sql = "select st_Area(Geography(st_intersection(a.geom,b.geom)))/st_Area(geography(a.geom)) as per100 " + "from sqs.sqs_city a, "+tablename +" b " + "where st_Intersects(a.geom,b.geom) and b."+id+"='"+ storeid+ "' and b.username='"+username +"' and b.status=1" ; String sql = "select ceil(totarea) as totarea,st_asgeojson(b.geom) as geometry, ceil(st_distance(Geography(ST_PointFromText('POINT(" + x + " " + y +")',4326)),Geography(b.geom))) as distance from sqs.sqs_newhouse b " + "where ST_Intersects(ST_PolygonFromText('POLYGON(("+p+"))',4326), b.geom) and " +"dlvydate > '2010-1-1' " + "order by tothh desc " + "limit 10";
    学习资源
    · PostgreSQL 10.1 手册
    · PostgreSQL 学习手册笔记
    https://www.cnblogs.com/stephen-liu74/archive/2012/06/08/2315679.html
    · PostgreSQL 常用操作总结
    · PostGIS Reference
     
     
    未完待续...
     

  • 相关阅读:
    Goahead在linux环境下安装部署
    vim卡住怎么办
    Clickhouse 实现 row number功能
    JavaScript ES6 模块化
    MySQL012事务的四个基本特征是什么
    MySQL015简述mysql中索引类型有哪些,以及对数据库的性能的影响
    MySQL010MySQL执行计划怎么看
    JavaScript ES6 Promise
    MySQL011如何处理MySQL的慢查询
    MySQL009MySQL为什么需要主从复制和读写分离
  • 原文地址:https://www.cnblogs.com/geo-will/p/9535421.html
Copyright © 2020-2023  润新知