1 <?xml version="1.0" encoding="UTF-8"?>
\r
2 <chapter xml:id="intro_to_sql" xmlns="http://docbook.org/ns/docbook" version="5.0" xml:lang="EN"
\r
3 xmlns:xi="http://www.w3.org/2001/XInclude" xmlns:xlink="http://www.w3.org/1999/xlink">
\r
5 <title>Introduction to SQL for Evergreen Administrators</title>
\r
7 <abstract id="itnroSQL_abstract">
\r
8 <simpara>This chapter was taken from Dan Scott's <emphasis>Introduction to SQL for Evergreen Administrators</emphasis>, February 2010.</simpara>
\r
10 <section id="intro_to_databases">
\r
11 <title>Introduction to SQL Databases</title>
\r
13 <title>Introduction</title>
\r
14 <simpara>Over time, the SQL database has become the standard method of storing,
\r
15 retrieving, and processing raw data for applications. Ranging from embedded
\r
16 databases such as SQLite and Apache Derby, to enterprise databases such as
\r
17 Oracle and IBM DB2, any SQL database offers basic advantages to application
\r
18 developers such as standard interfaces (Structured Query Language (SQL), Java
\r
19 Database Connectivity (JDBC), Open Database Connectivity (ODBC), Perl Database
\r
20 Independent Interface (DBI)), a standard conceptual model of data (tables,
\r
21 fields, relationships, constraints, etc), performance in storing and retrieving
\r
22 data, concurrent access, etc.</simpara>
\r
23 <simpara>Evergreen is built on PostgreSQL, an open source SQL database that began as
\r
24 <literal>POSTGRES</literal> at the University of California at Berkeley in 1986 as a research
\r
25 project led by Professor Michael Stonebraker. A SQL interface was added to a
\r
26 fork of the original POSTGRES Berkelely code in 1994, and in 1996 the project
\r
27 was renamed PostgreSQL.</simpara>
\r
29 <simplesect id="_tables">
\r
30 <title>Tables</title>
\r
31 <simpara>The table is the cornerstone of a SQL database. Conceptually, a database table
\r
32 is similar to a single sheet in a spreadsheet: every table has one or more
\r
33 columns, with each row in the table containing values for each column. Each
\r
34 column in a table defines an attribute corresponding to a particular data type.</simpara>
\r
35 <simpara>We’ll insert a row into a table, then display the resulting contents. Don’t
\r
36 worry if the INSERT statement is completely unfamiliar, we’ll talk more about
\r
37 the syntax of the insert statement later.</simpara>
\r
38 <formalpara><title><literal>actor.usr_note</literal> database table</title><para>
\r
39 <programlisting language="sql" linenumbering="unnumbered">evergreen=# INSERT INTO actor.usr_note (usr, creator, pub, title, value)
\r
40 VALUES (1, 1, TRUE, 'Who is this guy?', 'He''s the administrator!');
\r
42 evergreen=# select id, usr, creator, pub, title, value from actor.usr_note;
\r
43 id | usr | creator | pub | title | value
\r
44 ----+-----+---------+-----+------------------+-------------------------
\r
45 1 | 1 | 1 | t | Who is this guy? | He's the administrator!
\r
46 (1 rows)</programlisting>
\r
47 </para></formalpara>
\r
48 <simpara>PostgreSQL supports table inheritance, which lets you define tables that
\r
49 inherit the column definitions of a given parent table. A search of the data in
\r
50 the parent table includes the data in the child tables. Evergreen uses table
\r
51 inheritance: for example, the <literal>action.circulation</literal> table is a child of the
\r
52 <literal>money.billable_xact</literal> table, and the <literal>money.*_payment</literal> tables all inherit from
\r
53 the <literal>money.payment</literal> parent table.</simpara>
\r
55 <simplesect id="_schemas">
\r
56 <title>Schemas</title>
\r
57 <simpara>PostgreSQL, like most SQL databases, supports the use of schema names to group
\r
58 collections of tables and other database objects together. You might think of
\r
59 schemas as namespaces if you’re a programmer; or you might think of the schema
\r
60 / table / column relationship like the area code / exchange / local number
\r
61 structure of a telephone number.</simpara>
\r
64 rowsep="1" colsep="1"
\r
66 <title>Examples: database object names</title>
\r
67 <?dbhtml table-width="80%"?>
\r
68 <?dbfo table-width="80%"?>
\r
70 <colspec colname="col_1" colwidth="85*"/>
\r
71 <colspec colname="col_2" colwidth="85*"/>
\r
72 <colspec colname="col_3" colwidth="85*"/>
\r
73 <colspec colname="col_4" colwidth="85*"/>
\r
76 <entry align="left" valign="top">Full name </entry>
\r
77 <entry align="left" valign="top">Schema name </entry>
\r
78 <entry align="left" valign="top">Table name </entry>
\r
79 <entry align="left" valign="top">Field name</entry>
\r
84 <entry align="left" valign="top"><simpara>actor.usr_note.title</simpara></entry>
\r
85 <entry align="left" valign="top"><simpara>actor</simpara></entry>
\r
86 <entry align="left" valign="top"><simpara>usr_note</simpara></entry>
\r
87 <entry align="left" valign="top"><simpara>title</simpara></entry>
\r
90 <entry align="left" valign="top"><simpara>biblio.record_entry.marc</simpara></entry>
\r
91 <entry align="left" valign="top"><simpara>biblio</simpara></entry>
\r
92 <entry align="left" valign="top"><simpara>record_entry</simpara></entry>
\r
93 <entry align="left" valign="top"><simpara>marc</simpara></entry>
\r
98 <simpara>The default schema name in PostgreSQL is <literal>public</literal>, so if you do not specify a
\r
99 schema name when creating or accessing a database object, PostgreSQL will use
\r
100 the <literal>public</literal> schema. As a result, you might not find the object that you’re
\r
101 looking for if you don’t use the appropriate schema.</simpara>
\r
102 <formalpara><title>Example: Creating a table without a specific schema</title><para>
\r
103 <programlisting language="sql" linenumbering="unnumbered">evergreen=# CREATE TABLE foobar (foo TEXT, bar TEXT);
\r
105 evergreen=# \d foobar
\r
106 Table "public.foobar"
\r
107 Column | Type | Modifiers
\r
108 --------+------+-----------
\r
110 bar | text |</programlisting>
\r
111 </para></formalpara>
\r
112 <formalpara><title>Example: Trying to access a unqualified table outside of the public schema</title><para>
\r
113 <programlisting language="sql" linenumbering="unnumbered">evergreen=# SELECT * FROM usr_note;
\r
114 ERROR: relation "usr_note" does not exist
\r
115 LINE 1: SELECT * FROM usr_note;
\r
117 </para></formalpara>
\r
118 <simpara>Evergreen uses schemas to organize all of its tables with mostly intuitive,
\r
119 if short, schema names. Here’s the current (as of 2010-01-03) list of schemas
\r
120 used by Evergreen:</simpara>
\r
123 rowsep="1" colsep="1"
\r
125 <title>Evergreen schema names</title>
\r
126 <?dbhtml table-width="80%"?>
\r
127 <?dbfo table-width="80%"?>
\r
129 <colspec colname="col_1" colwidth="170*"/>
\r
130 <colspec colname="col_2" colwidth="170*"/>
\r
133 <entry align="left" valign="top">Schema name </entry>
\r
134 <entry align="left" valign="top">Description</entry>
\r
139 <entry align="left" valign="top"><simpara><literal>acq</literal></simpara></entry>
\r
140 <entry align="left" valign="top"><simpara>Acquisitions</simpara></entry>
\r
143 <entry align="left" valign="top"><simpara><literal>action</literal></simpara></entry>
\r
144 <entry align="left" valign="top"><simpara>Circulation actions</simpara></entry>
\r
147 <entry align="left" valign="top"><simpara><literal>action_trigger</literal></simpara></entry>
\r
148 <entry align="left" valign="top"><simpara>Event mechanisms</simpara></entry>
\r
151 <entry align="left" valign="top"><simpara><literal>actor</literal></simpara></entry>
\r
152 <entry align="left" valign="top"><simpara>Evergreen users and organization units</simpara></entry>
\r
155 <entry align="left" valign="top"><simpara><literal>asset</literal></simpara></entry>
\r
156 <entry align="left" valign="top"><simpara>Call numbers and copies</simpara></entry>
\r
159 <entry align="left" valign="top"><simpara><literal>auditor</literal></simpara></entry>
\r
160 <entry align="left" valign="top"><simpara>Track history of changes to selected tables</simpara></entry>
\r
163 <entry align="left" valign="top"><simpara><literal>authority</literal></simpara></entry>
\r
164 <entry align="left" valign="top"><simpara>Authority records</simpara></entry>
\r
167 <entry align="left" valign="top"><simpara><literal>biblio</literal></simpara></entry>
\r
168 <entry align="left" valign="top"><simpara>Bibliographic records</simpara></entry>
\r
171 <entry align="left" valign="top"><simpara><literal>booking</literal></simpara></entry>
\r
172 <entry align="left" valign="top"><simpara>Resource bookings</simpara></entry>
\r
175 <entry align="left" valign="top"><simpara><literal>config</literal></simpara></entry>
\r
176 <entry align="left" valign="top"><simpara>Evergreen configurable options</simpara></entry>
\r
179 <entry align="left" valign="top"><simpara><literal>container</literal></simpara></entry>
\r
180 <entry align="left" valign="top"><simpara>Buckets for records, call numbers, copies, and users</simpara></entry>
\r
183 <entry align="left" valign="top"><simpara><literal>extend_reporter</literal></simpara></entry>
\r
184 <entry align="left" valign="top"><simpara>Extra views for report definitions</simpara></entry>
\r
187 <entry align="left" valign="top"><simpara><literal>metabib</literal></simpara></entry>
\r
188 <entry align="left" valign="top"><simpara>Metadata about bibliographic records</simpara></entry>
\r
191 <entry align="left" valign="top"><simpara><literal>money</literal></simpara></entry>
\r
192 <entry align="left" valign="top"><simpara>Fines and bills</simpara></entry>
\r
195 <entry align="left" valign="top"><simpara><literal>offline</literal></simpara></entry>
\r
196 <entry align="left" valign="top"><simpara>Offline transactions</simpara></entry>
\r
199 <entry align="left" valign="top"><simpara><literal>permission</literal></simpara></entry>
\r
200 <entry align="left" valign="top"><simpara>User permissions</simpara></entry>
\r
203 <entry align="left" valign="top"><simpara><literal>query</literal></simpara></entry>
\r
204 <entry align="left" valign="top"><simpara>Stored SQL statements</simpara></entry>
\r
207 <entry align="left" valign="top"><simpara><literal>reporter</literal></simpara></entry>
\r
208 <entry align="left" valign="top"><simpara>Report definitions</simpara></entry>
\r
211 <entry align="left" valign="top"><simpara><literal>search</literal></simpara></entry>
\r
212 <entry align="left" valign="top"><simpara>Search functions</simpara></entry>
\r
215 <entry align="left" valign="top"><simpara><literal>serial</literal></simpara></entry>
\r
216 <entry align="left" valign="top"><simpara>Serial MFHD records</simpara></entry>
\r
219 <entry align="left" valign="top"><simpara><literal>stats</literal></simpara></entry>
\r
220 <entry align="left" valign="top"><simpara>Convenient views of circulation and asset statistics</simpara></entry>
\r
223 <entry align="left" valign="top"><simpara><literal>vandelay</literal></simpara></entry>
\r
224 <entry align="left" valign="top"><simpara>MARC batch importer and exporter</simpara></entry>
\r
229 <note><simpara>The term <emphasis>schema</emphasis> has two meanings in the world of SQL databases. We have
\r
230 discussed the schema as a conceptual grouping of tables and other database
\r
231 objects within a given namespace; for example, "the <emphasis role="strong">actor</emphasis> schema contains the
\r
232 tables and functions related to users and organizational units". Another common
\r
233 usage of <emphasis>schema</emphasis> is to refer to the entire data model for a given database;
\r
234 for example, "the Evergreen database schema".</simpara></note>
\r
236 <simplesect id="_columns">
\r
237 <title>Columns</title>
\r
238 <simpara>Each column definition consists of:</simpara>
\r
247 (optionally) a default value to be used whenever a row is inserted that
\r
248 does not contain a specific value
\r
253 (optionally) one or more constraints on the values beyond data type
\r
257 <simpara>Although PostgreSQL supports dozens of data types, Evergreen makes our life
\r
258 easier by only using a handful.</simpara>
\r
261 rowsep="1" colsep="1"
\r
263 <title>PostgreSQL data types used by Evergreen</title>
\r
264 <?dbhtml table-width="90%"?>
\r
265 <?dbfo table-width="90%"?>
\r
267 <colspec colname="col_1" colwidth="77*"/>
\r
268 <colspec colname="col_2" colwidth="77*"/>
\r
269 <colspec colname="col_3" colwidth="230*"/>
\r
272 <entry align="left" valign="top">Type name </entry>
\r
273 <entry align="left" valign="top">Description </entry>
\r
274 <entry align="left" valign="top">Limits</entry>
\r
279 <entry align="left" valign="top"><simpara><literal>INTEGER</literal></simpara></entry>
\r
280 <entry align="left" valign="top"><simpara>Medium integer</simpara></entry>
\r
281 <entry align="left" valign="top"><simpara>-2147483648 to +2147483647</simpara></entry>
\r
284 <entry align="left" valign="top"><simpara><literal>BIGINT</literal></simpara></entry>
\r
285 <entry align="left" valign="top"><simpara>Large integer</simpara></entry>
\r
286 <entry align="left" valign="top"><simpara>-9223372036854775808 to 9223372036854775807</simpara></entry>
\r
289 <entry align="left" valign="top"><simpara><literal>SERIAL</literal></simpara></entry>
\r
290 <entry align="left" valign="top"><simpara>Sequential integer</simpara></entry>
\r
291 <entry align="left" valign="top"><simpara>1 to 2147483647</simpara></entry>
\r
294 <entry align="left" valign="top"><simpara><literal>BIGSERIAL</literal></simpara></entry>
\r
295 <entry align="left" valign="top"><simpara>Large sequential integer</simpara></entry>
\r
296 <entry align="left" valign="top"><simpara>1 to 9223372036854775807</simpara></entry>
\r
299 <entry align="left" valign="top"><simpara><literal>TEXT</literal></simpara></entry>
\r
300 <entry align="left" valign="top"><simpara>Variable length character data</simpara></entry>
\r
301 <entry align="left" valign="top"><simpara>Unlimited length</simpara></entry>
\r
304 <entry align="left" valign="top"><simpara><literal>BOOL</literal></simpara></entry>
\r
305 <entry align="left" valign="top"><simpara>Boolean</simpara></entry>
\r
306 <entry align="left" valign="top"><simpara>TRUE or FALSE</simpara></entry>
\r
309 <entry align="left" valign="top"><simpara><literal>TIMESTAMP WITH TIME ZONE</literal></simpara></entry>
\r
310 <entry align="left" valign="top"><simpara>Timestamp</simpara></entry>
\r
311 <entry align="left" valign="top"><simpara>4713 BC to 294276 AD</simpara></entry>
\r
314 <entry align="left" valign="top"><simpara><literal>TIME</literal></simpara></entry>
\r
315 <entry align="left" valign="top"><simpara>Time</simpara></entry>
\r
316 <entry align="left" valign="top"><simpara>Expressed in HH:MM:SS</simpara></entry>
\r
319 <entry align="left" valign="top"><simpara><literal>NUMERIC</literal>(precision, scale)</simpara></entry>
\r
320 <entry align="left" valign="top"><simpara>Decimal</simpara></entry>
\r
321 <entry align="left" valign="top"><simpara>Up to 1000 digits of precision. In Evergreen mostly used for money
\r
322 values, with a precision of 6 and a scale of 2 (<literal>####.##</literal>).</simpara></entry>
\r
327 <simpara>Full details about these data types are available from the
\r
328 <ulink url="http://www.postgresql.org/docs/8.4/static/datatype.html">data types section of
\r
329 the PostgreSQL manual</ulink>.</simpara>
\r
331 <simplesect id="_constraints">
\r
332 <title>Constraints</title>
\r
333 <simplesect id="_prevent_null_values">
\r
334 <title>Prevent NULL values</title>
\r
335 <simpara>A column definition may include the constraint <literal>NOT NULL</literal> to prevent NULL
\r
336 values. In PostgreSQL, a NULL value is not the equivalent of zero or false or
\r
337 an empty string; it is an explicit non-value with special properties. We’ll
\r
338 talk more about how to work with NULL values when we get to queries.</simpara>
\r
340 <simplesect id="_primary_key">
\r
341 <title>Primary key</title>
\r
342 <simpara>Every table can have at most one primary key. A primary key consists of one or
\r
343 more columns which together uniquely identify each row in a table. If you
\r
344 attempt to insert a row into a table that would create a duplicate or NULL
\r
345 primary key entry, the database rejects the row and returns an error.</simpara>
\r
346 <simpara>Natural primary keys are drawn from the intrinsic properties of the data being
\r
347 modelled. For example, some potential natural primary keys for a table that
\r
348 contains people would be:</simpara>
\r
351 rowsep="1" colsep="1"
\r
353 <title>Example: Some potential natural primary keys for a table of people</title>
\r
354 <?dbhtml table-width="90%"?>
\r
355 <?dbfo table-width="90%"?>
\r
357 <colspec colname="col_1" colwidth="77*"/>
\r
358 <colspec colname="col_2" colwidth="153*"/>
\r
359 <colspec colname="col_3" colwidth="153*"/>
\r
362 <entry align="left" valign="top">Natural key </entry>
\r
363 <entry align="left" valign="top">Pros </entry>
\r
364 <entry align="left" valign="top">Cons</entry>
\r
369 <entry align="left" valign="top"><simpara>First name, last name, address</simpara></entry>
\r
370 <entry align="left" valign="top"><simpara>No two people with the same name would ever live at the same address, right?</simpara></entry>
\r
371 <entry align="left" valign="top"><simpara>Lots of columns force data duplication in referencing tables</simpara></entry>
\r
374 <entry align="left" valign="top"><simpara>SSN or driver’s license</simpara></entry>
\r
375 <entry align="left" valign="top"><simpara>These are guaranteed to be unique</simpara></entry>
\r
376 <entry align="left" valign="top"><simpara>Lots of people don’t have an SSN or a driver’s license</simpara></entry>
\r
381 <simpara>To avoid problems with natural keys, many applications instead define surrogate
\r
382 primary keys. A surrogate primary keys is a column with an autoincrementing
\r
383 integer value added to a table definition that ensures uniqueness.</simpara>
\r
384 <simpara>Evergreen uses surrogate keys (a column named <literal>id</literal> with a <literal>SERIAL</literal> data type)
\r
385 for most of its tables.</simpara>
\r
387 <simplesect id="_foreign_keys">
\r
388 <title>Foreign keys</title>
\r
389 <simpara>Every table can contain zero or more foreign keys: one or more columns that
\r
390 refer to the primary key of another table.</simpara>
\r
391 <simpara>For example, let’s consider Evergreen’s modelling of the basic relationship
\r
392 between copies, call numbers, and bibliographic records. Bibliographic records
\r
393 contained in the <literal>biblio.record_entry</literal> table can have call numbers attached to
\r
394 them. Call numbers are contained in the <literal>asset.call_number</literal> table, and they can
\r
395 have copies attached to them. Copies are contained in the <literal>asset.copy</literal> table.</simpara>
\r
398 rowsep="1" colsep="1"
\r
400 <title>Example: Evergreen’s copy / call number / bibliographic record relationships</title>
\r
401 <?dbhtml table-width="100%"?>
\r
402 <?dbfo table-width="100%"?>
\r
404 <colspec colname="col_1" colwidth="106*"/>
\r
405 <colspec colname="col_2" colwidth="106*"/>
\r
406 <colspec colname="col_3" colwidth="106*"/>
\r
407 <colspec colname="col_4" colwidth="106*"/>
\r
410 <entry align="left" valign="top">Table </entry>
\r
411 <entry align="left" valign="top">Primary key </entry>
\r
412 <entry align="left" valign="top">Column with a foreign key </entry>
\r
413 <entry align="left" valign="top">Points to</entry>
\r
418 <entry align="left" valign="top"><simpara>asset.copy</simpara></entry>
\r
419 <entry align="left" valign="top"><simpara>asset.copy.id</simpara></entry>
\r
420 <entry align="left" valign="top"><simpara>asset.copy.call_number</simpara></entry>
\r
421 <entry align="left" valign="top"><simpara>asset.call_number.id</simpara></entry>
\r
424 <entry align="left" valign="top"><simpara>asset.call_number</simpara></entry>
\r
425 <entry align="left" valign="top"><simpara>asset.call_number.id</simpara></entry>
\r
426 <entry align="left" valign="top"><simpara>asset.call_number.record</simpara></entry>
\r
427 <entry align="left" valign="top"><simpara>biblio.record_entry.id</simpara></entry>
\r
430 <entry align="left" valign="top"><simpara>biblio.record_entry</simpara></entry>
\r
431 <entry align="left" valign="top"><simpara>biblio.record_entry.id</simpara></entry>
\r
432 <entry align="left" valign="top"><simpara></simpara></entry>
\r
433 <entry align="left" valign="top"><simpara></simpara></entry>
\r
439 <simplesect id="_check_constraints">
\r
440 <title>Check constraints</title>
\r
441 <simpara>PostgreSQL enables you to define rules to ensure that the value to be inserted
\r
442 or updated meets certain conditions. For example, you can ensure that an
\r
443 incoming integer value is within a specific range, or that a ZIP code matches a
\r
444 particular pattern.</simpara>
\r
447 <simplesect id="_deconstructing_a_table_definition_statement">
\r
448 <title>Deconstructing a table definition statement</title>
\r
449 <simpara>The <literal>actor.org_address</literal> table is a simple table in the Evergreen schema that
\r
450 we can use as a concrete example of many of the properties of databases that
\r
451 we have discussed so far.</simpara>
\r
452 <programlisting language="sql" linenumbering="unnumbered">CREATE TABLE actor.org_address (
\r
453 id SERIAL PRIMARY KEY, <co id="sqlCO1-1"/>
\r
454 valid BOOL NOT NULL DEFAULT TRUE, <co id="sqlCO1-2"/>
\r
455 address_type TEXT NOT NULL DEFAULT 'MAILING', <co id="sqlCO1-3"/>
\r
456 org_unit INT NOT NULL REFERENCES actor.org_unit (id) <co id="sqlCO1-4"/>
\r
457 DEFERRABLE INITIALLY DEFERRED,
\r
458 street1 TEXT NOT NULL,
\r
459 street2 TEXT, <co id="sqlCO1-5"/>
\r
460 city TEXT NOT NULL,
\r
462 state TEXT NOT NULL,
\r
463 country TEXT NOT NULL,
\r
464 post_code TEXT NOT NULL
\r
465 );</programlisting>
\r
467 <callout arearefs="sqlCO1-1">
\r
469 The column named <literal>id</literal> is defined with a special data type of <literal>SERIAL</literal>; if
\r
470 given no value when a row is inserted into a table, the database automatically
\r
471 generates the next sequential integer value for the column. <literal>SERIAL</literal> is a
\r
472 popular data type for a primary key because it is guaranteed to be unique - and
\r
473 indeed, the constraint for this column identifies it as the <literal>PRIMARY KEY</literal>.
\r
476 <callout arearefs="sqlCO1-2">
\r
478 The data type <literal>BOOL</literal> defines a boolean value: <literal>TRUE</literal> or <literal>FALSE</literal> are the only
\r
479 acceptable values for the column. The constraint <literal>NOT NULL</literal> instructs the
\r
480 database to prevent the column from ever containing a NULL value. The column
\r
481 property <literal>DEFAULT TRUE</literal> instructs the database to automatically set the value
\r
482 of the column to <literal>TRUE</literal> if no value is provided.
\r
485 <callout arearefs="sqlCO1-3">
\r
487 The data type <literal>TEXT</literal> defines a text column of practically unlimited length.
\r
488 As with the previous column, there is a <literal>NOT NULL</literal> constraint, and a default
\r
489 value of <literal>'MAILING'</literal> will result if no other value is supplied.
\r
492 <callout arearefs="sqlCO1-4">
\r
494 The <literal>REFERENCES actor.org_unit (id)</literal> clause indicates that this column has a
\r
495 foreign key relationship to the <literal>actor.org_unit</literal> table, and that the value of
\r
496 this column in every row in this table must have a corresponding value in the
\r
497 <literal>id</literal> column in the referenced table (<literal>actor.org_unit</literal>).
\r
500 <callout arearefs="sqlCO1-5">
\r
502 The column named <literal>street2</literal> demonstrates that not all columns have constraints
\r
503 beyond data type. In this case, the column is allowed to be NULL or to contain a
\r
504 <literal>TEXT</literal> value.
\r
509 <simplesect id="_displaying_a_table_definition_using_literal_psql_literal">
\r
510 <title>Displaying a table definition using <literal>psql</literal></title>
\r
511 <simpara>The <literal>psql</literal> command-line interface is the preferred method for accessing
\r
512 PostgreSQL databases. It offers features like tab-completion, readline support
\r
513 for recalling previous commands, flexible input and output formats, and
\r
514 is accessible via a standard SSH session.</simpara>
\r
515 <simpara>If you press the <literal>Tab</literal> key once after typing one or more characters of the
\r
516 database object name, <literal>psql</literal> automatically completes the name if there are no
\r
517 other matches. If there are other matches for your current input, nothing
\r
518 happens until you press the <literal>Tab</literal> key a second time, at which point <literal>psql</literal>
\r
519 displays all of the matches for your current input.</simpara>
\r
520 <simpara>To display the definition of a database object such as a table, issue the
\r
521 command <literal>\d _object-name_</literal>. For example, to display the definition of the
\r
522 actor.usr_note table:</simpara>
\r
523 <programlisting language="sh" linenumbering="unnumbered">$ psql evergreen <co id="sqlCO2-1"/>
\r
525 Type "help" for help.
\r
527 evergreen=# \d actor.usr_note <co id="sqlCO2-2"/>
\r
528 Table "actor.usr_note"
\r
529 Column | Type | Modifiers
\r
530 -------------+--------------------------+-------------------------------------------------------------
\r
531 id | bigint | not null default nextval('actor.usr_note_id_seq'::regclass)
\r
532 usr | bigint | not null
\r
533 creator | bigint | not null
\r
534 create_date | timestamp with time zone | default now()
\r
535 pub | boolean | not null default false
\r
536 title | text | not null
\r
537 value | text | not null
\r
539 "usr_note_pkey" PRIMARY KEY, btree (id)
\r
540 "actor_usr_note_creator_idx" btree (creator)
\r
541 "actor_usr_note_usr_idx" btree (usr)
\r
542 Foreign-key constraints:
\r
543 "usr_note_creator_fkey" FOREIGN KEY (creator) REFERENCES actor.usr(id) ON DELETE CASCADE DEFERRABLE INITIALLY DEFERRED
\r
544 "usr_note_usr_fkey" FOREIGN KEY (usr) REFERENCES actor.usr(id) ON DELETE CASCADE DEFERRABLE INITIALLY DEFERRED
\r
546 evergreen=# \q <co id="sqlCO2-3"/>
\r
549 <callout arearefs="sqlCO2-1">
\r
551 This is the most basic connection to a PostgreSQL database. You can use a
\r
552 number of other flags to specify user name, hostname, port, and other options.
\r
555 <callout arearefs="sqlCO2-2">
\r
557 The <literal>\d</literal> command displays the definition of a database object.
\r
560 <callout arearefs="sqlCO2-3">
\r
562 The <literal>\q</literal> command quits the <literal>psql</literal> session and returns you to the shell prompt.
\r
568 <section id="basic_sql_queries">
\r
569 <title>Basic SQL queries</title>
\r
570 <simplesect id="_the_select_statement">
\r
571 <title>The SELECT statement</title>
\r
572 <simpara>The SELECT statement is the basic tool for retrieving information from a
\r
573 database. The syntax for most SELECT statements is:</simpara>
\r
575 <literallayout><literal>SELECT</literal> [<emphasis>columns(s)</emphasis>]
\r
576 <literal>FROM</literal> [<emphasis>table(s)</emphasis>]
\r
577 [<literal>WHERE</literal> <emphasis>condition(s)</emphasis>]
\r
578 [<literal>GROUP BY</literal> <emphasis>columns(s)</emphasis>]
\r
579 [<literal>HAVING</literal> <emphasis>grouping-condition(s)</emphasis>]
\r
580 [<literal>ORDER BY</literal> <emphasis>column(s)</emphasis>]
\r
581 [<literal>LIMIT</literal> <emphasis>maximum-results</emphasis>]
\r
582 [<literal>OFFSET</literal> <emphasis>start-at-result-#</emphasis>]
\r
585 <simpara>For example, to select all of the columns for each row in the
\r
586 <literal>actor.usr_address</literal> table, issue the following query:</simpara>
\r
587 <programlisting language="sql" linenumbering="unnumbered">SELECT *
\r
588 FROM actor.usr_address
\r
591 <simplesect id="_selecting_particular_columns_from_a_table">
\r
592 <title>Selecting particular columns from a table</title>
\r
593 <simpara><literal>SELECT *</literal> returns all columns from all of the tables included in your query.
\r
594 However, quite often you will want to return only a subset of the possible
\r
595 columns. You can retrieve specific columns by listing the names of the columns
\r
596 you want after the <literal>SELECT</literal> keyword. Separate each column name with a comma.</simpara>
\r
597 <simpara>For example, to select just the city, county, and state from the
\r
598 actor.usr_address table, issue the following query:</simpara>
\r
599 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
600 FROM actor.usr_address
\r
603 <simplesect id="_sorting_results_with_the_order_by_clause">
\r
604 <title>Sorting results with the ORDER BY clause</title>
\r
605 <simpara>By default, a SELECT statement returns rows matching your query with no
\r
606 guarantee of any particular order in which they are returned. To force
\r
607 the rows to be returned in a particular order, use the ORDER BY clause
\r
608 to specify one or more columns to determine the sorting priority of the
\r
610 <simpara>For example, to sort the rows returned from your <literal>actor.usr_address</literal> query by
\r
611 city, with county and then zip code as the tie breakers, issue the
\r
612 following query:</simpara>
\r
613 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
614 FROM actor.usr_address
\r
615 ORDER BY city, county, post_code
\r
618 <simplesect id="_filtering_results_with_the_where_clause">
\r
619 <title>Filtering results with the WHERE clause</title>
\r
620 <simpara>Thus far, your results have been returning all of the rows in the table.
\r
621 Normally, however, you would want to restrict the rows that are returned to the
\r
622 subset of rows that match one or more conditions of your search. The <literal>WHERE</literal>
\r
623 clause enables you to specify a set of conditions that filter your query
\r
624 results. Each condition in the <literal>WHERE</literal> clause is an SQL expression that returns
\r
625 a boolean (true or false) value.</simpara>
\r
626 <simpara>For example, to restrict the results returned from your <literal>actor.usr_address</literal>
\r
627 query to only those rows containing a state value of <emphasis>Connecticut</emphasis>, issue the
\r
628 following query:</simpara>
\r
629 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
630 FROM actor.usr_address
\r
631 WHERE state = 'Connecticut'
\r
632 ORDER BY city, county, post_code
\r
634 <simpara>You can include more conditions in the <literal>WHERE</literal> clause with the <literal>OR</literal> and <literal>AND</literal>
\r
635 operators. For example, to further restrict the results returned from your
\r
636 <literal>actor.usr_address</literal> query to only those rows where the state column contains a
\r
637 value of <emphasis>Connecticut</emphasis> and the city column contains a value of <emphasis>Hartford</emphasis>,
\r
638 issue the following query:</simpara>
\r
639 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
640 FROM actor.usr_address
\r
641 WHERE state = 'Connecticut'
\r
642 AND city = 'Hartford'
\r
643 ORDER BY city, county, post_code
\r
645 <note><simpara>To return rows where the state is <emphasis>Connecticut</emphasis> and the city is <emphasis>Hartford</emphasis> or
\r
646 <emphasis>New Haven</emphasis>, you must use parentheses to explicitly group the city value
\r
647 conditions together, or else the database will evaluate the <literal>OR city = 'New
\r
648 Haven'</literal> clause entirely on its own and match all rows where the city column is
\r
649 <emphasis>New Haven</emphasis>, even though the state might not be <emphasis>Connecticut</emphasis>.</simpara></note>
\r
650 <formalpara><title>Trouble with OR</title><para>
\r
651 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
652 FROM actor.usr_address
\r
653 WHERE state = 'Connecticut'
\r
654 AND city = 'Hartford' OR city = 'New Haven'
\r
655 ORDER BY city, county, post_code
\r
658 -- Can return unwanted rows because the OR is not grouped!</programlisting>
\r
659 </para></formalpara>
\r
660 <formalpara><title>Grouped OR’ed conditions</title><para>
\r
661 <programlisting language="sql" linenumbering="unnumbered">SELECT city, county, state
\r
662 FROM actor.usr_address
\r
663 WHERE state = 'Connecticut'
\r
664 AND (city = 'Hartford' OR city = 'New Haven')
\r
665 ORDER BY city, county, post_code
\r
668 -- The parentheses ensure that the OR is applied to the cities, and the
\r
669 -- state in either case must be 'Connecticut'</programlisting>
\r
670 </para></formalpara>
\r
671 <simplesect id="_comparison_operators">
\r
672 <title>Comparison operators</title>
\r
673 <simpara>Here is a partial list of comparison operators that are commonly used in
\r
674 <literal>WHERE</literal> clauses:</simpara>
\r
675 <simplesect id="_comparing_two_scalar_values">
\r
676 <title>Comparing two scalar values</title>
\r
680 <literal>x = y</literal> (equal to)
\r
685 <literal>x != y</literal> (not equal to)
\r
690 <literal>x < y</literal> (less than)
\r
695 <literal>x > y</literal> (greater than)
\r
700 <literal>x LIKE y</literal> (TEXT value x matches a subset of TEXT y, where y is a string that
\r
701 can contain <emphasis>%</emphasis> as a wildcard for 0 or more characters, and <emphasis>_</emphasis> as a wildcard
\r
702 for a single character. For example, <literal>WHERE 'all you can eat fish and chips
\r
703 and a big stick' LIKE '%fish%stick'</literal> would return TRUE)
\r
708 <literal>x ILIKE y</literal> (like LIKE, but the comparison ignores upper-case / lower-case)
\r
713 <literal>x IN y</literal> (x is in the list of values y, where y can be a list or a SELECT
\r
714 statement that returns a list)
\r
721 <simplesect id="_null_values">
\r
722 <title>NULL values</title>
\r
723 <simpara>SQL databases have a special way of representing the value of a column that has
\r
724 no value: <literal>NULL</literal>. A <literal>NULL</literal> value is not equal to zero, and is not an empty
\r
725 string; it is equal to nothing, not even another <literal>NULL</literal>, because it has no value
\r
726 that can be compared.</simpara>
\r
727 <simpara>To return rows from a table where a given column is not <literal>NULL</literal>, use the
\r
728 <literal>IS NOT NULL</literal> comparison operator.</simpara>
\r
729 <formalpara><title>Retrieving rows where a column is not <literal>NULL</literal></title><para>
\r
730 <programlisting language="sql" linenumbering="unnumbered">SELECT id, first_given_name, family_name
\r
732 WHERE second_given_name IS NOT NULL
\r
734 </para></formalpara>
\r
735 <simpara>Similarly, to return rows from a table where a given column is <literal>NULL</literal>, use
\r
736 the <literal>IS NULL</literal> comparison operator.</simpara>
\r
737 <formalpara><title>Retrieving rows where a column is <literal>NULL</literal></title><para>
\r
738 <programlisting language="sql" linenumbering="unnumbered">SELECT id, first_given_name, second_given_name, family_name
\r
740 WHERE second_given_name IS NULL
\r
743 id | first_given_name | second_given_name | family_name
\r
744 ----+------------------+-------------------+----------------
\r
745 1 | Administrator | | System Account
\r
746 (1 row)</programlisting>
\r
747 </para></formalpara>
\r
748 <simpara>Notice that the <literal>NULL</literal> value in the output is displayed as empty space,
\r
749 indistinguishable from an empty string; this is the default display method in
\r
750 <literal>psql</literal>. You can change the behaviour of <literal>psql</literal> using the <literal>pset</literal> command:</simpara>
\r
751 <formalpara><title>Changing the way <literal>NULL</literal> values are displayed in <literal>psql</literal></title><para>
\r
752 <programlisting language="sql" linenumbering="unnumbered">evergreen=# \pset null '(null)'
\r
753 Null display is '(null)'.
\r
755 SELECT id, first_given_name, second_given_name, family_name
\r
757 WHERE second_given_name IS NULL
\r
760 id | first_given_name | second_given_name | family_name
\r
761 ----+------------------+-------------------+----------------
\r
762 1 | Administrator | (null) | System Account
\r
763 (1 row)</programlisting>
\r
764 </para></formalpara>
\r
765 <simpara>Database queries within programming languages such as Perl and C have
\r
766 special methods of checking for <literal>NULL</literal> values in returned results.</simpara>
\r
768 <simplesect id="_text_delimiter">
\r
769 <title>Text delimiter: '</title>
\r
770 <simpara>You might have noticed that we have been using the <literal>'</literal> character to delimit
\r
771 TEXT values and values such as dates and times that are TEXT values. Sometimes,
\r
772 however, your TEXT value itself contains a <literal>'</literal> character, such as the word
\r
773 <literal>you’re</literal>. To prevent the database from prematurely ending the TEXT value at the
\r
774 first <literal>'</literal> character and returning a syntax error, use another <literal>'</literal> character to
\r
775 escape the following <literal>'</literal> character.</simpara>
\r
776 <simpara>For example, to change the last name of a user in the <literal>actor.usr</literal> table to
\r
777 <literal>L’estat</literal>, issue the following SQL:</simpara>
\r
778 <formalpara><title>Escaping <literal>'</literal> in TEXT values</title><para>
\r
779 <programlisting language="sql" linenumbering="unnumbered">UPDATE actor.usr
\r
780 SET family_name = 'L''estat'
\r
783 FROM permission.grp_tree
\r
784 WHERE name = 'Vampire'
\r
787 </para></formalpara>
\r
788 <simpara>When you retrieve the row from the database, the value is displayed with just
\r
789 a single <literal>'</literal> character:</simpara>
\r
790 <programlisting language="sql" linenumbering="unnumbered">SELECT id, family_name
\r
792 WHERE family_name = 'L''estat'
\r
798 (1 row)</programlisting>
\r
800 <simplesect id="_grouping_and_eliminating_results_with_the_group_by_and_having_clauses">
\r
801 <title>Grouping and eliminating results with the GROUP BY and HAVING clauses</title>
\r
802 <simpara>The GROUP BY clause returns a unique set of results for the desired columns.
\r
803 This is most often used in conjunction with an aggregate function to present
\r
804 results for a range of values in a single query, rather than requiring you to
\r
805 issue one query per target value.</simpara>
\r
806 <formalpara><title>Returning unique results of a single column with <literal>GROUP BY</literal></title><para>
\r
807 <programlisting language="sql" linenumbering="unnumbered">SELECT grp
\r
808 FROM permission.grp_perm_map
\r
822 (8 rows)</programlisting>
\r
823 </para></formalpara>
\r
824 <simpara>While <literal>GROUP BY</literal> can be useful for a single column, it is more often used
\r
825 to return the distinct results across multiple columns. For example, the
\r
826 following query shows us which groups have permissions at each depth in
\r
827 the library hierarchy:</simpara>
\r
828 <formalpara><title>Returning unique results of multiple columns with <literal>GROUP BY</literal></title><para>
\r
829 <programlisting language="sql" linenumbering="unnumbered">SELECT grp, depth
\r
830 FROM permission.grp_perm_map
\r
831 GROUP BY grp, depth
\r
832 ORDER BY depth, grp;
\r
851 (15 rows)</programlisting>
\r
852 </para></formalpara>
\r
853 <simpara>Extending this further, you can use the <literal>COUNT()</literal> aggregate function to
\r
854 also return the number of times each unique combination of <literal>grp</literal> and <literal>depth</literal>
\r
855 appears in the table. <emphasis>Yes, this is a sneak peek at the use of aggregate
\r
856 functions! Keeners.</emphasis></simpara>
\r
857 <formalpara><title>Counting unique column combinations with <literal>GROUP BY</literal></title><para>
\r
858 <programlisting language="sql" linenumbering="unnumbered">SELECT grp, depth, COUNT(grp)
\r
859 FROM permission.grp_perm_map
\r
860 GROUP BY grp, depth
\r
861 ORDER BY depth, grp;
\r
863 grp | depth | count
\r
864 -----+-------+-------
\r
880 (15 rows)</programlisting>
\r
881 </para></formalpara>
\r
882 <simpara>You can use the <literal>WHERE</literal> clause to restrict the returned results before grouping
\r
883 is applied to the results. The following query restricts the results to those
\r
884 rows that have a depth of 0.</simpara>
\r
885 <formalpara><title>Using the <literal>WHERE</literal> clause with <literal>GROUP BY</literal></title><para>
\r
886 <programlisting language="sql" linenumbering="unnumbered">SELECT grp, COUNT(grp)
\r
887 FROM permission.grp_perm_map
\r
901 (6 rows)</programlisting>
\r
902 </para></formalpara>
\r
903 <simpara>To restrict results after grouping has been applied to the rows, use the
\r
904 <literal>HAVING</literal> clause; this is typically used to restrict results based on
\r
905 a comparison to the value returned by an aggregate function. For example,
\r
906 the following query restricts the returned rows to those that have more than
\r
907 5 occurrences of the same value for <literal>grp</literal> in the table.</simpara>
\r
908 <formalpara><title><literal>GROUP BY</literal> restricted by a <literal>HAVING</literal> clause</title><para>
\r
909 <programlisting language="sql" linenumbering="unnumbered">SELECT grp, COUNT(grp)
\r
910 FROM permission.grp_perm_map
\r
912 HAVING COUNT(grp) > 5
\r
923 (6 rows)</programlisting>
\r
924 </para></formalpara>
\r
926 <simplesect id="_eliminating_duplicate_results_with_the_distinct_keyword">
\r
927 <title>Eliminating duplicate results with the DISTINCT keyword</title>
\r
928 <simpara><literal>GROUP BY</literal> is one way of eliminating duplicate results from the rows returned
\r
929 by your query. The purpose of the <literal>DISTINCT</literal> keyword is to remove duplicate
\r
930 rows from the results of your query. However, it works, and it is easy - so if
\r
931 you just want a quick list of the unique set of values for a column or set of
\r
932 columns, the <literal>DISTINCT</literal> keyword might be appropriate.</simpara>
\r
933 <simpara>On the other hand, if you are getting duplicate rows back when you don’t expect
\r
934 them, then applying the <literal>DISTINCT</literal> keyword might be a sign that you are
\r
935 papering over a real problem.</simpara>
\r
936 <formalpara><title>Returning unique results of multiple columns with <literal>DISTINCT</literal></title><para>
\r
937 <programlisting language="sql" linenumbering="unnumbered">SELECT DISTINCT grp, depth
\r
938 FROM permission.grp_perm_map
\r
939 ORDER BY depth, grp
\r
959 (15 rows)</programlisting>
\r
960 </para></formalpara>
\r
962 <simplesect id="_paging_through_results_with_the_limit_and_offset_clauses">
\r
963 <title>Paging through results with the LIMIT and OFFSET clauses</title>
\r
964 <simpara>The <literal>LIMIT</literal> clause restricts the total number of rows returned from your query
\r
965 and is useful if you just want to list a subset of a large number of rows. For
\r
966 example, in the following query we list the five most frequently used
\r
967 circulation modifiers:</simpara>
\r
968 <formalpara><title>Using the <literal>LIMIT</literal> clause to restrict results</title><para>
\r
969 <programlisting language="sql" linenumbering="unnumbered">SELECT circ_modifier, COUNT(circ_modifier)
\r
971 GROUP BY circ_modifier
\r
976 circ_modifier | count
\r
977 ---------------+--------
\r
983 (5 rows)</programlisting>
\r
984 </para></formalpara>
\r
985 <simpara>When you use the <literal>LIMIT</literal> clause to restrict the total number of rows returned
\r
986 by your query, you can also use the <literal>OFFSET</literal> clause to determine which subset
\r
987 of the rows will be returned. The use of the <literal>OFFSET</literal> clause assumes that
\r
988 you’ve used the <literal>ORDER BY</literal> clause to impose order on the results.</simpara>
\r
989 <simpara>In the following example, we use the <literal>OFFSET</literal> clause to get results 6 through
\r
990 10 from the same query that we prevously executed.</simpara>
\r
991 <formalpara><title>Using the <literal>OFFSET</literal> clause to return a specific subset of rows</title><para>
\r
992 <programlisting language="sql" linenumbering="unnumbered">SELECT circ_modifier, COUNT(circ_modifier)
\r
994 GROUP BY circ_modifier
\r
1000 circ_modifier | count
\r
1001 ---------------+--------
\r
1002 LAW SERIAL | 102758
\r
1007 (5 rows)</programlisting>
\r
1008 </para></formalpara>
\r
1011 <section id="advanced_sql_queries">
\r
1012 <title>Advanced SQL queries</title>
\r
1013 <simplesect id="_transforming_column_values_with_functions">
\r
1014 <title>Transforming column values with functions</title>
\r
1015 <simpara>PostgreSQL includes many built-in functions for manipulating column data.
\r
1016 You can also create your own functions (and Evergreen does make use of
\r
1017 many custom functions). There are two types of functions used in
\r
1018 databases: scalar functions and aggregate functions.</simpara>
\r
1019 <simplesect id="_scalar_functions">
\r
1020 <title>Scalar functions</title>
\r
1021 <simpara>Scalar functions transform each value of the target column. If your query
\r
1022 would return 50 values for a column in a given query, and you modify your
\r
1023 query to apply a scalar function to the values returned for that column,
\r
1024 it will still return 50 values. For example, the UPPER() function,
\r
1025 used to convert text values to upper-case, modifies the results in the
\r
1026 following set of queries:</simpara>
\r
1027 <formalpara><title>Using the UPPER() scalar function to convert text values to upper-case</title><para>
\r
1028 <programlisting language="sql" linenumbering="unnumbered">-- First, without the UPPER() function for comparison
\r
1029 SELECT shortname, name
\r
1030 FROM actor.org_unit
\r
1035 -----------+-----------------------
\r
1036 CONS | Example Consortium
\r
1037 SYS1 | Example System 1
\r
1038 SYS2 | Example System 2
\r
1041 -- Now apply the UPPER() function to the name column
\r
1042 SELECT shortname, UPPER(name)
\r
1043 FROM actor.org_unit
\r
1048 -----------+--------------------
\r
1049 CONS | EXAMPLE CONSORTIUM
\r
1050 SYS1 | EXAMPLE SYSTEM 1
\r
1051 SYS2 | EXAMPLE SYSTEM 2
\r
1052 (3 rows)</programlisting>
\r
1053 </para></formalpara>
\r
1054 <simpara>There are so many scalar functions in PostgreSQL that we cannot cover them
\r
1055 all here, but we can list some of the most commonly used functions:</simpara>
\r
1059 || - concatenates two text values together
\r
1064 COALESCE() - returns the first non-NULL value from the list of arguments
\r
1069 LOWER() - returns a text value converted to lower-case
\r
1074 REPLACE() - returns a text value after replacing all occurrences of a given text value with a different text value
\r
1079 REGEXP_REPLACE() - returns a text value after being transformed by a regular expression
\r
1084 UPPER() - returns a text value converted to upper-case
\r
1088 <simpara>For a complete list of scalar functions, see
\r
1089 <ulink url="http://www.postgresql.org/docs/8.3/interactive/functions.html">the PostgreSQL function documentation</ulink>.</simpara>
\r
1091 <simplesect id="_aggregate_functions">
\r
1092 <title>Aggregate functions</title>
\r
1093 <simpara>Aggregate functions return a single value computed from the the complete set of
\r
1094 values returned for the specified column.</simpara>
\r
1124 <simplesect id="_sub_selects">
\r
1125 <title>Sub-selects</title>
\r
1126 <simpara>A sub-select is the technique of using the results of one query to feed
\r
1127 into another query. You can, for example, return a set of values from
\r
1128 one column in a SELECT statement to be used to satisfy the IN() condition
\r
1129 of another SELECT statement; or you could return the MAX() value of a
\r
1130 column in a SELECT statement to match the = condition of another SELECT
\r
1131 statement.</simpara>
\r
1132 <simpara>For example, in the following query we use a sub-select to restrict the copies
\r
1133 returned by the main SELECT statement to only those locations that have an
\r
1134 <literal>opac_visible</literal> value of <literal>TRUE</literal>:</simpara>
\r
1135 <formalpara><title>Sub-select example</title><para>
\r
1136 <programlisting language="sql" linenumbering="unnumbered">SELECT call_number
\r
1138 WHERE deleted IS FALSE
\r
1141 FROM asset.copy_location
\r
1142 WHERE opac_visible IS TRUE
\r
1144 ;</programlisting>
\r
1145 </para></formalpara>
\r
1146 <simpara>Sub-selects can be an approachable way to breaking down a problem that
\r
1147 requires matching values between different tables, and often result in
\r
1148 a clearly expressed solution to a problem. However, if you start writing
\r
1149 sub-selects within sub-selects, you should consider tackling the problem
\r
1150 with joins instead.</simpara>
\r
1152 <simplesect id="_joins">
\r
1153 <title>Joins</title>
\r
1154 <simpara>Joins enable you to access the values from multiple tables in your query
\r
1155 results and comparison operators. For example, joins are what enable you to
\r
1156 relate a bibliographic record to a barcoded copy via the <literal>biblio.record_entry</literal>,
\r
1157 <literal>asset.call_number</literal>, and <literal>asset.copy</literal> tables. In this section, we discuss the
\r
1158 most common kind of join—the inner join—as well as the less common outer join
\r
1159 and some set operations which can compare and contrast the values returned by
\r
1160 separate queries.</simpara>
\r
1161 <simpara>When we talk about joins, we are going to talk about the left-hand table and
\r
1162 the right-hand table that participate in the join. Every join brings together
\r
1163 just two tables - but you can use an unlimited (for our purposes) number
\r
1164 of joins in a single SQL statement. Each time you use a join, you effectively
\r
1165 create a new table, so when you add a second join clause to a statement,
\r
1166 table 1 and table 2 (which were the left-hand table and the right-hand table
\r
1167 for the first join) now act as a merged left-hand table and the new table
\r
1168 in the second join clause is the right-hand table.</simpara>
\r
1169 <simpara>Clear as mud? Okay, let’s look at some examples.</simpara>
\r
1170 <simplesect id="_inner_joins">
\r
1171 <title>Inner joins</title>
\r
1172 <simpara>An inner join returns all of the columns from the left-hand table in the join
\r
1173 with all of the columns from the right-hand table in the joins that match a
\r
1174 condition in the ON clause. Typically, you use the <literal>=</literal> operator to match the
\r
1175 foreign key of the left-hand table with the primary key of the right-hand
\r
1176 table to follow the natural relationship between the tables.</simpara>
\r
1177 <simpara>In the following example, we return all of columns from the <literal>actor.usr</literal> and
\r
1178 <literal>actor.org_unit</literal> tables, joined on the relationship between the user’s home
\r
1179 library and the library’s ID. Notice in the results that some columns, like
\r
1180 <literal>id</literal> and <literal>mailing_address</literal>, appear twice; this is because both the <literal>actor.usr</literal>
\r
1181 and <literal>actor.org_unit</literal> tables include columns with these names. This is also why
\r
1182 we have to fully qualify the column names in our queries with the schema and
\r
1183 table names.</simpara>
\r
1184 <formalpara><title>A simple inner join</title><para>
\r
1185 <programlisting language="sql" linenumbering="unnumbered">SELECT *
\r
1187 INNER JOIN actor.org_unit ON actor.usr.home_ou = actor.org_unit.id
\r
1188 WHERE actor.org_unit.shortname = 'CONS'
\r
1191 -[ RECORD 1 ]------------------+---------------------------------
\r
1202 claims_never_checked_out_count | 0
\r
1208 mailing_address | 1
\r
1209 billing_address | 1
\r
1211 name | Example Consortium
\r
1215 fiscal_calendar | 1</programlisting>
\r
1216 </para></formalpara>
\r
1217 <simpara>Of course, you do not have to return every column from the joined tables;
\r
1218 you can (and should) continue to specify only the columns that you want to
\r
1219 return. In the following example, we count the number of borrowers for
\r
1220 every user profile in a given library by joining the <literal>permission.grp_tree</literal>
\r
1221 table where profiles are defined against the <literal>actor.usr</literal> table, and then
\r
1222 joining the <literal>actor.org_unit</literal> table to give us access to the user’s home
\r
1223 library:</simpara>
\r
1224 <formalpara><title>Borrower Count by Profile (Adult, Child, etc)/Library</title><para>
\r
1225 <programlisting language="sql" linenumbering="unnumbered">SELECT permission.grp_tree.name, actor.org_unit.name, COUNT(permission.grp_tree.name)
\r
1227 INNER JOIN permission.grp_tree
\r
1228 ON actor.usr.profile = permission.grp_tree.id
\r
1229 INNER JOIN actor.org_unit
\r
1230 ON actor.org_unit.id = actor.usr.home_ou
\r
1231 WHERE actor.usr.deleted IS FALSE
\r
1232 GROUP BY permission.grp_tree.name, actor.org_unit.name
\r
1233 ORDER BY actor.org_unit.name, permission.grp_tree.name
\r
1236 name | name | count
\r
1237 -------+--------------------+-------
\r
1238 Users | Example Consortium | 1
\r
1239 (1 row)</programlisting>
\r
1240 </para></formalpara>
\r
1242 <simplesect id="_aliases">
\r
1243 <title>Aliases</title>
\r
1244 <simpara>So far we have been fully-qualifying all of our table names and column names to
\r
1245 prevent any confusion. This quickly gets tiring with lengthy qualified
\r
1246 table names like <literal>permission.grp_tree</literal>, so the SQL syntax enables us to assign
\r
1247 aliases to table names and column names. When you define an alias for a table
\r
1248 name, you can access its column throughout the rest of the statement by simply
\r
1249 appending the column name to the alias with a period; for example, if you assign
\r
1250 the alias <literal>au</literal> to the <literal>actor.usr</literal> table, you can access the <literal>actor.usr.id</literal>
\r
1251 column through the alias as <literal>au.id</literal>.</simpara>
\r
1252 <simpara>The formal syntax for declaring an alias for a column is to follow the column
\r
1253 name in the result columns clause with <literal>AS</literal> <emphasis>alias</emphasis>. To declare an alias for a table name,
\r
1254 follow the table name in the FROM clause (including any JOIN statements) with
\r
1255 <literal>AS</literal> <emphasis>alias</emphasis>. However, the <literal>AS</literal> keyword is optional for tables (and columns as
\r
1256 of PostgreSQL 8.4), and in practice most SQL statements leave it out. For
\r
1257 example, we can write the previous INNER JOIN statement example using aliases
\r
1258 instead of fully-qualified identifiers:</simpara>
\r
1259 <formalpara><title>Borrower Count by Profile (using aliases)</title><para>
\r
1260 <programlisting language="sql" linenumbering="unnumbered">SELECT pgt.name AS "Profile", aou.name AS "Library", COUNT(pgt.name) AS "Count"
\r
1262 INNER JOIN permission.grp_tree pgt
\r
1263 ON au.profile = pgt.id
\r
1264 INNER JOIN actor.org_unit aou
\r
1265 ON aou.id = au.home_ou
\r
1266 WHERE au.deleted IS FALSE
\r
1267 GROUP BY pgt.name, aou.name
\r
1268 ORDER BY aou.name, pgt.name
\r
1271 Profile | Library | Count
\r
1272 ---------+--------------------+-------
\r
1273 Users | Example Consortium | 1
\r
1274 (1 row)</programlisting>
\r
1275 </para></formalpara>
\r
1276 <simpara>A nice side effect of declaring an alias for your columns is that the alias
\r
1277 is used as the column header in the results table. The previous version of
\r
1278 the query, which didn’t use aliased column names, had two columns named
\r
1279 <literal>name</literal>; this version of the query with aliases results in a clearer
\r
1280 categorization.</simpara>
\r
1282 <simplesect id="_outer_joins">
\r
1283 <title>Outer joins</title>
\r
1284 <simpara>An outer join returns all of the rows from one or both of the tables
\r
1285 participating in the join.</simpara>
\r
1289 For a LEFT OUTER JOIN, the join returns all of the rows from the left-hand
\r
1290 table and the rows matching the join condition from the right-hand table, with
\r
1291 NULL values for the rows with no match in the right-hand table.
\r
1296 A RIGHT OUTER JOIN behaves in the same way as a LEFT OUTER JOIN, with the
\r
1297 exception that all rows are returned from the right-hand table participating in
\r
1303 For a FULL OUTER JOIN, the join returns all the rows from both the left-hand
\r
1304 and right-hand tables, with NULL values for the rows with no match in either
\r
1305 the left-hand or right-hand table.
\r
1309 <formalpara><title>Base tables for the OUTER JOIN examples</title><para>
\r
1310 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM aaa;
\r
1321 SELECT * FROM bbb;
\r
1324 ----+-------+----------
\r
1327 5 | five | fivefive
\r
1329 (4 rows)</programlisting>
\r
1330 </para></formalpara>
\r
1331 <formalpara><title>Example of a LEFT OUTER JOIN</title><para>
\r
1332 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM aaa
\r
1333 LEFT OUTER JOIN bbb ON aaa.id = bbb.id
\r
1335 id | stuff | id | stuff | foo
\r
1336 ----+-------+----+-------+----------
\r
1337 1 | one | 1 | one | oneone
\r
1338 2 | two | 2 | two | twotwo
\r
1341 5 | five | 5 | five | fivefive
\r
1342 (5 rows)</programlisting>
\r
1343 </para></formalpara>
\r
1344 <formalpara><title>Example of a RIGHT OUTER JOIN</title><para>
\r
1345 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM aaa
\r
1346 RIGHT OUTER JOIN bbb ON aaa.id = bbb.id
\r
1348 id | stuff | id | stuff | foo
\r
1349 ----+-------+----+-------+----------
\r
1350 1 | one | 1 | one | oneone
\r
1351 2 | two | 2 | two | twotwo
\r
1352 5 | five | 5 | five | fivefive
\r
1353 | | 6 | six | sixsix
\r
1354 (4 rows)</programlisting>
\r
1355 </para></formalpara>
\r
1356 <formalpara><title>Example of a FULL OUTER JOIN</title><para>
\r
1357 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM aaa
\r
1358 FULL OUTER JOIN bbb ON aaa.id = bbb.id
\r
1360 id | stuff | id | stuff | foo
\r
1361 ----+-------+----+-------+----------
\r
1362 1 | one | 1 | one | oneone
\r
1363 2 | two | 2 | two | twotwo
\r
1366 5 | five | 5 | five | fivefive
\r
1367 | | 6 | six | sixsix
\r
1368 (6 rows)</programlisting>
\r
1369 </para></formalpara>
\r
1371 <simplesect id="_self_joins">
\r
1372 <title>Self joins</title>
\r
1373 <simpara>It is possible to join a table to itself. You can, in fact you must, use
\r
1374 aliases to disambiguate the references to the table.</simpara>
\r
1377 <simplesect id="_set_operations">
\r
1378 <title>Set operations</title>
\r
1379 <simpara>Relational databases are effectively just an efficient mechanism for
\r
1380 manipulating sets of values; they are implementations of set theory. There are
\r
1381 three operators for sets (tables) in which each set must have the same number
\r
1382 of columns with compatible data types: the union, intersection, and difference
\r
1383 operators.</simpara>
\r
1384 <formalpara><title>Base tables for the set operation examples</title><para>
\r
1385 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM aaa;
\r
1396 SELECT * FROM bbb;
\r
1399 ----+-------+----------
\r
1402 5 | five | fivefive
\r
1404 (4 rows)</programlisting>
\r
1405 </para></formalpara>
\r
1406 <simplesect id="_union">
\r
1407 <title>Union</title>
\r
1408 <simpara>The <literal>UNION</literal> operator returns the distinct set of rows that are members of
\r
1409 either or both of the left-hand and right-hand tables. The <literal>UNION</literal> operator
\r
1410 does not return any duplicate rows. To return duplicate rows, use the
\r
1411 <literal>UNION ALL</literal> operator.</simpara>
\r
1412 <formalpara><title>Example of a UNION set operation</title><para>
\r
1413 <programlisting language="sql" linenumbering="unnumbered">-- The parentheses are not required, but are intended to help
\r
1414 -- illustrate the sets participating in the set operation
\r
1435 (6 rows)</programlisting>
\r
1436 </para></formalpara>
\r
1438 <simplesect id="_intersection">
\r
1439 <title>Intersection</title>
\r
1440 <simpara>The <literal>INTERSECT</literal> operator returns the distinct set of rows that are common to
\r
1441 both the left-hand and right-hand tables. To return duplicate rows, use the
\r
1442 <literal>INTERSECT ALL</literal> operator.</simpara>
\r
1443 <formalpara><title>Example of an INTERSECT set operation</title><para>
\r
1444 <programlisting language="sql" linenumbering="unnumbered">(
\r
1461 (3 rows)</programlisting>
\r
1462 </para></formalpara>
\r
1464 <simplesect id="_difference">
\r
1465 <title>Difference</title>
\r
1466 <simpara>The <literal>EXCEPT</literal> operator returns the rows in the left-hand table that do not
\r
1467 exist in the right-hand table. You are effectively subtracting the common
\r
1468 rows from the left-hand table.</simpara>
\r
1469 <formalpara><title>Example of an EXCEPT set operation</title><para>
\r
1470 <programlisting language="sql" linenumbering="unnumbered">(
\r
1488 -- Order matters: switch the left-hand and right-hand tables
\r
1489 -- and you get a different result
\r
1505 (1 row)</programlisting>
\r
1506 </para></formalpara>
\r
1509 <simplesect id="_views">
\r
1510 <title>Views</title>
\r
1511 <simpara>A view is a persistent <literal>SELECT</literal> statement that acts like a read-only table.
\r
1512 To create a view, issue the <literal>CREATE VIEW</literal> statement, giving the view a name
\r
1513 and a <literal>SELECT</literal> statement on which the view is built.</simpara>
\r
1514 <simpara>The following example creates a view based on our borrower profile count:</simpara>
\r
1515 <formalpara><title>Creating a view</title><para>
\r
1516 <programlisting language="sql" linenumbering="unnumbered">CREATE VIEW actor.borrower_profile_count AS
\r
1517 SELECT pgt.name AS "Profile", aou.name AS "Library", COUNT(pgt.name) AS "Count"
\r
1519 INNER JOIN permission.grp_tree pgt
\r
1520 ON au.profile = pgt.id
\r
1521 INNER JOIN actor.org_unit aou
\r
1522 ON aou.id = au.home_ou
\r
1523 WHERE au.deleted IS FALSE
\r
1524 GROUP BY pgt.name, aou.name
\r
1525 ORDER BY aou.name, pgt.name
\r
1526 ;</programlisting>
\r
1527 </para></formalpara>
\r
1528 <simpara>When you subsequently select results from the view, you can apply additional
\r
1529 <literal>WHERE</literal> clauses to filter the results, or <literal>ORDER BY</literal> clauses to change the
\r
1530 order of the returned rows. In the following examples, we issue a simple
\r
1531 <literal>SELECT *</literal> statement to show that the default results are returned in the
\r
1532 same order from the view as the equivalent SELECT statement would be returned.
\r
1533 Then we issue a <literal>SELECT</literal> statement with a <literal>WHERE</literal> clause to further filter the
\r
1534 results.</simpara>
\r
1535 <formalpara><title>Selecting results from a view</title><para>
\r
1536 <programlisting language="sql" linenumbering="unnumbered">SELECT * FROM actor.borrower_profile_count;
\r
1538 Profile | Library | Count
\r
1539 ----------------------------+----------------------------+-------
\r
1540 Faculty | University Library | 208
\r
1541 Graduate | University Library | 16
\r
1542 Patrons | University Library | 62
\r
1545 -- You can still filter your results with WHERE clauses
\r
1547 FROM actor.borrower_profile_count
\r
1548 WHERE "Profile" = 'Faculty';
\r
1550 Profile | Library | Count
\r
1551 ---------+----------------------------+-------
\r
1552 Faculty | University Library | 208
\r
1553 Faculty | College Library | 64
\r
1554 Faculty | College Library 2 | 102
\r
1555 Faculty | University Library 2 | 776
\r
1556 (4 rows)</programlisting>
\r
1557 </para></formalpara>
\r
1559 <simplesect id="_inheritance">
\r
1560 <title>Inheritance</title>
\r
1561 <simpara>PostgreSQL supports table inheritance: that is, a child table inherits its
\r
1562 base definition from a parent table, but can add additional columns to its
\r
1563 own definition. The data from any child tables is visible in queries against
\r
1564 the parent table.</simpara>
\r
1565 <simpara>Evergreen uses table inheritance in several areas:
\r
1566 * In the Vandelay MARC batch importer / exporter, Evergreen defines base
\r
1567 tables for generic queues and queued records for which authority record and
\r
1568 bibliographic record child tables
\r
1569 * Billable transactions are based on the <literal>money.billable_xact</literal> table;
\r
1570 child tables include <literal>action.circulation</literal> for circulation transactions
\r
1571 and <literal>money.grocery</literal> for general bills.
\r
1572 * Payments are based on the <literal>money.payment</literal> table; its child table is
\r
1573 <literal>money.bnm_payment</literal> (for brick-and-mortar payments), which in turn has child
\r
1574 tables of <literal>money.forgive_payment</literal>, <literal>money.work_payment</literal>, <literal>money.credit_payment</literal>,
\r
1575 <literal>money.goods_payment</literal>, and <literal>money.bnm_desk_payment</literal>. The
\r
1576 <literal>money.bnm_desk_payment</literal> table in turn has child tables of <literal>money.cash_payment</literal>,
\r
1577 <literal>money.check_payment</literal>, and <literal>money.credit_card_payment</literal>.
\r
1578 * Transits are based on the <literal>action.transit_copy</literal> table, which has a child
\r
1579 table of <literal>action.hold_transit_copy</literal> for transits initiated by holds.
\r
1580 * Generic acquisition line items are defined by the
\r
1581 <literal>acq.lineitem_attr_definition</literal> table, which in turn has a number of child
\r
1582 tables to define MARC attributes, generated attributes, user attributes, and
\r
1583 provider attributes.</simpara>
\r
1586 <section id="understanding_query_performance_with_explain">
\r
1587 <title>Understanding query performance with EXPLAIN</title>
\r
1588 <simpara>Some queries run for a long, long time. This can be the result of a poorly
\r
1589 written query—a query with a join condition that joins every
\r
1590 row in the <literal>biblio.record_entry</literal> table with every row in the <literal>metabib.full_rec</literal>
\r
1591 view would consume a massive amount of memory and disk space and CPU time—or
\r
1592 a symptom of a schema that needs some additional indexes. PostgreSQL provides
\r
1593 the <literal>EXPLAIN</literal> tool to estimate how long it will take to run a given query and
\r
1594 show you the <emphasis>query plan</emphasis> (how it plans to retrieve the results from the
\r
1595 database).</simpara>
\r
1596 <simpara>To generate the query plan without actually running the statement, simply
\r
1597 prepend the <literal>EXPLAIN</literal> keyword to your query. In the following example, we
\r
1598 generate the query plan for the poorly written query that would join every
\r
1599 row in the <literal>biblio.record_entry</literal> table with every row in the <literal>metabib.full_rec</literal>
\r
1601 <formalpara><title>Query plan for a terrible query</title><para>
\r
1602 <programlisting language="sql" linenumbering="unnumbered">EXPLAIN SELECT *
\r
1603 FROM biblio.record_entry
\r
1604 FULL OUTER JOIN metabib.full_rec ON 1=1
\r
1608 -------------------------------------------------------------------------------//
\r
1609 Merge Full Join (cost=0.00..4959156437783.60 rows=132415734100864 width=1379)
\r
1610 -> Seq Scan on record_entry (cost=0.00..400634.16 rows=2013416 width=1292)
\r
1611 -> Seq Scan on real_full_rec (cost=0.00..1640972.04 rows=65766704 width=87)
\r
1612 (3 rows)</programlisting>
\r
1613 </para></formalpara>
\r
1614 <simpara>This query plan shows that the query would return 132415734100864 rows, and it
\r
1615 plans to accomplish what you asked for by sequentially scanning (<emphasis>Seq Scan</emphasis>)
\r
1616 every row in each of the tables participating in the join.</simpara>
\r
1617 <simpara>In the following example, we have realized our mistake in joining every row of
\r
1618 the left-hand table with every row in the right-hand table and take the saner
\r
1619 approach of using an <literal>INNER JOIN</literal> where the join condition is on the record ID.</simpara>
\r
1620 <formalpara><title>Query plan for a less terrible query</title><para>
\r
1621 <programlisting language="sql" linenumbering="unnumbered">EXPLAIN SELECT *
\r
1622 FROM biblio.record_entry bre
\r
1623 INNER JOIN metabib.full_rec mfr ON mfr.record = bre.id;
\r
1625 ----------------------------------------------------------------------------------------//
\r
1626 Hash Join (cost=750229.86..5829273.98 rows=65766704 width=1379)
\r
1627 Hash Cond: (real_full_rec.record = bre.id)
\r
1628 -> Seq Scan on real_full_rec (cost=0.00..1640972.04 rows=65766704 width=87)
\r
1629 -> Hash (cost=400634.16..400634.16 rows=2013416 width=1292)
\r
1630 -> Seq Scan on record_entry bre (cost=0.00..400634.16 rows=2013416 width=1292)
\r
1631 (5 rows)</programlisting>
\r
1632 </para></formalpara>
\r
1633 <simpara>This time, we will return 65766704 rows - still way too many rows. We forgot
\r
1634 to include a <literal>WHERE</literal> clause to limit the results to something meaningful. In
\r
1635 the following example, we will limit the results to deleted records that were
\r
1636 modified in the last month.</simpara>
\r
1637 <formalpara><title>Query plan for a realistic query</title><para>
\r
1638 <programlisting language="sql" linenumbering="unnumbered">EXPLAIN SELECT *
\r
1639 FROM biblio.record_entry bre
\r
1640 INNER JOIN metabib.full_rec mfr ON mfr.record = bre.id
\r
1641 WHERE bre.deleted IS TRUE
\r
1642 AND DATE_TRUNC('MONTH', bre.edit_date) >
\r
1643 DATE_TRUNC ('MONTH', NOW() - '1 MONTH'::INTERVAL)
\r
1647 ----------------------------------------------------------------------------------------//
\r
1648 Hash Join (cost=5058.86..2306218.81 rows=201669 width=1379)
\r
1649 Hash Cond: (real_full_rec.record = bre.id)
\r
1650 -> Seq Scan on real_full_rec (cost=0.00..1640972.04 rows=65766704 width=87)
\r
1651 -> Hash (cost=4981.69..4981.69 rows=6174 width=1292)
\r
1652 -> Index Scan using biblio_record_entry_deleted on record_entry bre
\r
1653 (cost=0.00..4981.69 rows=6174 width=1292)
\r
1654 Index Cond: (deleted = true)
\r
1655 Filter: ((deleted IS TRUE) AND (date_trunc('MONTH'::text, edit_date)
\r
1656 > date_trunc('MONTH'::text, (now() - '1 mon'::interval))))
\r
1657 (7 rows)</programlisting>
\r
1658 </para></formalpara>
\r
1659 <simpara>We can see that the number of rows returned is now only 201669; that’s
\r
1660 something we can work with. Also, the overall cost of the query is 2306218,
\r
1661 compared to 4959156437783 in the original query. The <literal>Index Scan</literal> tells us
\r
1662 that the query planner will use the index that was defined on the <literal>deleted</literal>
\r
1663 column to avoid having to check every row in the <literal>biblio.record_entry</literal> table.</simpara>
\r
1664 <simpara>However, we are still running a sequential scan over the
\r
1665 <literal>metabib.real_full_rec</literal> table (the table on which the <literal>metabib.full_rec</literal>
\r
1666 view is based). Given that linking from the bibliographic records to the
\r
1667 flattened MARC subfields is a fairly common operation, we could create a
\r
1668 new index and see if that speeds up our query plan.</simpara>
\r
1669 <formalpara><title>Query plan with optimized access via a new index</title><para>
\r
1670 <programlisting language="sql" linenumbering="unnumbered">-- This index will take a long time to create on a large database
\r
1671 -- of bibliographic records
\r
1672 CREATE INDEX bib_record_idx ON metabib.real_full_rec (record);
\r
1675 FROM biblio.record_entry bre
\r
1676 INNER JOIN metabib.full_rec mfr ON mfr.record = bre.id
\r
1677 WHERE bre.deleted IS TRUE
\r
1678 AND DATE_TRUNC('MONTH', bre.edit_date) >
\r
1679 DATE_TRUNC ('MONTH', NOW() - '1 MONTH'::INTERVAL)
\r
1683 ----------------------------------------------------------------------------------------//
\r
1684 Nested Loop (cost=0.00..1558330.46 rows=201669 width=1379)
\r
1685 -> Index Scan using biblio_record_entry_deleted on record_entry bre
\r
1686 (cost=0.00..4981.69 rows=6174 width=1292)
\r
1687 Index Cond: (deleted = true)
\r
1688 Filter: ((deleted IS TRUE) AND (date_trunc('MONTH'::text, edit_date) >
\r
1689 date_trunc('MONTH'::text, (now() - '1 mon'::interval))))
\r
1690 -> Index Scan using bib_record_idx on real_full_rec
\r
1691 (cost=0.00..240.89 rows=850 width=87)
\r
1692 Index Cond: (real_full_rec.record = bre.id)
\r
1693 (6 rows)</programlisting>
\r
1694 </para></formalpara>
\r
1695 <simpara>We can see that the resulting number of rows is still the same (201669), but
\r
1696 the execution estimate has dropped to 1558330 because the query planner can
\r
1697 use the new index (<literal>bib_record_idx</literal>) rather than scanning the entire table.
\r
1698 Success!</simpara>
\r
1699 <note><simpara>While indexes can significantly speed up read access to tables for common
\r
1700 filtering conditions, every time a row is created or updated the corresponding
\r
1701 indexes also need to be maintained - which can decrease the performance of
\r
1702 writes to the database. Be careful to keep the balance of read performance
\r
1703 versus write performance in mind if you plan to create custom indexes in your
\r
1704 Evergreen database.</simpara></note>
\r
1706 <section id="inserting_updating_and_deleting_data">
\r
1707 <title>Inserting, updating, and deleting data</title>
\r
1708 <simplesect id="_inserting_data">
\r
1709 <title>Inserting data</title>
\r
1710 <simpara>To insert one or more rows into a table, use the INSERT statement to identify
\r
1711 the target table and list the columns in the table for which you are going to
\r
1712 provide values for each row. If you do not list one or more columns contained
\r
1713 in the table, the database will automatically supply a <literal>NULL</literal> value for those
\r
1714 columns. The values for each row follow the <literal>VALUES</literal> clause and are grouped in
\r
1715 parentheses and delimited by commas. Each row, in turn, is delimited by commas
\r
1716 (<emphasis>this multiple row syntax requires PostgreSQL 8.2 or higher</emphasis>).</simpara>
\r
1717 <simpara>For example, to insert two rows into the <literal>permission.usr_grp_map</literal> table:</simpara>
\r
1718 <formalpara><title>Inserting rows into the <literal>permission.usr_grp_map</literal> table</title><para>
\r
1719 <programlisting language="sql" linenumbering="unnumbered">INSERT INTO permission.usr_grp_map (usr, grp)
\r
1720 VALUES (2, 10), (2, 4)
\r
1721 ;</programlisting>
\r
1722 </para></formalpara>
\r
1723 <simpara>Of course, as with the rest of SQL, you can replace individual column values
\r
1724 with one or more use sub-selects:</simpara>
\r
1725 <formalpara><title>Inserting rows using sub-selects instead of integers</title><para>
\r
1726 <programlisting language="sql" linenumbering="unnumbered">INSERT INTO permission.usr_grp_map (usr, grp)
\r
1728 (SELECT id FROM actor.usr
\r
1729 WHERE family_name = 'Scott' AND first_given_name = 'Daniel'),
\r
1730 (SELECT id FROM permission.grp_tree
\r
1731 WHERE name = 'Local System Administrator')
\r
1733 (SELECT id FROM actor.usr
\r
1734 WHERE family_name = 'Scott' AND first_given_name = 'Daniel'),
\r
1735 (SELECT id FROM permission.grp_tree
\r
1736 WHERE name = 'Circulator')
\r
1738 ;</programlisting>
\r
1739 </para></formalpara>
\r
1741 <simplesect id="_inserting_data_using_a_select_statement">
\r
1742 <title>Inserting data using a SELECT statement</title>
\r
1743 <simpara>Sometimes you want to insert a bulk set of data into a new table based on
\r
1744 a query result. Rather than a <literal>VALUES</literal> clause, you can use a <literal>SELECT</literal>
\r
1745 statement to insert one or more rows matching the column definitions. This
\r
1746 is a good time to point out that you can include explicit values, instead
\r
1747 of just column identifiers, in the return columns of the <literal>SELECT</literal> statement.
\r
1748 The explicit values are returned in every row of the result set.</simpara>
\r
1749 <simpara>In the following example, we insert 6 rows into the <literal>permission.usr_grp_map</literal>
\r
1750 table; each row will have a <literal>usr</literal> column value of 1, with varying values for
\r
1751 the <literal>grp</literal> column value based on the <literal>id</literal> column values returned from
\r
1752 <literal>permission.grp_tree</literal>:</simpara>
\r
1753 <formalpara><title>Inserting rows via a <literal>SELECT</literal> statement</title><para>
\r
1754 <programlisting language="sql" linenumbering="unnumbered">INSERT INTO permission.usr_grp_map (usr, grp)
\r
1756 FROM permission.grp_tree
\r
1760 INSERT 0 6</programlisting>
\r
1761 </para></formalpara>
\r
1763 <simplesect id="_deleting_rows">
\r
1764 <title>Deleting rows</title>
\r
1765 <simpara>Deleting data from a table is normally fairly easy. To delete rows from a table,
\r
1766 issue a <literal>DELETE</literal> statement identifying the table from which you want to delete
\r
1767 rows and a <literal>WHERE</literal> clause identifying the row or rows that should be deleted.</simpara>
\r
1768 <simpara>In the following example, we delete all of the rows from the
\r
1769 <literal>permission.grp_perm_map</literal> table where the permission maps to
\r
1770 <literal>UPDATE_ORG_UNIT_CLOSING</literal> and the group is anything other than administrators:</simpara>
\r
1771 <formalpara><title>Deleting rows from a table</title><para>
\r
1772 <programlisting language="sql" linenumbering="unnumbered">DELETE FROM permission.grp_perm_map
\r
1775 FROM permission.grp_tree
\r
1776 WHERE name != 'Local System Administrator'
\r
1779 FROM permission.perm_list
\r
1780 WHERE code = 'UPDATE_ORG_UNIT_CLOSING'
\r
1782 ;</programlisting>
\r
1783 </para></formalpara>
\r
1784 <note><simpara>There are two main reasons that a <literal>DELETE</literal> statement may not actually
\r
1785 delete rows from a table, even when the rows meet the conditional clause.</simpara></note>
\r
1786 <orderedlist numeration="arabic">
\r
1790 If the row contains a value that is the target of a relational constraint,
\r
1791 for example, if another table has a foreign key pointing at your target
\r
1792 table, you will be prevented from deleting a row with a value corresponding
\r
1793 to a row in the dependent table.
\r
1798 If the table has a rule that substitutes a different action for a <literal>DELETE</literal>
\r
1799 statement, the deletion will not take place. In Evergreen it is common for a
\r
1800 table to have a rule that substitutes the action of setting a <literal>deleted</literal> column
\r
1801 to <literal>TRUE</literal>. For example, if a book is discarded, deleting the row representing
\r
1802 the copy from the <literal>asset.copy</literal> table would severely affect circulation statistics,
\r
1803 bills, borrowing histories, and their corresponding tables in the database that
\r
1804 have foreign keys pointing at the <literal>asset.copy</literal> table (<literal>action.circulation</literal> and
\r
1805 <literal>money.billing</literal> and its children respectively). Instead, the <literal>deleted</literal> column
\r
1806 value is set to <literal>TRUE</literal> and Evergreen’s application logic skips over these rows
\r
1812 <simplesect id="_updating_rows">
\r
1813 <title>Updating rows</title>
\r
1814 <simpara>To update rows in a table, issue an <literal>UPDATE</literal> statement identifying the table
\r
1815 you want to update, the column or columns that you want to set with their
\r
1816 respective new values, and (optionally) a <literal>WHERE</literal> clause identifying the row or
\r
1817 rows that should be updated.</simpara>
\r
1818 <simpara>Following is the syntax for the <literal>UPDATE</literal> statement:</simpara>
\r
1820 <literallayout><literal>UPDATE</literal> [<emphasis>table-name</emphasis>]
\r
1821 <literal>SET</literal> [<emphasis>column</emphasis>] <literal>TO</literal> [<emphasis>new-value</emphasis>]
\r
1822 <literal>WHERE</literal> [<emphasis>condition</emphasis>]
\r
1827 <section id="query_requests">
\r
1828 <title>Query requests</title>
\r
1829 <simpara>The following queries were requested by Bibliomation, but might be reusable
\r
1830 by other libraries.</simpara>
\r
1831 <simplesect id="_monthly_circulation_stats_by_collection_code_library">
\r
1832 <title>Monthly circulation stats by collection code / library</title>
\r
1833 <formalpara><title>Monthly Circulation Stats by Collection Code/Library</title><para>
\r
1834 <programlisting language="sql" linenumbering="unnumbered">SELECT COUNT(acirc.id) AS "COUNT", aou.name AS "Library", acl.name AS "Copy Location"
\r
1835 FROM asset.copy ac
\r
1836 INNER JOIN asset.copy_location acl ON ac.location = acl.id
\r
1837 INNER JOIN action.circulation acirc ON acirc.target_copy = ac.id
\r
1838 INNER JOIN actor.org_unit aou ON acirc.circ_lib = aou.id
\r
1839 WHERE DATE_TRUNC('MONTH', acirc.create_time) = DATE_TRUNC('MONTH', NOW() - INTERVAL '3 month')
\r
1840 AND acirc.desk_renewal IS FALSE
\r
1841 AND acirc.opac_renewal IS FALSE
\r
1842 AND acirc.phone_renewal IS FALSE
\r
1843 GROUP BY aou.name, acl.name
\r
1844 ORDER BY aou.name, acl.name, 1
\r
1845 ;</programlisting>
\r
1846 </para></formalpara>
\r
1848 <simplesect id="_monthly_circulation_stats_by_borrower_stat_library">
\r
1849 <title>Monthly circulation stats by borrower stat / library</title>
\r
1850 <formalpara><title>Monthly Circulation Stats by Borrower Stat/Library</title><para>
\r
1851 <programlisting language="sql" linenumbering="unnumbered">SELECT COUNT(acirc.id) AS "COUNT", aou.name AS "Library", asceum.stat_cat_entry AS "Borrower Stat"
\r
1852 FROM action.circulation acirc
\r
1853 INNER JOIN actor.org_unit aou ON acirc.circ_lib = aou.id
\r
1854 INNER JOIN actor.stat_cat_entry_usr_map asceum ON asceum.target_usr = acirc.usr
\r
1855 INNER JOIN actor.stat_cat astat ON asceum.stat_cat = astat.id
\r
1856 WHERE DATE_TRUNC('MONTH', acirc.create_time) = DATE_TRUNC('MONTH', NOW() - INTERVAL '3 month')
\r
1857 AND astat.name = 'Preferred language'
\r
1858 AND acirc.desk_renewal IS FALSE
\r
1859 AND acirc.opac_renewal IS FALSE
\r
1860 AND acirc.phone_renewal IS FALSE
\r
1861 GROUP BY aou.name, asceum.stat_cat_entry
\r
1862 ORDER BY aou.name, asceum.stat_cat_entry, 1
\r
1863 ;</programlisting>
\r
1864 </para></formalpara>
\r
1866 <simplesect id="_monthly_intralibrary_loan_stats_by_library">
\r
1867 <title>Monthly intralibrary loan stats by library</title>
\r
1868 <formalpara><title>Monthly Intralibrary Loan Stats by Library</title><para>
\r
1869 <programlisting language="sql" linenumbering="unnumbered">SELECT aou.name AS "Library", COUNT(acirc.id)
\r
1870 FROM action.circulation acirc
\r
1871 INNER JOIN actor.org_unit aou ON acirc.circ_lib = aou.id
\r
1872 INNER JOIN asset.copy ac ON acirc.target_copy = ac.id
\r
1873 INNER JOIN asset.call_number acn ON ac.call_number = acn.id
\r
1874 WHERE acirc.circ_lib != acn.owning_lib
\r
1875 AND DATE_TRUNC('MONTH', acirc.create_time) = DATE_TRUNC('MONTH', NOW() - INTERVAL '3 month')
\r
1876 AND acirc.desk_renewal IS FALSE
\r
1877 AND acirc.opac_renewal IS FALSE
\r
1878 AND acirc.phone_renewal IS FALSE
\r
1880 ORDER BY aou.name, 2
\r
1881 ;</programlisting>
\r
1882 </para></formalpara>
\r
1884 <simplesect id="_monthly_borrowers_added_by_profile_adult_child_etc_library">
\r
1885 <title>Monthly borrowers added by profile (adult, child, etc) / library</title>
\r
1886 <formalpara><title>Monthly Borrowers Added by Profile (Adult, Child, etc)/Library</title><para>
\r
1887 <programlisting language="sql" linenumbering="unnumbered">SELECT pgt.name AS "Profile", aou.name AS "Library", COUNT(pgt.name) AS "Count"
\r
1889 INNER JOIN permission.grp_tree pgt
\r
1890 ON au.profile = pgt.id
\r
1891 INNER JOIN actor.org_unit aou
\r
1892 ON aou.id = au.home_ou
\r
1893 WHERE au.deleted IS FALSE
\r
1894 AND DATE_TRUNC('MONTH', au.create_date) = DATE_TRUNC('MONTH', NOW() - '3 months'::interval)
\r
1895 GROUP BY pgt.name, aou.name
\r
1896 ORDER BY aou.name, pgt.name
\r
1897 ;</programlisting>
\r
1898 </para></formalpara>
\r
1900 <simplesect id="_borrower_count_by_profile_adult_child_etc_library">
\r
1901 <title>Borrower count by profile (adult, child, etc) / library</title>
\r
1902 <formalpara><title>Borrower Count by Profile (Adult, Child, etc)/Library</title><para>
\r
1903 <programlisting language="sql" linenumbering="unnumbered">SELECT pgt.name AS "Profile", aou.name AS "Library", COUNT(pgt.name) AS "Count"
\r
1905 INNER JOIN permission.grp_tree pgt
\r
1906 ON au.profile = pgt.id
\r
1907 INNER JOIN actor.org_unit aou
\r
1908 ON aou.id = au.home_ou
\r
1909 WHERE au.deleted IS FALSE
\r
1910 GROUP BY pgt.name, aou.name
\r
1911 ORDER BY aou.name, pgt.name
\r
1912 ;</programlisting>
\r
1913 </para></formalpara>
\r
1915 <simplesect id="_monthly_items_added_by_collection_library">
\r
1916 <title>Monthly items added by collection / library</title>
\r
1917 <simpara>We define a "collection" as a shelving location in Evergreen.</simpara>
\r
1918 <formalpara><title>Monthly Items Added by Collection/Library</title><para>
\r
1919 <programlisting language="sql" linenumbering="unnumbered">SELECT aou.name AS "Library", acl.name, COUNT(ac.barcode)
\r
1920 FROM actor.org_unit aou
\r
1921 INNER JOIN asset.call_number acn ON acn.owning_lib = aou.id
\r
1922 INNER JOIN asset.copy ac ON ac.call_number = acn.id
\r
1923 INNER JOIN asset.copy_location acl ON ac.location = acl.id
\r
1924 WHERE ac.deleted IS FALSE
\r
1925 AND acn.deleted IS FALSE
\r
1926 AND DATE_TRUNC('MONTH', ac.create_date) = DATE_TRUNC('MONTH', NOW() - '1 month'::interval)
\r
1927 GROUP BY aou.name, acl.name
\r
1928 ORDER BY aou.name, acl.name
\r
1929 ;</programlisting>
\r
1930 </para></formalpara>
\r
1932 <simplesect id="_hold_purchase_alert_by_library">
\r
1933 <title>Hold purchase alert by library</title>
\r
1934 <simpara>in the following set of queries, we bring together the active title, volume,
\r
1935 and copy holds and display those that have more than a certain number of holds
\r
1936 per title. The goal is to UNION ALL the three queries, then group by the
\r
1937 bibliographic record ID and display the title / author information for those
\r
1938 records that have more than a given threshold of holds.</simpara>
\r
1939 <formalpara><title>Hold Purchase Alert by Library</title><para>
\r
1940 <programlisting language="sql" linenumbering="unnumbered">-- Title holds
\r
1941 SELECT all_holds.bib_id, aou.name, rmsr.title, rmsr.author, COUNT(all_holds.bib_id)
\r
1945 SELECT target, request_lib
\r
1946 FROM action.hold_request
\r
1947 WHERE hold_type = 'T'
\r
1948 AND fulfillment_time IS NULL
\r
1949 AND cancel_time IS NULL
\r
1954 SELECT bre.id, request_lib
\r
1955 FROM action.hold_request ahr
\r
1956 INNER JOIN asset.call_number acn ON ahr.target = acn.id
\r
1957 INNER JOIN biblio.record_entry bre ON acn.record = bre.id
\r
1958 WHERE ahr.hold_type = 'V'
\r
1959 AND ahr.fulfillment_time IS NULL
\r
1960 AND ahr.cancel_time IS NULL
\r
1965 SELECT bre.id, request_lib
\r
1966 FROM action.hold_request ahr
\r
1967 INNER JOIN asset.copy ac ON ahr.target = ac.id
\r
1968 INNER JOIN asset.call_number acn ON ac.call_number = acn.id
\r
1969 INNER JOIN biblio.record_entry bre ON acn.record = bre.id
\r
1970 WHERE ahr.hold_type = 'C'
\r
1971 AND ahr.fulfillment_time IS NULL
\r
1972 AND ahr.cancel_time IS NULL
\r
1974 ) AS all_holds(bib_id, request_lib)
\r
1975 INNER JOIN reporter.materialized_simple_record rmsr
\r
1976 INNER JOIN actor.org_unit aou ON aou.id = all_holds.request_lib
\r
1977 ON rmsr.id = all_holds.bib_id
\r
1978 GROUP BY all_holds.bib_id, aou.name, rmsr.id, rmsr.title, rmsr.author
\r
1979 HAVING COUNT(all_holds.bib_id) > 2
\r
1981 ;</programlisting>
\r
1982 </para></formalpara>
\r
1984 <simplesect id="_update_borrower_records_with_a_different_home_library">
\r
1985 <title>Update borrower records with a different home library</title>
\r
1986 <simpara>In this example, the library has opened a new branch in a growing area,
\r
1987 and wants to reassign the home library for the patrons in the vicinity of
\r
1988 the new branch to the new branch. To accomplish this, we create a staging table
\r
1989 that holds a set of city names and the corresponding branch shortname for the home
\r
1990 library for each city.</simpara>
\r
1991 <simpara>Then we issue an <literal>UPDATE</literal> statement to set the home library for patrons with a
\r
1992 physical address with a city that matches the city names in our staging table.</simpara>
\r
1993 <formalpara><title>Update borrower records with a different home library</title><para>
\r
1994 <programlisting language="sql" linenumbering="unnumbered">CREATE SCHEMA staging;
\r
1995 CREATE TABLE staging.city_home_ou_map (city TEXT, ou_shortname TEXT,
\r
1996 FOREIGN KEY (ou_shortname) REFERENCES actor.org_unit (shortname));
\r
1997 INSERT INTO staging.city_home_ou_map (city, ou_shortname)
\r
1998 VALUES ('Southbury', 'BR1'), ('Middlebury', 'BR2'), ('Hartford', 'BR3');
\r
2001 UPDATE actor.usr au SET home_ou = COALESCE(
\r
2004 FROM actor.org_unit aou
\r
2005 INNER JOIN staging.city_home_ou_map schom ON schom.ou_shortname = aou.shortname
\r
2006 INNER JOIN actor.usr_address aua ON aua.city = schom.city
\r
2007 WHERE au.id = aua.usr
\r
2012 FROM actor.org_unit aou
\r
2013 INNER JOIN staging.city_home_ou_map schom ON schom.ou_shortname = aou.shortname
\r
2014 INNER JOIN actor.usr_address aua ON aua.city = schom.city
\r
2015 WHERE au.id = aua.usr
\r
2017 ) IS NOT NULL;</programlisting>
\r
2018 </para></formalpara>
\r