home > db.select

db.select

Other languages: 简体中文

Problem

You want to select data from a database

Solution

With web.py version 0.3 and later releases, databases are defined like this:

db = web.database(dbn='postgres', db='mydata', user='dbuser', pw='')

Once the database is defined as such, you can perform selects like this:

# Select all rows from table 'mytable'.
results = db.select('mytable')

The query result (results in above example) is an iterator of dict-like storage items, you can access them like below:

len(results)        # number of rows returned
results[0]          # first row returned
results[0]['name']  # value of sql column `name` in first row returned

Note: since the query result is a iterator, you cannot access it repeatly:

print(results[0])   # first access works.
print(results[1])   # FAILED with IndexError

If you need to access it repeatly, please convert it to a list first:

qr = db.select('mytable')
results = list(qr)
print(results[0])   # it works
print(results[1])   # works too

If SQL column is defined as binary format, e.g. VARBINARY in MySQL/MariaDB, BYTEA in PostgreSQL, returned value will be a bytes string, not str. For example:

  1. Create a SQL table in MySQL/MariaDB with command: CREATE TABLE mytable (email VARBINARY(255));
  2. Insert a sample record: INSERT INTO mytable (email) VALUES ("test@domain.com");
  3. Query it with web.py db module:
qr = db.select("mytable", what="email", limit=1)
email = qr[0]['email']
print(email)         // result is bytes: b'test@domain.com', not str 'test@domain.com'

The select statement takes the following keyword arguments:

  • vars
  • what
  • where
  • order
  • group
  • limit
  • offset
  • _test

vars

The vars variable is used to populate the rest of the statements. For example:

myvar = dict(name="Bob")

# Same as SQL statement: SELECT * FROM mytable WHERE name="Bob"
results = db.select('mytable', vars=myvar, where="name = $name")

what

The what variable defaults to *, but can take a list of items you want selected if the entire entry isn’t desired.

# Same as SQL statement: SELECT id, name FROM mytable
results = db.select('mytable', what="id,name")

where

The where variable lets you pass where clauses to the SQL select, such as:

# Same as SQL statement: SELECT * FROM mytable WHERE id>100
results = db.select('mytable', where="id>100")

order

The order variable lets the order be specified. For instance:

# Same as SQL statement: SELECT * FROM mytable ORDER BY post_date DESC
results = db.select('mytable', order="post_date DESC")

group

Grouping lets you combine things that are common.

# Same as SQL statement: SELECT * FROM mytable GROUP BY color
results = db.select('mytable', group="color")

limit

Limits set how many results are returned.

# Same as SQL statement: SELECT * FROM mytable LIMIT 10
results = db.select('mytable', limit=10)

offset

Offsets start returning results after a certain point; they’re often used with limits to do something like show 10 entries per page, and then see the next 10.

# Same as SQL statement: SELECT * FROM mytable OFFSET 10 LIMIT 20
results = db.select('mytable', offset=10, limit=20)

# Same as SQL statement: SELECT * FROM mytable OFFSET 30 LIMIT 20
results = db.select('mytable', offset=30, limit=20)

_test

The _test variable lets you see the SQL produced by the statement, no statement is actually performed:

results = db.select('mytable', offset=10, _test=True)
# `results` is a string: <sql: 'SELECT * FROM mytable OFFSET 10'>