postgresql-simple-0.5.4.0: Mid-Level PostgreSQL client library

Copyright(c) 2011 MailRank Inc.
(c) 2011-2012 Leon P Smith
LicenseBSD3
MaintainerLeon P Smith <leon@melding-monads.com>
Stabilityexperimental
Safe HaskellNone
LanguageHaskell2010

Database.PostgreSQL.Simple

Contents

Description

A mid-level client library for the PostgreSQL database, aimed at ease of use and high performance.

Synopsis

Writing queries

SQL-based applications are somewhat notorious for their susceptibility to attacks through the injection of maliciously crafted data. The primary reason for widespread vulnerability to SQL injections is that many applications are sloppy in handling user data when constructing SQL queries.

This library provides a Query type and a parameter substitution facility to address both ease of use and security.

The Query type

A Query is a newtype-wrapped ByteString. It intentionally exposes a tiny API that is not compatible with the ByteString API; this makes it difficult to construct queries from fragments of strings. The query and execute functions require queries to be of type Query.

To most easily construct a query, enable GHC's OverloadedStrings language extension and write your query as a normal literal string.

{-# LANGUAGE OverloadedStrings #-}

import Database.PostgreSQL.Simple

hello :: IO Int
hello = do
  conn <- connectPostgreSQL ""
  [Only i] <- query_ conn "select 2 + 2"
  return i

A Query value does not represent the actual query that will be executed, but is a template for constructing the final query.

Parameter substitution

Since applications need to be able to construct queries with parameters that change, this library provides a query substitution capability.

The Query template accepted by query and execute can contain any number of "?" characters. Both query and execute accept a third argument, typically a tuple. When constructing the real query to execute, these functions replace the first "?" in the template with the first element of the tuple, the second "?" with the second element, and so on. If necessary, each tuple element will be quoted and escaped prior to substitution; this defeats the single most common injection vector for malicious data.

For example, given the following Query template:

select * from user where first_name = ? and age > ?

And a tuple of this form:

("Boris" :: String, 37 :: Int)

The query to be executed will look like this after substitution:

select * from user where first_name = 'Boris' and age > 37

If there is a mismatch between the number of "?" characters in your template and the number of elements in your tuple, a FormatError will be thrown.

Note that the substitution functions do not attempt to parse or validate your query. It's up to you to write syntactically valid SQL, and to ensure that each "?" in your query template is matched with the right tuple element.

Type inference

Automated type inference means that you will often be able to avoid supplying explicit type signatures for the elements of a tuple. However, sometimes the compiler will not be able to infer your types. Consider a case where you write a numeric literal in a parameter tuple:

query conn "select ? + ?" (40,2)

The above query will be rejected by the compiler, because it does not know the specific numeric types of the literals 40 and 2. This is easily fixed:

query conn "select ? + ?" (40 :: Double, 2 :: Double)

The same kind of problem can arise with string literals if you have the OverloadedStrings language extension enabled. Again, just use an explicit type signature if this happens.

Finally, remember that the compiler must be able to infer the type of a query's results as well as its parameters. We might like the following example to work:

print =<< query_ conn "select 2 + 2"

Unfortunately, while a quick glance tells us that the result type should be a single row containing a single numeric column, the compiler has no way to infer what the types are. We can easily fix this by providing an explicit type annotation:

xs <- query_ conn "select 2 + 2"
print (xs :: [Only Int])

Substituting a single parameter

Haskell lacks a single-element tuple type, so if you have just one value you want substituted into a query or a single-column result, what should you do?

The obvious approach would appear to be something like this:

instance (ToField a) => ToRow a where
    ...

Unfortunately, this wreaks havoc with type inference, so we take a different tack. To represent a single value val as a parameter, write a singleton list [val], use Just val, or use Only val.

Here's an example using a singleton list:

execute conn "insert into users (first_name) values (?)"
             ["Nuala"]

A row of n query results is represented using an n-tuple, so you should use Only to represent a single-column result.

Representing a list of values

Suppose you want to write a query using an IN clause:

select * from users where first_name in ('Anna', 'Boris', 'Carla')

In such cases, it's common for both the elements and length of the list after the IN keyword to vary from query to query.

To address this case, use the In type wrapper, and use a single "?" character to represent the list. Omit the parentheses around the list; these will be added for you.

Here's an example:

query conn "select * from users where first_name in ?" $
      Only $ In ["Anna", "Boris", "Carla"]

If your In-wrapped list is empty, the string "(null)" will be substituted instead, to ensure that your clause remains syntactically valid.

Modifying multiple rows at once

If you know that you have many rows of data to insert into a table, it is much more efficient to perform all the insertions in a single multi-row INSERT statement than individually.

The executeMany function is intended specifically for helping with multi-row INSERT and UPDATE statements. Its rules for query substitution are different than those for execute.

What executeMany searches for in your Query template is a single substring of the form:

values (?,?,?)

The rules are as follows:

  • The keyword VALUES is matched case insensitively.
  • There must be no other "?" characters anywhere in your template.
  • There must be one or more "?" in the parentheses.
  • Extra white space is fine.

The last argument to executeMany is a list of parameter tuples. These will be substituted into the query where the (?,?) string appears, in a form suitable for use in a multi-row INSERT or UPDATE.

Here is an example:

executeMany conn
  "insert into users (first_name,last_name) values (?,?)"
  [("Boris","Karloff"),("Ed","Wood")]

The query that will be executed here will look like this (reformatted for tidiness):

insert into users (first_name,last_name) values
  ('Boris','Karloff'),('Ed','Wood')

RETURNING: modifications that return results

PostgreSQL supports returning values from data manipulation statements such as INSERT and UPDATE. You can use these statements by using query instead of execute. For multi-tuple inserts, use returning instead of executeMany.

For example, were there an auto-incrementing id column and timestamp column t that defaulted to the present time for the sales table, then the following query would insert two new sales records and also return their new ids and timestamps.

let q = "insert into sales (amount, label) values (?,?) returning id, t"
xs :: [(Int, UTCTime)] <- query conn q (15,"Sawdust")
ys :: [(Int, UTCTime)] <- returning conn q [(20,"Chips"),(300,"Wood")]

Extracting results

The query and query_ functions return a list of values in the FromRow typeclass. This class performs automatic extraction and type conversion of rows from a query result.

Here is a simple example of how to extract results:

import qualified Data.Text as Text

xs <- query_ conn "select name,age from users"
forM_ xs $ \(name,age) ->
  putStrLn $ Text.unpack name ++ " is " ++ show (age :: Int)

Notice two important details about this code:

  • The number of columns we ask for in the query template must exactly match the number of elements we specify in a row of the result tuple. If they do not match, a ResultError exception will be thrown.
  • Sometimes, the compiler needs our help in specifying types. It can infer that name must be a Text, due to our use of the unpack function. However, we have to tell it the type of age, as it has no other information to determine the exact type.

Handling null values

The type of a result tuple will look something like this:

(Text, Int, Int)

Although SQL can accommodate NULL as a value for any of these types, Haskell cannot. If your result contains columns that may be NULL, be sure that you use Maybe in those positions of your tuple.

(Text, Maybe Int, Int)

If query encounters a NULL in a row where the corresponding Haskell type is not Maybe, it will throw a ResultError exception.

Type conversions

Conversion of SQL values to Haskell values is somewhat permissive. Here are the rules.

  • For numeric types, any Haskell type that can accurately represent all values of the given PostgreSQL type is considered "compatible". For instance, you can always extract a PostgreSQL 16-bit SMALLINT column to a Haskell Int. The Haskell Float type can accurately represent a SMALLINT, so it is considered compatble with those types.
  • A numeric compatibility check is based only on the type of a column, not on its values. For instance, a PostgreSQL 64-bit BIGINT column will be considered incompatible with a Haskell Int16, even if it contains the value 1.
  • If a numeric incompatibility is found, query will throw a ResultError.
  • The String and Text types are assumed to be encoded as UTF-8. If you use some other encoding, decoding may fail or give wrong results. In such cases, write a newtype wrapper and a custom Result instance to handle your encoding.

Types

data Query Source #

A query string. This type is intended to make it difficult to construct a SQL query by concatenating string fragments, as that is an extremely common way to accidentally introduce SQL injection vulnerabilities into an application.

This type is an instance of IsString, so the easiest way to construct a query is to enable the OverloadedStrings language extension and then simply write the query in double quotes.

{-# LANGUAGE OverloadedStrings #-}

import Database.PostgreSQL.Simple

q :: Query
q = "select ?"

The underlying type is a ByteString, and literal Haskell strings that contain Unicode characters will be correctly transformed to UTF-8.

class ToRow a Source #

A collection type that can be turned into a list of rendering Actions.

Instances should use the toField method of the ToField class to perform conversion of each element of the collection.

Instances

ToRow () Source # 

Methods

toRow :: () -> [Action] Source #

ToField a => ToRow [a] Source # 

Methods

toRow :: [a] -> [Action] Source #

ToField a => ToRow (Only a) Source # 

Methods

toRow :: Only a -> [Action] Source #

(ToField a, ToField b) => ToRow (a, b) Source # 

Methods

toRow :: (a, b) -> [Action] Source #

(ToRow a, ToRow b) => ToRow ((:.) a b) Source # 

Methods

toRow :: (a :. b) -> [Action] Source #

(ToField a, ToField b, ToField c) => ToRow (a, b, c) Source # 

Methods

toRow :: (a, b, c) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d) => ToRow (a, b, c, d) Source # 

Methods

toRow :: (a, b, c, d) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e) => ToRow (a, b, c, d, e) Source # 

Methods

toRow :: (a, b, c, d, e) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e, ToField f) => ToRow (a, b, c, d, e, f) Source # 

Methods

toRow :: (a, b, c, d, e, f) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e, ToField f, ToField g) => ToRow (a, b, c, d, e, f, g) Source # 

Methods

toRow :: (a, b, c, d, e, f, g) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e, ToField f, ToField g, ToField h) => ToRow (a, b, c, d, e, f, g, h) Source # 

Methods

toRow :: (a, b, c, d, e, f, g, h) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e, ToField f, ToField g, ToField h, ToField i) => ToRow (a, b, c, d, e, f, g, h, i) Source # 

Methods

toRow :: (a, b, c, d, e, f, g, h, i) -> [Action] Source #

(ToField a, ToField b, ToField c, ToField d, ToField e, ToField f, ToField g, ToField h, ToField i, ToField j) => ToRow (a, b, c, d, e, f, g, h, i, j) Source # 

Methods

toRow :: (a, b, c, d, e, f, g, h, i, j) -> [Action] Source #

class FromRow a Source #

A collection type that can be converted from a sequence of fields. Instances are provided for tuples up to 10 elements and lists of any length.

Note that instances can be defined outside of postgresql-simple, which is often useful. For example, here's an instance for a user-defined pair:

data User = User { name :: String, fileQuota :: Int }

instance FromRow User where
    fromRow = User <$> field <*> field

The number of calls to field must match the number of fields returned in a single row of the query result. Otherwise, a ConversionFailed exception will be thrown.

Note that field evaluates its result to WHNF, so the caveats listed in mysql-simple and very early versions of postgresql-simple no longer apply. Instead, look at the caveats associated with user-defined implementations of fromField.

Instances

FromField a => FromRow [a] Source # 

Methods

fromRow :: RowParser [a] Source #

FromField a => FromRow (Maybe [a]) Source # 

Methods

fromRow :: RowParser (Maybe [a]) Source #

(FromField a, FromField b) => FromRow (Maybe (a, b)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b)) Source #

(FromField a, FromField b, FromField c) => FromRow (Maybe (a, b, c)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c)) Source #

(FromField a, FromField b, FromField c, FromField d) => FromRow (Maybe (a, b, c, d)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e) => FromRow (Maybe (a, b, c, d, e)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f) => FromRow (Maybe (a, b, c, d, e, f)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e, f)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g) => FromRow (Maybe (a, b, c, d, e, f, g)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e, f, g)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h) => FromRow (Maybe (a, b, c, d, e, f, g, h)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e, f, g, h)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h, FromField i) => FromRow (Maybe (a, b, c, d, e, f, g, h, i)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e, f, g, h, i)) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h, FromField i, FromField j) => FromRow (Maybe (a, b, c, d, e, f, g, h, i, j)) Source # 

Methods

fromRow :: RowParser (Maybe (a, b, c, d, e, f, g, h, i, j)) Source #

FromField a => FromRow (Maybe (Vector a)) Source # 
FromField a => FromRow (Maybe (Only a)) Source # 
FromField a => FromRow (Vector a) Source # 
FromField a => FromRow (Only a) Source # 

Methods

fromRow :: RowParser (Only a) Source #

(FromField a, FromField b) => FromRow (a, b) Source # 

Methods

fromRow :: RowParser (a, b) Source #

(FromRow a, FromRow b) => FromRow ((:.) a b) Source # 

Methods

fromRow :: RowParser (a :. b) Source #

(FromField a, FromField b, FromField c) => FromRow (a, b, c) Source # 

Methods

fromRow :: RowParser (a, b, c) Source #

(FromField a, FromField b, FromField c, FromField d) => FromRow (a, b, c, d) Source # 

Methods

fromRow :: RowParser (a, b, c, d) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e) => FromRow (a, b, c, d, e) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f) => FromRow (a, b, c, d, e, f) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e, f) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g) => FromRow (a, b, c, d, e, f, g) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e, f, g) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h) => FromRow (a, b, c, d, e, f, g, h) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e, f, g, h) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h, FromField i) => FromRow (a, b, c, d, e, f, g, h, i) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e, f, g, h, i) Source #

(FromField a, FromField b, FromField c, FromField d, FromField e, FromField f, FromField g, FromField h, FromField i, FromField j) => FromRow (a, b, c, d, e, f, g, h, i, j) Source # 

Methods

fromRow :: RowParser (a, b, c, d, e, f, g, h, i, j) Source #

newtype In a Source #

Wrap a list of values for use in an IN clause. Replaces a single "?" character with a parenthesized list of rendered values.

Example:

query c "select * from whatever where id in ?" (Only (In [3,4,5]))

Note that In [] expands to (null), which works as expected in the query above, but evaluates to the logical null value on every row instead of TRUE. This means that changing the query above to ... id NOT in ? and supplying the empty list as the parameter returns zero rows, instead of all of them as one would expect.

Since postgresql doesn't seem to provide a syntax for actually specifying an empty list, which could solve this completely, there are two workarounds particularly worth mentioning, namely:

  1. Use postgresql-simple's Values type instead, which can handle the empty case correctly. Note however that while specifying the postgresql type "int4" is mandatory in the empty case, specifying the haskell type Values (Only Int) would not normally be needed in realistic use cases.

    query c "select * from whatever where id not in ?"
            (Only (Values ["int4"] [] :: Values (Only Int)))
  2. Use sql's COALESCE operator to turn a logical null into the correct boolean. Note however that the correct boolean depends on the use case:

    query c "select * from whatever where coalesce(id NOT in ?, TRUE)"
            (Only (In [] :: In [Int]))
    query c "select * from whatever where coalesce(id IN ?, FALSE)"
            (Only (In [] :: In [Int]))

    Note that at as of PostgreSQL 9.4, the query planner cannot see inside the COALESCE operator, so if you have an index on id then you probably don't want to write the last example with COALESCE, which would result in a table scan. There are further caveats if id can be null or you want null treated sensibly as a component of IN or NOT IN.

Constructors

In a 

Instances

Functor In Source # 

Methods

fmap :: (a -> b) -> In a -> In b #

(<$) :: a -> In b -> In a #

Eq a => Eq (In a) Source # 

Methods

(==) :: In a -> In a -> Bool #

(/=) :: In a -> In a -> Bool #

Ord a => Ord (In a) Source # 

Methods

compare :: In a -> In a -> Ordering #

(<) :: In a -> In a -> Bool #

(<=) :: In a -> In a -> Bool #

(>) :: In a -> In a -> Bool #

(>=) :: In a -> In a -> Bool #

max :: In a -> In a -> In a #

min :: In a -> In a -> In a #

Read a => Read (In a) Source # 
Show a => Show (In a) Source # 

Methods

showsPrec :: Int -> In a -> ShowS #

show :: In a -> String #

showList :: [In a] -> ShowS #

ToField a => ToField (In [a]) Source # 

Methods

toField :: In [a] -> Action Source #

newtype Binary a Source #

Wrap binary data for use as a bytea value.

Constructors

Binary 

Fields

newtype Only a Source #

A single-value "collection".

This is useful if you need to supply a single parameter to a SQL query, or extract a single column from a SQL result.

Parameter example:

query c "select x from scores where x > ?" (Only (42::Int))

Result example:

xs <- query_ c "select id from users"
forM_ xs $ \(Only id) -> {- ... -}

Constructors

Only 

Fields

Instances

Functor Only Source # 

Methods

fmap :: (a -> b) -> Only a -> Only b #

(<$) :: a -> Only b -> Only a #

Eq a => Eq (Only a) Source # 

Methods

(==) :: Only a -> Only a -> Bool #

(/=) :: Only a -> Only a -> Bool #

Ord a => Ord (Only a) Source # 

Methods

compare :: Only a -> Only a -> Ordering #

(<) :: Only a -> Only a -> Bool #

(<=) :: Only a -> Only a -> Bool #

(>) :: Only a -> Only a -> Bool #

(>=) :: Only a -> Only a -> Bool #

max :: Only a -> Only a -> Only a #

min :: Only a -> Only a -> Only a #

Read a => Read (Only a) Source # 
Show a => Show (Only a) Source # 

Methods

showsPrec :: Int -> Only a -> ShowS #

show :: Only a -> String #

showList :: [Only a] -> ShowS #

ToField a => ToRow (Only a) Source # 

Methods

toRow :: Only a -> [Action] Source #

FromField a => FromRow (Maybe (Only a)) Source # 
FromField a => FromRow (Only a) Source # 

Methods

fromRow :: RowParser (Only a) Source #

data h :. t infixr 3 Source #

A composite type to parse your custom data structures without having to define dummy newtype wrappers every time.

instance FromRow MyData where ...
instance FromRow MyData2 where ...

then I can do the following for free:

res <- query' c "..."
forM res $ \(MyData{..} :. MyData2{..}) -> do
  ....

Constructors

h :. t infixr 3 

Instances

(Eq t, Eq h) => Eq ((:.) h t) Source # 

Methods

(==) :: (h :. t) -> (h :. t) -> Bool #

(/=) :: (h :. t) -> (h :. t) -> Bool #

(Ord t, Ord h) => Ord ((:.) h t) Source # 

Methods

compare :: (h :. t) -> (h :. t) -> Ordering #

(<) :: (h :. t) -> (h :. t) -> Bool #

(<=) :: (h :. t) -> (h :. t) -> Bool #

(>) :: (h :. t) -> (h :. t) -> Bool #

(>=) :: (h :. t) -> (h :. t) -> Bool #

max :: (h :. t) -> (h :. t) -> h :. t #

min :: (h :. t) -> (h :. t) -> h :. t #

(Read t, Read h) => Read ((:.) h t) Source # 

Methods

readsPrec :: Int -> ReadS (h :. t) #

readList :: ReadS [h :. t] #

readPrec :: ReadPrec (h :. t) #

readListPrec :: ReadPrec [h :. t] #

(Show t, Show h) => Show ((:.) h t) Source # 

Methods

showsPrec :: Int -> (h :. t) -> ShowS #

show :: (h :. t) -> String #

showList :: [h :. t] -> ShowS #

(ToRow a, ToRow b) => ToRow ((:.) a b) Source # 

Methods

toRow :: (a :. b) -> [Action] Source #

(FromRow a, FromRow b) => FromRow ((:.) a b) Source # 

Methods

fromRow :: RowParser (a :. b) Source #

Exceptions

data ExecStatus :: * #

Constructors

EmptyQuery

The string sent to the server was empty.

CommandOk

Successful completion of a command returning no data.

TuplesOk

Successful completion of a command returning data (such as a SELECT or SHOW).

CopyOut

Copy Out (from server) data transfer started.

CopyIn

Copy In (to server) data transfer started.

CopyBoth

Copy In/Out data transfer started.

BadResponse

The server's response was not understood.

NonfatalError

A nonfatal error (a notice or warning) occurred.

FatalError

A fatal error occurred.

SingleTuple

The PGresult contains a single result tuple from the current command. This status occurs only when single-row mode has been selected for the query.

data FormatError Source #

Exception thrown if a Query could not be formatted correctly. This may occur if the number of '?' characters in the query string does not match the number of parameters provided.

Constructors

FormatError 

data QueryError Source #

Exception thrown if query is used to perform an INSERT-like operation, or execute is used to perform a SELECT-like operation.

Constructors

QueryError 

Fields

data ResultError Source #

Exception thrown if conversion from a SQL value to a Haskell value fails.

Constructors

Incompatible

The SQL and Haskell types are not compatible.

UnexpectedNull

A SQL NULL was encountered when the Haskell type did not permit it.

ConversionFailed

The SQL value could not be parsed, or could not be represented as a valid Haskell value, or an unexpected low-level error occurred (e.g. mismatch between metadata and actual data in a row).

Connection management

connectPostgreSQL :: ByteString -> IO Connection Source #

Attempt to make a connection based on a libpq connection string. See https://www.postgresql.org/docs/9.5/static/libpq-connect.html#LIBPQ-CONNSTRING for more information. Also note that environment variables also affect parameters not provided, parameters provided as the empty string, and a few other things; see https://www.postgresql.org/docs/9.5/static/libpq-envars.html for details. Here is an example with some of the most commonly used parameters:

host='db.somedomain.com' port=5432 ...

This attempts to connect to db.somedomain.com:5432. Omitting the port will normally default to 5432.

On systems that provide unix domain sockets, omitting the host parameter will cause libpq to attempt to connect via unix domain sockets. The default filesystem path to the socket is constructed from the port number and the DEFAULT_PGSOCKET_DIR constant defined in the pg_config_manual.h header file. Connecting via unix sockets tends to use the peer authentication method, which is very secure and does not require a password.

On Windows and other systems without unix domain sockets, omitting the host will default to localhost.

... dbname='postgres' user='postgres' password='secret \' \\ pw'

This attempts to connect to a database named postgres with user postgres and password secret ' \ pw. Backslash characters will have to be double-quoted in literal Haskell strings, of course. Omitting dbname and user will both default to the system username that the client process is running as.

Omitting password will default to an appropriate password found in the pgpass file, or no password at all if a matching line is not found. See https://www.postgresql.org/docs/9.5/static/libpq-pgpass.html for more information regarding this file.

As all parameters are optional and the defaults are sensible, the empty connection string can be useful for development and exploratory use, assuming your system is set up appropriately.

On Unix, such a setup would typically consist of a local postgresql server listening on port 5432, as well as a system user, database user, and database sharing a common name, with permissions granted to the user on the database.

On Windows, in addition you will either need pg_hba.conf to specify the use of the trust authentication method for the connection, which may not be appropriate for multiuser or production machines, or you will need to use a pgpass file with the password or md5 authentication methods.

See https://www.postgresql.org/docs/9.5/static/client-authentication.html for more information regarding the authentication process.

SSL/TLS will typically "just work" if your postgresql server supports or requires it. However, note that libpq is trivially vulnerable to a MITM attack without setting additional SSL connection parameters. In particular, sslmode needs to be set to require, verify-ca, or verify-full in order to perform certificate validation. When sslmode is require, then you will also need to specify a sslrootcert file, otherwise no validation of the server's identity will be performed. Client authentication via certificates is also possible via the sslcert and sslkey parameters. See https://www.postgresql.org/docs/9.5/static/libpq-ssl.html for detailed information regarding libpq and SSL.

connect :: ConnectInfo -> IO Connection Source #

Connect with the given username to the given database. Will throw an exception if it cannot connect.

defaultConnectInfo :: ConnectInfo Source #

Default information for setting up a connection.

Defaults are as follows:

  • Server on localhost
  • Port on 5432
  • User postgres
  • No password
  • Database postgres

Use as in the following example:

connect defaultConnectInfo { connectHost = "db.example.com" }

postgreSQLConnectionString :: ConnectInfo -> ByteString Source #

Turns a ConnectInfo data structure into a libpq connection string.

Queries that return results

query :: (ToRow q, FromRow r) => Connection -> Query -> q -> IO [r] Source #

Perform a SELECT or other SQL query that is expected to return results. All results are retrieved and converted before this function returns.

When processing large results, this function will consume a lot of client-side memory. Consider using fold instead.

Exceptions that may be thrown:

  • FormatError: the query string could not be formatted correctly.
  • QueryError: the result contains no columns (i.e. you should be using execute instead of query).
  • ResultError: result conversion failed.
  • SqlError: the postgresql backend returned an error, e.g. a syntax or type error, or an incorrect table or column name.

query_ :: FromRow r => Connection -> Query -> IO [r] Source #

A version of query that does not perform query substitution.

Queries taking parser as argument

queryWith :: ToRow q => RowParser r -> Connection -> Query -> q -> IO [r] Source #

A version of query taking parser as argument

queryWith_ :: RowParser r -> Connection -> Query -> IO [r] Source #

A version of query_ taking parser as argument

Queries that stream results

data FetchQuantity Source #

Number of rows to fetch at a time. Automatic currently defaults to 256 rows, although it might be nice to make this more intelligent based on e.g. the average size of the rows.

Constructors

Automatic 
Fixed !Int 

fold :: (FromRow row, ToRow params) => Connection -> Query -> params -> a -> (a -> row -> IO a) -> IO a Source #

Perform a SELECT or other SQL query that is expected to return results. Results are streamed incrementally from the server, and consumed via a left fold.

When dealing with small results, it may be simpler (and perhaps faster) to use query instead.

This fold is not strict. The stream consumer is responsible for forcing the evaluation of its result to avoid space leaks.

This is implemented using a database cursor. As such, this requires a transaction. This function will detect whether or not there is a transaction in progress, and will create a ReadCommitted ReadOnly transaction if needed. The cursor is given a unique temporary name, so the consumer may itself call fold.

Exceptions that may be thrown:

  • FormatError: the query string could not be formatted correctly.
  • QueryError: the result contains no columns (i.e. you should be using execute instead of query).
  • ResultError: result conversion failed.
  • SqlError: the postgresql backend returned an error, e.g. a syntax or type error, or an incorrect table or column name.

foldWithOptions :: (FromRow row, ToRow params) => FoldOptions -> Connection -> Query -> params -> a -> (a -> row -> IO a) -> IO a Source #

The same as fold, but this provides a bit more control over lower-level details. Currently, the number of rows fetched per round-trip to the server and the transaction mode may be adjusted accordingly. If the connection is already in a transaction, then the existing transaction is used and thus the transactionMode option is ignored.

fold_ Source #

Arguments

:: FromRow r 
=> Connection 
-> Query

Query.

-> a

Initial state for result consumer.

-> (a -> r -> IO a)

Result consumer.

-> IO a 

A version of fold that does not perform query substitution.

foldWithOptions_ Source #

Arguments

:: FromRow r 
=> FoldOptions 
-> Connection 
-> Query

Query.

-> a

Initial state for result consumer.

-> (a -> r -> IO a)

Result consumer.

-> IO a 

forEach Source #

Arguments

:: (ToRow q, FromRow r) 
=> Connection 
-> Query

Query template.

-> q

Query parameters.

-> (r -> IO ())

Result consumer.

-> IO () 

A version of fold that does not transform a state value.

forEach_ Source #

Arguments

:: FromRow r 
=> Connection 
-> Query

Query template.

-> (r -> IO ())

Result consumer.

-> IO () 

A version of forEach that does not perform query substitution.

returning :: (ToRow q, FromRow r) => Connection -> Query -> [q] -> IO [r] Source #

Execute INSERT ... RETURNING, UPDATE ... RETURNING, or other SQL query that accepts multi-row input and is expected to return results. Note that it is possible to write query conn "INSERT ... RETURNING ..." ... in cases where you are only inserting a single row, and do not need functionality analogous to executeMany.

If the list of parameters is empty, this function will simply return [] without issuing the query to the backend. If this is not desired, consider using the Values constructor instead.

Throws FormatError if the query could not be formatted correctly.

Queries that stream results taking a parser as an argument

foldWith :: ToRow params => RowParser row -> Connection -> Query -> params -> a -> (a -> row -> IO a) -> IO a Source #

A version of fold taking a parser as an argument

foldWithOptionsAndParser :: ToRow params => FoldOptions -> RowParser row -> Connection -> Query -> params -> a -> (a -> row -> IO a) -> IO a Source #

A version of foldWithOptions taking a parser as an argument

foldWith_ :: RowParser r -> Connection -> Query -> a -> (a -> r -> IO a) -> IO a Source #

A version of fold_ taking a parser as an argument

foldWithOptionsAndParser_ Source #

Arguments

:: FoldOptions 
-> RowParser r 
-> Connection 
-> Query

Query.

-> a

Initial state for result consumer.

-> (a -> r -> IO a)

Result consumer.

-> IO a 

A version of foldWithOptions_ taking a parser as an argument

forEachWith :: ToRow q => RowParser r -> Connection -> Query -> q -> (r -> IO ()) -> IO () Source #

A version of forEach taking a parser as an argument

forEachWith_ :: RowParser r -> Connection -> Query -> (r -> IO ()) -> IO () Source #

returningWith :: ToRow q => RowParser r -> Connection -> Query -> [q] -> IO [r] Source #

Statements that do not return results

execute :: ToRow q => Connection -> Query -> q -> IO Int64 Source #

Execute an INSERT, UPDATE, or other SQL query that is not expected to return results.

Returns the number of rows affected.

Throws FormatError if the query could not be formatted correctly, or a SqlError exception if the backend returns an error.

execute_ :: Connection -> Query -> IO Int64 Source #

A version of execute that does not perform query substitution.

executeMany :: ToRow q => Connection -> Query -> [q] -> IO Int64 Source #

Execute a multi-row INSERT, UPDATE, or other SQL query that is not expected to return results.

Returns the number of rows affected. If the list of parameters is empty, this function will simply return 0 without issuing the query to the backend. If this is not desired, consider using the Values constructor instead.

Throws FormatError if the query could not be formatted correctly, or a SqlError exception if the backend returns an error.

For example, here's a command that inserts two rows into a table with two columns:

executeMany c [sql|
    INSERT INTO sometable VALUES (?,?)
 |] [(1, "hello"),(2, "world")]

Here's an canonical example of a multi-row update command:

executeMany c [sql|
    UPDATE sometable
       SET sometable.y = upd.y
      FROM (VALUES (?,?)) as upd(x,y)
     WHERE sometable.x = upd.x
 |] [(1, "hello"),(2, "world")]

Transaction handling

withTransaction :: Connection -> IO a -> IO a Source #

Execute an action inside a SQL transaction.

This function initiates a transaction with a "begin transaction" statement, then executes the supplied action. If the action succeeds, the transaction will be completed with commit before this function returns.

If the action throws any kind of exception (not just a PostgreSQL-related exception), the transaction will be rolled back using rollback, then the exception will be rethrown.

For nesting transactions, see withSavepoint.

withSavepoint :: Connection -> IO a -> IO a Source #

Create a savepoint, and roll back to it if an error occurs. This may only be used inside of a transaction, and provides a sort of "nested transaction".

See https://www.postgresql.org/docs/9.5/static/sql-savepoint.html

begin :: Connection -> IO () Source #

Begin a transaction.

commit :: Connection -> IO () Source #

Commit a transaction.

rollback :: Connection -> IO () Source #

Rollback a transaction.

Helper functions

formatMany :: ToRow q => Connection -> Query -> [q] -> IO ByteString Source #

Format a query string with a variable number of rows.

This function is exposed to help with debugging and logging. Do not use it to prepare queries for execution.

The query string must contain exactly one substitution group, identified by the SQL keyword "VALUES" (case insensitive) followed by an "(" character, a series of one or more "?" characters separated by commas, and a ")" character. White space in a substitution group is permitted.

Throws FormatError if the query string could not be formatted correctly.

formatQuery :: ToRow q => Connection -> Query -> q -> IO ByteString Source #

Format a query string.

This function is exposed to help with debugging and logging. Do not use it to prepare queries for execution.

String parameters are escaped according to the character set in use on the Connection.

Throws FormatError if the query string could not be formatted correctly.