pike.git / lib / modules / Sql.pmod / pgsql.pike

version» Context lines:

pike.git/lib/modules/Sql.pmod/pgsql.pike:1808:   //! binary method for simple queries with small or no result sets.   //! Note that this mode causes all but the first query result of a list   //! of semicolon separated statements to be discarded.   //! @member int ":_sync"   //! Forces synchronous parsing on or off for statements.   //! Setting this to off can cause surprises because statements could   //! be parsed before the previous statements have been executed.   //! This can speed up parsing by increased parallelism.   //! @endmapping   //! + //! @note + //! The bindings-parameter passed to this function must remain unaltered + //! until the parameters have been sent to the database. The driver + //! currently does not expose this moment, but to avoid a race condition + //! it is sufficient to keep them unaltered until the first resultrow + //! has been fetched (or EOF is reached, in case of no resultrows). + //!   //! @returns   //! A @[Sql.pgsql_util.sql_result] object (which conforms to the   //! @[Sql.sql_result] standard interface for accessing data). It is   //! recommended to use @[Sql.Sql()->query()] for simpler queries (because   //! it is easier to handle, but stores all the result in memory), and   //! @[Sql.Sql()->big_query()] for queries you expect to return huge amounts of   //! data (it's harder to handle, but fetches results on demand).   //!   //! @note   //! This function @b{can@} raise exceptions.