1
- check that the fix for in-progress queries during rollback is threadsafe. In
2
particular, we shouldn't be resetting queries from another thread! Does
3
sqlite3_next_stmt() only return statements from this thread?
5
- turn on extended errcodes in open() and handle them in sqlite_error
7
- make basic_statement and database keep a shared pointer to the database handle
8
so the classes can be made copyable. The wrappers around the handle
9
(implemented in sqlite::detail) can clean them up after use. This will also
10
make the implementation of rows (to get round the forced non-dependency of
11
rows on querys) a little easier to swallow.
12
- A similar wrapper should be created for statement handles, making
13
basic_statements, querys and commands copyable. Could weak_ptrs to these
14
also be used in the database's list active querys?
16
- add columns() to row that returns a boost::tuple of various types so multple
17
columns can be fetched at once (look in to using BOOST_PP_ITERATE macro)
19
- use sqlite3_db_mutex() to provide extended error information during
20
sqlite_error construction. The general procedure would be to lock the db
21
mutex, perform some sqlite3 command, check the error code, throw an
22
sqlite_error (whilst obtaining extended error info) and then unlock the db
24
- a macro would be simple
25
- a templated safe-calling object (passing the comman's arg types as
26
template params) may be overkill
28
- expand sqlite_error - perhaps use boost::system_error (see
29
boost/asio/error.hpp for an example of extending system_error)
31
- see if we can #include "sqlite.h" in to a namespace.
33
we better encapsulate the library
34
we can reuse "sqlite3" as a namespace
36
makes access to real sqlite stuff awkward to sqlite3cc users, but does
37
this matter? they can't access database._handle anyway!
38
potential incompatibility when linking to libraries that also link
41
- query::prepare() isn't being called during construction (form
42
basic_statement's constructor)