1
- add columns() to row that returns a boost::tuple of various types so multple columns can be fetched at once
2
- look in to using BOOST_PP_ITERATE macro
3
- make basic_statement and database keep a shared pointer to the database handle so the classes can be made copyable. Also:
4
- the wrappers around the handle can clean them up after use
5
- the actual wrappers wround the handles can be made in sqlite::detail
6
- this will also make the implementation od rows (to get round the forced non-dependency of rows on querys) a little easier to swallow.
7
- committing a transaction during a query (i.e., when sqlite3_step() has returned SQLITE_ROW) causes an error. To counter this:
8
- need to check it's not fixed in latest sqlite; write test program
9
- calling query.reset() before the commit fixes the issue
10
- will need to keep a list of querys that need resetting in the database :o(
1
- add columns() to row that returns a boost::tuple of various types so multple
2
columns can be fetched at once (look in to using BOOST_PP_ITERATE macro)
4
- make basic_statement and database keep a shared pointer to the database handle
5
so the classes can be made copyable. The wrappers around the handle
6
(implemented in sqlite::detail) can clean them up after use. This will also
7
make the implementation of rows (to get round the forced non-dependency of
8
rows on querys) a little easier to swallow.
9
- A similar wrapper should be created for statement handles, making
10
basic_statements, querys and commands copyable. Could weak_ptrs to these
11
also be used in the database's list active querys?
13
- fix to force the finalisation of queries in progress for transactions causes
14
errors; queries are now finalised twice, the second from basic_statement's
15
dtor, which causes a segfault. We could:
16
- keep a list of force-finalised sqlite3_stmt pointers in the database which
17
we use to check queries against before finalising them to make sure we
18
don't finalise them a second time
19
- an efficient implementation, but not very OO
20
- keep a map of active queries in the database (using the sqlite3_stmt
21
pointer as the key), so that we can obtain the query and tell it to
23
- this seems like a messy and complicated implementation
11
25
- turn on extended errcodes in open() and handle them in sqlite_error
12
- use sqlite3_db_mutex() to provide extended error information during sqlite_error construction - see sqlite::query::step() for example
13
- expand sqlite_error - perhaps use boost::system_error (see boost/asio/error.hpp for an example of extending system_error)
27
- use sqlite3_db_mutex() to provide extended error information during
28
sqlite_error construction. The genreeal procedure would be to lock the db
29
mutex, perform some sqlite3 command, check the error code, throw an
30
sqlite_error (whilst obtaining extended error info) and then unlock the db
32
- a macro would be simple
33
- a templated safe-calling object (passing the comman's arg types as
34
template params) may be overkill
36
- expand sqlite_error - perhaps use boost::system_error (see
37
boost/asio/error.hpp for an example of extending system_error)
14
39
- see if we can #include "sqlite.h" in to a namespace.
16
41
we better encapsulate the library
17
42
we can reuse "sqlite3" as a namespace
19
makes access to real sqlite stuff awkward to sqlite3cc users, but does this matter? they can't access database._handle anyway!
20
potential incompatibility when linking to libraries that also link against sqlite
21
- fix step() inconsistency - query::step() returns a row, whereas basic_statement::step() and command::step() return an int return code
44
makes access to real sqlite stuff awkward to sqlite3cc users, but does
45
this matter? they can't access database._handle anyway!
46
potential incompatibility when linking to libraries that also link
22
49
- query::prepare() isn't being called during construction (form
23
50
basic_statement's constructor)
added
removed
TODO
