Also switch BOOLEAN constructor from sal_Bool to bool.
old/new signed/unsigned storage situation:
-------------------------------------------------------
SQL type | signed | unsigned old | unsigned new
-------------------------------------------------------
TINYINT | sal_Int8 | sal_Int16 | sal_uInt8
SMALLINT | sal_Int16 | sal_Int32 | sal_uInt16
INTEGER | sal_Int32 | sal_Int64 | sal_uInt32
BIGINT | sal_Int64 | pValue (String*) | sal_uInt64
-------------------------------------------------------
When sticking an UNSIGNED TINYINT into an Any,
silently promote it to UNSIGNED SMALLINT (that is sal_uInt16),
else Any would take it as a sal_Bool and normalise to
sal_True (1) or sal_False (0).
When constructing an ORowSetValue from a sal_Bool,
silently keep it as an unsigned 8 bit integer
(that is understand it as a sal_uInt8).
This will work in most cases,
since when asked back for a bool or sal_Bool,
we'll give back the right value.
Only code looking at the type tag could possibly
make a "wrong" decision.
The main (hopefully only?) path
through which this would happen is
through an implementation of
XParameters::setBoolean
XRowUpdate::updateBoolean
that would use its sal_Bool argument
to construct an ORowSetValue.
So make sure each implementation
constructs a proper BOOLEAN so as not to get confused.
For authorship/copyright purposes, this patch is a cooperation between
Lionel Elie Mamane <lionel@mamane.lu>
and
David Ostrovsky <david@ostrovsky.org>
Change-Id: I3f1f08716127147f077bff4edb6ec558b1b09e09
- Sanitized some OUStringBuilder, avoiding creating a new one and after
make some appends
- Removed some ::rtl prefixes
- Remove RTL_* macro
Change-Id: Ide3d78f20c68774cd4864b82cb8d29784228d1cd
Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/1552
Reviewed-by: Miklos Vajna <vmiklos@suse.cz>
Tested-by: Miklos Vajna <vmiklos@suse.cz>
This was done for the sake of ODBC,
but the cost was imposed on all backends.
The ODBC problems are now solved cleanly (and more efficiently)
in the SDBC<->ODBC layer.
Change-Id: Ib8a864da08deaaacc96a379fb72b3b7cbb34598c
For example dbaccess::OSingleSelectQueryComposer::appendOrderByColumn expects it
(via impl_getColumnName_throw via getTableAlias)
There is some vagueness:
Should the TableName property contain just the table name,
or the *composed* table name
(that is with catalog and/or schema if used by this DB)?
In the case of a query, should it contain the table name (alias)
*in* *the* *query* or of the original table?
In the former case, what meaning do SchemaName and CatalogName have?
They should be empty?
For now, commit as such and deal with the fallout, if any,
when it hits the fan.
If we really need to store these *different* values,
(that is, some code validly needs them)
it would be easier / cleaner / ...
to define *different* properties
for these *different* notions.
Change-Id: I032e619a60e7563cd51478db16cb5e0e5452bfde
Especially since the rest of the function is prepared to handle
no/empty (Catalog|Schema|Table)Name.
Change-Id: Ic0bb59ead5789e671c90887ef850588f4924f5e7
* Made XDatabaseContext inherit XDatabaseRegistrations non-optionally, adapted
call-sites to just use XDatabaseContext w/o querying. (The previous commit
had inadvertantly effectively removed support for XDatabaseRegistrations from
the ODatabaseContext implementation, as an optional UNO super-interface does
not lead to a super-class in the corresponding C++ class hierarchy, but making
the super-interface non-optional fixes that anyway.)
* Adapted some more call-sites to just use XDatabaseContext w/o querying.
* Added @since tag.
* Replaced new uses of comphelper::ComponentContext::getUNOContext with
comphelper::getComponentContext (see 03a9f139bd9ea1a4f9096fc982e6b326def58532
"ComponentContext::getUnoContext -> getComponentContext simplification;" I
intend to get rid of comphelper/componentcontext.hxx much sooner than of
comphelper/processfactory.hxx).
Change-Id: I68d09f2dbe651629f79ed21cd40cdb6d6b32c624
Create a merged XDatabaseContext interface for this service to implement.
Which is backwards-compatible, but does not require creating a new service.
Quite a few IDL files had to be marked as published for this to work.
Change-Id: Ie9a0da88d8c33cc83fc9d2334ff83ab2744c222f
The implementation of the LocaleData implements the optional XLocaleData4,
so rather than creating a new interface for the new-style service, we simply
make the service implement XLocaleData4, which in turn implements
XLocaleData3, XLocaleData2, XLocaleData.
Change-Id: I3e9a48b031be6b2aa5e04b376b3940b942add85a
Create a merged XNumberFormatter2 interface for this service to implement.
Which is backwards-compatible, but does not require creating a new service.
Change-Id: I57f35cde0a9dbbe91c1d2c3d068cb3a97c7245e3
This avoids fetching data that will not be requested when the "cursor" is only moved and no data requested. That is typically what RowSetCache does when its own cursor is moved within its window.
This basically makes the whole {next,previous,absolute,...}_checked story obsolete, by basically always being as fast as the i_bFetchRow==false case, but in a safer way.
Change-Id: I89eaf277069736b3077bde8b45325929db290f2d
... BOOST_STATIC_ASSERT_MSG was added in boost 1.46, while the internal
boost is still at version 1.44, so use BOOST_STATIC_ASSERT instead.
Change-Id: I14f8e48e31956b34a1a907cd2c4e454a5715889b
