This improves commit c20c58595a to
definitely make sure the primary key values are stored and handled
correctly by simply using the same data type as SQLite itself does.
Apparently SQLite supports empty table names for whatever reason. This
commit patches SQLiteBrowser to support databases with a table with an
empty name. I haven't removed the check for an empty table name in the
Create/Edit table dialog though because you probably really shouldn't
create such a table.
See issue #194.
Primary keys in SQLite are 64bit in size. We, however, used the int
datatype which often is 32bit only. Also the conversion from QString and
other Qt datatypes to numbers was done by the toInt() method which fails
on these large numbers. These issues are fixed by this commit, adding
support for databases with big primary key values.
See issue #172.
When attaching another database, SQLCipher assumes that it shares
passphrase and salt with the main database. Usually this isn't the case,
though. With this commit SQLiteBrowser tries to open the database to attach
before actually attaching it and asks for encryption details if needed.
This allows the user to attach unencrypted databases to an encrypted
master database and vice-versa. Note though, that attaching an encrypted
database with a non-default (i.e. <> 1024) cipher page size doesn't seem
to work - this however, is apparently a SQLCipher limitation.
See issue #174.
The behavior remains basically as before: when first launched
the application will try to load a translation for the user's
locale, and if one cannot be found it defaults to English.
The difference is that now this is remembered so that upon
further launches the program will go straight to the matching
locale.
See issue #182.
The argument count (first parameter) is a reference, and thus must be
kept alive for the whole lifetime of the QCoreApplication instance,
as also the Qt apidocs say.
Also properly create a "string array" for the actual args, instead
of badly casting a string to that.
This fixes sporadic crashes in this test.
Fix parsing of table definitions like this one where there is a complex
check constraint using string literals:
CREATE TABLE "a" (
`b` TEXT CHECK(`b`='A' or `b`='B')
);
The grammar parser would fail to parse this statement correctly prior to
this fix.
See issue #179.
When entering a default value like "=(strftime('%s','now'))", don't
quote it but remove the '=' character and use the following expression
as default value.
See issue #166.
Show 'NULL' for those fields in a table which contain a NULL value. This
allows the user to see the difference between an empty string field and
a NULL field.
Also change the format of the 'BLOB' information shown for binary data
to match the new 'NULL' information.
See issue #163.
When saving a SQL file in the Execute Query tab don't always ask for a
new file name but use the file name used when last saving the tab or
used when a file was opened. Add a 'Save As' menu option with the old
behaviour of always asking for a file name.
See issue #152.
Apparently after querying an encrypted database the handle is left in an
inconsistent state. So before setting any key or other encryption
details make sure to close and reopen the database to get a new, valid
handle.
Thanks to Nick (developernotes) and Stephen (sjlombardo) from the
SQLCipher forums for pointing this out!
Martin Kleusberg