* Empty destructors are replaced with default destructors
* A few loop variables made into const references to avoid copies
* Add missing `override` spec for some `Command::execute` methods
This new subcommand checks all passwords in the given database against a given list of SHA-1 password hashes. Such lists are available from the "Have I Been Pwned" project at https://haveibeenpwned.com/Passwords.
Note that this support offline checking only. The HIBP project also provides a web API for checking specific hash ranges; this is not currently supported.
* Fixes#925
* Add 'flatten' option to CLI ls command
* Add test for Group::hierarchy() and man page for ls --flatten
* Rename group sort test to align with others
This PR cleans up the `Command` classes in the CLI, introducing a
`DatabaseCommand` class for the commands operating on a database,
and a `getCommandLineParser` command to centralize the arguments
parsing and validation.
The opening of the database based on the CLI arguments and options
is now centralized in `DatabaseCommand.execute`, making it easy to
add new database opening features (like YubiKey support for the CLI).
Also a couple of bugs fixed:
* `Create` was still using `stdout` for some error messages.
* `Diceware` and `Generate` were not validating that the word count was an integer.
* `Diceware` was also using `stdout` for some error messages.
* Updated utilities to unlock KDBX with OS password manager on macOS and Linux
* Use a static stream on stdin for --pw-stdin otherwise buffer loss eliminates subsequent passwords
* Update INSTALL requirements
Previously, we enforced code page 850 for all console input and output,
which breaks with non-western scripts. Since more recent Windows shells
are able to display Unicode properly, this patch now enforces UTF-8 and
falls back to code page 850 only if UTF-8 is unsupported.
Non-Windows systems default to UTF-8, but can override the codec
by setting the LANG environment variable to something other than C.
Resolves#3049.
To reduce residual fragments of secret data in memory after
deallocation, this patch replaces the global delete operator with a
version that zeros out previously allocated memory. It makes use of
the new C++14 sized deallocation, but provides an unsized fallback
with platform-specific size deductions.
This change is only a minor mitigation and cannot protect against
buffer reallocations by the operating system or non-C++ libraries.
Thus, we still cannot guarantee all memory to be wiped after free.
As a further improvement, this patch uses libgcrypt and libsodium
to write long-lived master key component hashes into a secure
memory area and wipe it afterwards.
The patch also fixes compiler flags not being set properly on macOS.
Adding debug info to the CLI and the general option
of the main Qt app. Also took time to:
* use `EXIT_SUCCESS`/`EXIT_FAILURE` constants
for main.cpp (this is what is used in `src/cli`);
* fixed `m_initalized` typo;
* added info on debugging mode being disabled
or not;
* regrouped Qt related stuff in the debug output.
There are other man implementations beside man-db so it is not even sure
that the "mandb" binary even exists on all unices. Other than that, usually
there's a cron job running "mandb" on a daily basis.
Previously, extracting the XML from a database was done with the
`saveXml` attribute in the `KeePass2Reader` class.
This had several unfortunate consequences:
* The `KdbxReader` class had to import the `KdbxXmlWriter` class
in order to perform the export (bad separation of concerns);
* The CLI database unlocking logic had to be duplicated only
for the `Extract` command;
* The `xmlData` had to be stored in the `KeePass2Reader` as
a temporary result.
* Lots of `setSaveXml` functions were implemented only
to trickle down this functionality.
Also, the naming of the `saveXml` variable was not really
helpful to understand it's role.
Overall, this change will make it easier to maintain and expand
the CLI database unlocking logic (for example, adding a `--no-password`
option as requested in https://github.com/keepassxreboot/keepassxc/issues/1873)
It also opens to door to other types of extraction/exporting (for
example exporting to CSV, as requested in
https://github.com/keepassxreboot/keepassxc/issues/2572)
* CLI: fix missing check for correct credentials
Before this fix, most/all CLI commands had incorrect behaviour when bad
credentials were supplied: they would carry on regardless, with
potentially catastrophic results. In particular, the "add" subcommand
seemed to corrupt the database. "ls" would always report an empty
database. Haven't tested any others.
Also fixed a related missing check specific to the "merge" subcommand.
* Fixes including keepassxc-cli when building KeePassXC dmg on Mac.
resolves#1697
* Fix Qt search path and Properties display on macOS
* Simplify packaging POST_BUILD fixups
* Various fixes to get cli to run on macos
* Correct cli tests on macOS
* Several macOS related GUI test fixes
Many lines were not conformant with the project's formatting rules.
This patch should fix all formatting and whitespace issues in the code
base.
A clang-format directive was put around the connect() calls containing
SIGNALs and SLOTs whose signatures would be denormalized because of the
formatting rules.
The Database, DatabaseWidget, and DatabaseTabWidget classes share many responsibilities in inconsistent ways resulting in impenetrable and unmaintainable code and a diverse set of bugs and architecture restrictions. This patch reworks the architecture, responsibilities of, and dependencies between these classes.
The core changes are:
* Move loading and saving logic from widgets into the Database class
* Get rid of the DatabaseManagerStruct and move all the information contained in it into the Database
* Let database objects keep track of modifications and dirty/clean state instead of handing this to external widgets
* Move GUI interactions for loading and saving from the DatabaseTabWidget into the DatabaseWidget (resolves#2494 as a side-effect)
* Heavily clean up DatabaseTabWidget and degrade it to a slightly glorified QTabWidget
* Use QSharedPointers for all Database objects
* Remove the modifiedImmediate signal and replace it with a markAsModified() method
* Implement proper tabName() method instead of reading back titles from GUI widgets (resolves#1389 and its duplicates #2146#855)
* Fix unwanted AES-KDF downgrade if database uses Argon2 and has CustomData
* Improve code
This patch is also the first major step towards solving issues #476 and #2322.
* Add CLI commands show --totp and totp-clip for handling TOTPs, resolves#2429.
* Adding tests for new CLI TOTP commands
* Update keepassxc-cli man page.
This patch aims at reducing the number of copies for obejcts that could
be referenced rather than copied, because they're not modified during
the computation.
The arg() function of the QString class has a variable length argument
which allows to reduce the number of chained calls to the same function.
With proper formatting, readability is not affected.
Headers are not to be placed amongst the source files in the CMake
script. The preprocessor and the linker will take care of glue all the
files together. Also, the "include_directories()" statement at the top
of the file already tells CMake where to look for all the needed
header files.
QTextStream uses the system default locale, but this breaks in
various situations: (1) It does not work on the native Windows shell
(cmd.exe, Powershell), since the default Windows locale is Windows-1252,
but the shell uses Windows-850. (2) It also breaks on *nix systems where
the locale is Latin1 or C, which is the case for most CI systems or
build servers.
We allow overriding the detected codec by setting the ENCODING_OVERRIDE
environment variable, but otherwise prefer Windows-850 on Windows and
UTF-8 on any other system, even if LANG is set to something else.
This resolves#2413
The CLI module was lacking unit test coverage and showed some severe
coding style violations, which this patch addresses.
In addition, all uses of qCritical() with untranslatble raw char*
sequences were removed in favor of proper locale strings. These are
written to STDERR through QTextStreams and support output
redirection for testing purposes. With this change, error messages don't
depend on the global Qt logging settings and targets anymore and go
directly to the terminal or into a file if needed.
This patch also fixes a bug discovered during unit test development,
where the extract command would just dump the raw XML contents without
decrypting embedded Salsa20-protected values first, making the XML
export mostly useless, since passwords are scrambled.
Lastly, all CLI commands received a dedicated -h/--help option.
* Create history-based merging that keeps older data in history instead of discarding or deleting it
* Extract merge logic into the Merger class
* Allows special merge behavior
* Improve handling of deletion and changes on groups
* Enable basic change tracking while merging
* Prevent unintended timestamp changes while merging
* Handle differences in timestamp precision
* Introduce comparison operators to allow for more sophisticated comparisons (ignore special properties, ...)
* Introduce Clock class to handle datetime across the app
Merge Strategies:
* Default (use inherited/fallback method)
* Duplicate (duplicate conflicting nodes, apply all deletions)
* KeepLocal (use local values, but apply all deletions)
* KeepRemote (use remote values, but apply all deletions)
* KeepNewer (merge history only)
* Synchronize (merge history, newest value stays on top, apply all deletions)
This patch implements a new database wizard to guide users through the process
of setting up a new database and choosing sane encryption settings.
It also reimplements the master key settings to be more
user-friendly. Users can now add, change, or remove individual composite
key components instead of having to set all components at once. This
avoids confusion about a password being reset if the user only wants to
add a key file.
With these changes comes a major refactor of how database composite keys and key
components are handled. Copying of keys is prohibited and each key
exists only once in memory and is referenced via shared pointers. GUI
components for changing individual keys are encapsulated into separate
classes to be more reusable. The password edit and generator widgets
have also been refactored to be more reusable.
Note: This implementation is not yet connected to the
database itself and will corrupt existing kdbx3 db's.
* Implemented memory and parallelism parameters for Argon2Kdf
* Using libargon2; libsodium does not support Argon2d algorithm
* Moved basic rounds parameter into Kdf class
* Reimplemented benchmark algorithm; previous was utterly broken
* core: database: make UUID searching case-insensitive
4c4d8a5e84 ("Implement search for reference placeholder based on
fields other than ID") changed the semantics of searching-by-reference
in KeePassXC. Unforuntately it contained a bug where it implicitly
became case-sensitive to UUIDs, which broke existing databases that used
references (especially since the default reference format uses a
different case to the UUID used while searching).
The tests didn't catch this because ->toHex() preserves the case that it
was provided, they have been updated to check that UUIDs are case
insensitive.
* cli: show: resolve references in output
Previously, `keepassxc-cli show` would not resolve references. This
would make it quite hard to script around its output (since there's not
interface to resolve references manually either). Fix this by using
resolveMultiplePlaceholders as with all other users of ->password() and
related entry fields.
Fixes: keepassxreboot/keepassxc#1260
* tests: entry: add tests for ref-cloned entries
This ensures that the most "intuitive" current usage of references
(through the clone feature of the GUI) remains self-consistent and
always produces the correct results. In addition, explicitly test that
case insensitivity works as expected. These should avoid similar
regressions in reference handling in the future.
* http: resolve references in AccessControlDialog
The access control dialog previously would not show the "real" username
or "real" title when asking for permission to give access to entries.
Fix this by resolving it, as we do in many other places.
Fixes: keepassxreboot/keepassxc#1269
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
In order for scripting to be much simpler with `keepassxc-cli show`,
provide a simple --attributesk API which effectively is just a CLI
interface for entry->attributes()->value(...). This allows for more
extensibility and prevents changes in our output formatting from
breaking existing users of keepassxc-cli (if they use --attributes).
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
* Correct handling of --help and --version
* Moving arguments building up.
* Only manipulating argv is the command is valid.
* Not a failure when --help
* Not using showVersion()
