- fixed pagination when the last result set contains fewer rows than the
view can display and the user navigates back to the previous page.
and the user goes back to the previous batch of results.
- selectItem: don't try to select rows with empty data.
When selecting rows with the mouse, update the current index to the
first or last row of the selection range, depending on the selection
direction (up/down).
This allows keyboard-based selection expansion and shrinking to work
correctly without losing the existing selection.
- fixed navigating results with keys up/down and pageUp/Down.
- improved selecting/handling rows with shift + ctrl.
in both cases taking pagination into account.
Partially fixed a bug setting views limit, when the user had selected
several rows.
The issue occured in this scenario:
- the user selected several rows in a view.
- switched to a different view.
- returned to the previous view.
- the query was properly configured with the limits but returned all
results from the db.
- as a result, the scrollbar didn't respect the configured limit, and
instead, it was configured according to the total number of results
of the db.
Now if the number of results exceed the view/query limit and the query
limit is not 0 (show all results), we'll use the query limit as the
maximum number of rows of the view.
We should verify that the query returns the expected number of results.
Previously we loaded the ASN database unconditionally if the package
pyasn was installed, in order to lookup the network name associated with
an IP address.
As a result, the database was loaded into memory, even if you never
opened the Address tab, which we assume is not that common.
Now the DB will only be loaded when clicking on the Address tab, and
will be unloaded when closing the main window (loading the DB for the
first time takes a few milliseconds, or less).
This change reduces the memory usage by approximately 150MB of RAM for
most users that have the pyasn package installed.
Also:
- removed unused imports.
- imports reordered.
- fixed error that sent multicast addresses to analyze.
- new configuration options:
- allow to configure the plugin log level.
- allow to change icon if the outbound connection is malicious.
- allow to change the popup default action.
- use community score, reputation or suspicious option to
compute total malicious score.
- allow to define ips and domains exclusions.
Now row selection while pressing shift is not limited by the view
(query) limit. Users now can select rows across paginated results.
- fixed crash in some systems when paginating results (fillVisibleRows).
- now key up/down while pressing shift select rows.
- ensure that row numbers are refreshed when paginating results.
When a binary checksum mismatch occurs and you click on the checksums
warning:
- fixed error when updating a rule's checksum. Regardless of the
rule selected in the combobox, always the last one was updated.
- the last updated rule is now displayed first in the rules combobox.
In some cases, temporary rules were not being marked as disabled on the
database, making it difficult to see which temporary rules were active
or not (for example when the GUI was closed, a temporary rule expired in
the daemon, and the GUI was launched again).
Now when launching the GUI, we'll mark expired rules as not enabled,
including 'until-restart' rules.
When the daemon connects to the GUI, it'll send the list of active rules,
and the database will be updated accordingly.
This change will also reduce the number of rules displayed when
a binary checksum mismatch occurs (pop-ups, checksums). Sometimes when
the invalid checksum warning was displayed, we listed rules that
were no longer active.
the description for this rule, in the UI, vertically centers the 3 lines(2 text and one empty newline in the middle) thus showing only a few pixels of the first and last lines' text as the description for this rule making you believe there's something wrong with the font rendering. Removing one `\n` allows the two text lines to remain visible enough to be readable.
Loader.GetAll() returned the loader internal map by reference under an
RLock that was released as the call returned. getClientConfig() then
called GetAll() twice -- once for len() to size a slice, once to range
over the entries -- without holding any lock. A concurrent goroutine
(e.g. the loader still synchronously processing rules during
reloadConfiguration) writing to the map between the two calls makes the
iteration yield more entries than the pre-sized slice, panicking with
"index out of range [N] with length N". On bad luck the Go runtime
trips "concurrent map iteration and map write" instead.
The race is reachable on real systems whenever the UI poller, started
by reloadConfiguration via c.Connect(), reconnects before
c.rules.Reload(path) finishes loading rules from disk. Consistently
reproducible with ~2100 rules on disk.
Make GetAll() return a shallow snapshot of the rule map under the read
lock so length and iteration are guaranteed consistent for the caller.
Update the one caller (getClientConfig) to take the snapshot once.
Add a regression test running concurrent writers and readers against
the loader; -race catches the original bug, and the new GetAll
contract makes it pass.
* daemon-go: port layered list caching and matching strategies
* daemon-go: use immutable rule/list snapshots and add load benchmarks
* daemon-go: dropped gobwas, rule/list now use go filepath.Match for globbing
In ff05d94f41 python311-PyQt6 was added as
dependency, to help users install some needed dependencies automatically
on old versions of Tumbleweed.
However on new versions, it also install this package, which is not the
default python3 version. Besides, python3-qt6 exists as dependency on
latest versions.
So let's use again python3-qt6 for SuSe. If someone wants to install the
GUI on older versions of Tumbleweed (~2024-2025), they'll need to
install PyQt6 via pip.
Closes: #1576