While it's not the best solution, it should cover most use cases where a
custom format on the progress value is required. If any further
customization is needed, the only current alternative is to use rules
to override the format.
Closes#273
Displaying too heavy notifications can DoS dunst. For example bad
programs, which pipe raw image data into the notification.
Limiting the maximum character length to 5000 circumvents this.
5000 should be ridiculously high to prevent DoS while still not
truncating all correct notifications.
A middle click on an notification with a single or default action will
invoke it. If there are multiple actions and no default, the context
menu is opened. If there are no actions, proceed similarly with URLs.
x.c has grown into an unmaintainable blob. Moving it in its own
subdirectory is the first step to splitting it into submodules each of
which will have a well defined job so that a lot of the code can be
abstracted to improve readability and make it much easier to maintain.
This process will be done one small part at a time only when we need to
change significantly change a part of the file, there is no need to
break what already works.
It makes more sense to handle escapes in configuration strings when we
parse them, rather than after-the-fact when we handle notifications that
use the relevant config options.
Do this in the proper backslash handler; rather than a naive string
replace which doesn't allow the backslash to be escaped (`\\n` => 0x5C0x0A
rather than 0x5C0x6E).
it to the proper backslash handler
Merge the allow_markup and plain_text settings into a single setting.
These 2 settings had a similar function, allow_markup controlled whether
markup was parsed or stripped and plain_text whether the notification
was escaped and displayed as is.
To cover all the possible combinations of the settings mentioned above
`markup` can take the following values:
full: The equivalent of allow_markup yes, plain_text no.
Passes the text straight to pango with minimal parsing. All valid
pango tags will be parsed.
strip: The equivalent of allow_markup no, plain_text no.
Strips the markup using string_strip_delimited. The parsing is
simplistic and if there are any unescaped '<' and/or '>'
characters it might get tripped and strip out actual text.
According to the GNOME notification specification, if a server
doesn't support markup(and we don't advertise that we do if it is
turned off) it should be stripped clientside, so this setting
should rarely be used. It is mainly left in for compatibility
with broken clients that don't follow the specification.
no: The equivalent of allow_markup [yes/no](any value), plain_text
yes.
Makes the notification content be rendered as plain text
regardless if it contains markup. Any Markup will be shown as
regular text.
Markup inside 'format' will still be parsed regardless of what markup is
set to.
Closes#279
If a notification has a raw icon, icon is set to NULL which was passed
without checking to strcmp. We don't (yet) support comparing raw icons so
if a raw icon is set, we can safely assume the notification is not a
duplicate.
In my opinion, it's better to crash early when something is wrong than
to pretend it never happened. This change will allow us to catch more
in the long run.
notification_create currently simply allocates memory and zeroes it to
properly initialise a notification. Since we currently have at least 2
places in the code that create notifications(startup notification in
dunst.c and notifications from dbus in dbus.c) the creation should be
handled from a central location in case we ever need to change the
creation procedure later on.
Expand the duplication checking functionality to also check for urgency
and icons. If icons are turned off in the settings the icon check is
skipped.
Icons matching is done based on the given path, if the path differs or
an icon name is used for one notification but an absolute path for
another, the icons are considered different even if they refer to the
same file.
