This patch implements two new ZathuraDatabaseInterface functions, save_jumplist
and load_jumplist, for both the plain and sqlite backends (along with some
cleanups).
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
Since we are now saving the adjustments ratios in the jump structures, we need
to take care of the following scenario:
- We do an action that results in a new jump structure being added to the
jumplist while the inputbar is visible (e.g., search, jumping to a
specific page, jumping to a bookmark, or following a link).
- Since we are now storing the adjustments ratios in the jump structures,
all of the above actions would result in the vertical adjustment ratio
being saved while the inputbar and/or the completion menu is visible.
- Now we are exactly on the target of the jump (note that the inputbar and
completion menu now are hidden), so suppose that we want to go back using
^o (assuming that we didn't change the adjustments after jumping), then
the check at sc_jumplist that compares the current adjustments ratios
with that of the current jump (the jump that has just been added and
which we are currently on it's position) would fail, because after the
inputbar (with possibly the completion menu in case of bookmarks) is
activated it is hidden, which results in the vertical adjustment upper
bound to change, which in turn results in the vertical adjustment ratio
returned by zathura_adjustment_get_ratio to become different from what is
stored in the current jump structure, even though we haven't changed the
adjustments at all after the jump. This would always result in taking us
back to the exact position of the jump (which would be slightly different
from the current position) when we press ^o. This can be annoying,
because it would happen, for example, every time we need to go back
quickly after jumping to a link target, a search result, or a bookmark.
So, what this patch does is essentially to make the vertical adjustment ratio
reflecting the current vertical adjustment after a jump, to always be the same
as the one stored in the newly added jump structure, since both are calculated
with zathura_adjustment_get_ratio while the inputbar is _not_ visible, so they
should be the same.
I've elaborated just to make things clear, in case the purpose of the patch
isn't obvious.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This is just in case that zathura_link_evaluate returns without evaluating the
link, which would result in a new jump being added for the current position,
but without adding a corresponding jump for the target position (actually a
second one would be added but the check in zathura_jumplist_add would prevent
it from being appended on the linked list since it's position would be the same
as the current one, because the link haven't been evaluated).
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
Not that is really needed here since every argument is controlled by us. But
rather safe than sorry.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This is so that Zathura would compile without errors when enabling sqlite
support on systems that doesn't have sqlite compiled with
SQLITE_ENABLE_COLUMN_METADATA (which are the majority).
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This patch adds some enhancements/fixes to the bookmarking feature of Zathura:
- Bookmark the exact vertical and horizontal adjustments values, along with
the page number. This is done in a backward-compatible way for both the
plain and sqlite database backends, so that bookmarks that was taken
previously (bookmarking only the page number) will still work as expected
and won't be lost.
- Fix the issue of not being able to remove bookmarks from the plain
database; removing a bookmark in plain_remove_bookmark using
g_key_file_remove_key corrupts the bookmarks file. This is due to not
truncating the entire bookmarks file in zathura_db_write_key_file_to_file
prior to overriding it with a new one not containing the removed line.
This is why, I guess, plain_remove_bookmark hadn't been implemented as
expected using g_key_file_remove_key, because apparently, someone thought
that the problem is caused by this API.
- Fix not being able to update existing bookmarks persistently; updating a
bookmark works only during the current session, but after the file is
closed and reopened, the updated bookmark still has it's old value. This
is due to changing only the bookmark structure in memory; the proper way
to do it, is to call zathura_db_remove_bookmark on the old bookmark, then
follow that by a call to zathura_db_add_bookmark on a bookmark structure
containing the new position and the same old ID.
- Make zathura_bookmark_add updates the bookmark if it already exists,
rather than doing this externally in cmd_bookmark_create. This allows us
to have all the relevant girara_notify messages in cmd_bookmark_create.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
- Allow the user to always start with the very last (or first) jump on the
jumplist when pressing ^o (or ^i), while the jumplist iterator is
pointing to the last (or first) jump, and the document is not currently
at the position of this pointed-to jump. This is instead of jumping
directly to the previous/next jump, and skipping the last/first one.
- Don't jump to the current jump pointed to by the jumplist iterator, if
there is no more next jumps, and the jump direction is FORWARD.
- Don't jump to the current jump pointed to by the jumplist iterator, if
there is no more previous jumps, and the jump direction is BACKWARD.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
- Don't delete the elements on the right of the current one, when
appending a new jump to the jumplist, because this makes no sense at
all; the point of the jumplist in the first place is to remember
previously jumped-to positions in the document, so there is no need
to delete anythings except to trim the oldest entries from the
beginning to maintain the maximum size. This also makes us compatible
with the Vim way of doing things.
- Make the jumplist mechanism functional on the same page; if we
followed a link to a target on the same page, remember the
adjustments before and after following the link. The same holds for
navigating search results on the same page.
- Implement position_set and use it instead of position_set_delayed
when following links in order to give zathura_jumplist_save a chance
to record the exact adjustments of the link target. Otherwise, it
will always record the adjustments after going to the target page,
but before going to the exact position within it.
- Don't consider movements with ^i and ^o as jumps :)
- Don't use page_set followed by setting the adjustments in
sc_jumplist, because this is redundant and causes clutter when using
^i and ^o, as the adjustments is set twice this way (once in page_set
and again in position_set_delayed). It's enough to only update the
page number on the statusbar and then set the adjustments.
- Hide implementation details (zathura_jumplist_save and
zathura_jumplist_append), and make things more consistent by
exporting and using only zathura_jumplist_add for adding new entries.
The end result: A more slick jumping experience :-)
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This patch activates the last aborted search when pressing the search shortcuts
('n' or 'N').
To avoid confusion, and to make things more predictable, I've chosen to always
reactivate an aborted search starting from the beginning (or end, in case of
'N' or '?') of the current page, as opposed to Vim which continues from the
next search term each time the search is reactivated.
Searching using '/' or '?' doesn't center the view at the current search term
like when using 'n' or 'N', so we fix this here.
Also, I managed to work around the issue of the thin rectangular margins that
show around the previously-highlighted search terms after the search is aborted
(either explicitly or as a result of following links), by redrawing the page
widget (only if it's visible) instead of redrawing the rectangles covering the
highlighted search terms.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This uses the jumplist to implement a bisect functionality between the
last two jump points. ^j and ^k are bound to bisect forward and backward
respectively.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
Call cb_view_hadjustment_changed to set horizontal position in
navigation shortcuts: sc_navigate, sc_goto.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
Added config parameter "link-hadjust" with default value true. When set to false,
following internal links do not change the horizontal position of the page, only
the vertical position.
Also updates page number when following links, now.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This fixes the bug of the page cache acting weird (e.g. the same cache index is
invalidated over and over again) after being used for some time. This is
because when we read the last-view property, we only get the least-significant
32 bits of the last_view private member.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
The last_view member in zathura_page_widget_private_s is gint64, so we should
do the same here.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This will not scroll the document when pressing 'f' or 'F', just to make the
behavior consistent with activating the inputbar using ':'.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This also has the side effect of not flooding the console with debug messages
on every change of the vertical adjustment as a result of redundant cache
lookups.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
This patch implements a page cache that is invalidated in a LRU fashion.
Pages are added to the cache as soon as they become visible. When the cache is
full and a new page that isn't in the cache becomes visible, the least recently
viewed page in the cache is evicted from memory and the new one takes it's
place.
The cache size is configurable using the page-cache-size configuration
variable, with a default value of 15 pages. Very large values for the cache
size are not recommended, though, as it will stress the system memory out.
The old periodic page reclaiming code is no longer necessary with this patch,
so I removed it.
Special thanks to Ignas Anikevičius, and Sebastian Ramacher for the
inspirations.
Signed-off-by: Sebastian Ramacher <sebastian+dev@ramacher.at>
