Oops. Moving the check to the beginning of the function in order to avoid
partial execution in commit 69036a1bb7 ("Avoid resource leak by bailing
early") had a nasty side effect. Since Linus used 'i' both to hold the
dive number and as a loop variable, by moving this to the top of the
function the dive nr was overwritten which caused all kinds of problems.
My bad - but of course you shouldn't mix loop variables with meaningful
variables, either...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
While in the current use this won't happen, if someone were to call
split_dive_at with a dive that's not in the dive_table, let's bail right
away before doing any work.
Coverity CID 1325517 1325518
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
While the visual state didn't show it, our internal tracking of the
selected state was copied causing all kinds of unexpected behavior. With
this commit we get this right.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This case should be impossible to hit - but this seems better than
assuming this can never happen.
Coverity CID 1325458
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Right now this requires that
(a) the dive have only one divecomputer associated with it.
Trying to split a dive with multiple dive computers would be *much*
harder to do, since you'd have to try to line up the surface
interval between computers etc. So just don't do it after
downloading multiple dive computers for the same dive.
(b) there must be at least one minute between the sample that came up
to the surface and the sample that goes down again.
If you just peeked your head above the surface, don't try to split
things into two dives. Maybe we can relax this for freediving or
something.
also note that the split dive will only get new numbering if the dive
that was split was the very last dive in the divelist.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If the user added a dive manually we have a hard time trusting the start
time and duration. We just shouldn't automatically merge them with
anything else.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
pressure
At the beginning of the dive, let the cylinder cool down if the diver
starts off at the surface. And at the end of the dive, there may be
surface pressures where the diver has already turned off the air supply
(especially for computers like the Uemis Zurich that end up saving quite
a bit of samples after the dive has ended).
So just don't use surface samples to populate the beginning/ending
cylinder pressure data.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch changes the dive merging to be much more careful about
things, because it turns out that we had several small oddities that
caused big merge issues.
The oddities are:
- the dive "duration" is actually how long we spend under water.
But that means that when we do "dive->when + dive.duration.seconds"
to calculate the end of the dive, that is nonsensical if you came up
to the surface in the middle of a dive.
Now, normally you don't see profiles like that, but once you start
merging dives together, it can go from "small detail" to "dominant
factor".
- We have two different cases of merging: the automatic "merge new dive
computer download if it looks like the same dive" (which always has a
merge offset of 0, since we merge it as a new dive computer) and the
"merge two different dives into one longer dive.
The code assumed that it could look at the "downloaded" flag for the
dive to check one or the other, but that doesn't really work.
Reading a dive from an XML file isn't any different from downloading
it.
So we need to change the logic to determine what kind of merge it is
to actually check the passed-in time offset.
With this, Stuart Vernon's test-case of eight dives with short surface
intervals in between end up merging correctly into one dive.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Reported-by: Stuart Vernon <stuartv@force2.net>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Up to now, we only computed hashes of images upon actually displaying the images.
With this patch we start to compute hashes once we load the xml or from git. This
happens in the background, so the user should note an increased CPU load only once
per divelog.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Give the user the possibility to attach images to a dive even
when the times do not match
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Having random uuids seemed like a good idea, but there are several
situations where they really cause problems. One is merging dive file
imports from V2 logfiles. Another is testing such imports.
Instead of making the uuid random we now hash the name and add the
timestamp of the first dive associated with this dive site to the hash
(first in this context is "first encountered" with no guarantee that it is
the chronologically first). This way V2 imports create deterministic uuids
but uuid conflicts are still extremely unlikely, even if the user has
multiple dive sites with the same name.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This sets up a standard dive scenario (30 minutes at 260ft/79m, EAN36 and
Oxygen as deco gases, last stop at 20ft/6m) and calls the planner to set
up a dive plan given certain standard gases.
Instead of trying to verify the complete plans it checks that we switch to
the deco gases at the right depth and the complete duration of the dive
matches our expectation.
The test intentionally fails right now for imperial as we have the wrong
switch depth for Oxygen. See how useful tests are?
On the downside, the test does NOT produce the same plan as Subsurface
when I try to create a consistent setup for both - and I have not been
able to figure out why. There must be some other parameters that I'm not
setting, but I haven't identified them, yet. It's very small differences,
for example in the metric case the stops at 21m, 9m, and 6m are each one
minute shorter in the test than it what Subsurface calculates.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The longer I stared at the existing code the less it made sense.
So instead I rewrote it in a way that seems logical to me. And added a
boatload of debugging output (which needs to be removed, of course).
I tested this against more than a hundred dives and it seemed to always
pick the right fix.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Neither of these functions should ever be called with dc == NULL.
But it's easy to prevent the potential crash.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is re-run every time we update the dive list model for the UI, but we
should run it a first time right after loading the dives so that this data
is correct even without the UI running - this way the data in headless
standalone applications like the HTML exporter is correct as well.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Save and load a usually unused copy of the preferences with the units that
were active the last time the dive list was saved to git storage (this
isn't used in XML files); storing the unit preferences in the data file is
usually pointless (that's a setting of the software, not a property of the
data), but it's a great hint of what the user might expect to see when
creating a backend service that visualizes the dive list without
Subsurface running - so this is basically a functionality for the core
library that Subsurface itself doesn't use but that another consumer of
the library (like an HTML exporter) will need.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
taglist_added() simply figures out the tags that are in the new list but
not in the original list.
taglist_dump() makes debugging things easier.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This adds a new divelist context menu entry which asks for a URL. The file
is retrieved and if it is an image it is added to the cache and the url
is associated to dives as with local files.
NB this currently only works with URLs pointing directly to images. But it
should not be too hard to add the possibility to add a direction via an html
file and its image tags.
To test: open dives/test43.xml and delete the image and then add the URL
http://euve10195.vserver.de/~robert/wreck.jpg
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Factor out image load to find timestamp from loop over dives.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Use more explicit variabel names and make the get timestamp function actually
return the timestamp rather than getting a pointer argument
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When we delete a cylinder the gas changes in that dive may have to be
adjusted. We didn't do this at all in the past. With this commit we should
be doing this right for a single dive that is being edited.
This does NOT handle multiple dives being edited at the same time (or more
specifically - if you have multiple dives selected and delete a cylinder,
the dives that had the same set of cylinders (other than the displayed
dive) will get that particular cylinder deleted, but won't have their gas
change events (and sensor data in the samples) adapted.
Possibly we should simply prohibit deleting cylinders when more than one
dive are selected.
See #834
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Segfault did appear when picture had some gps coordinates but
there was no dive site assigned to the dive.
Signed-off-by: Jan Darowski <jan.darowski@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Seperated getting image timestamp from picture_load_exif_data() and
ShiftImageTimesDialog::syncCameraClicked() into picture_get_timestamp()
and seperated checking timestamp from dive_create_picture() to
dive_check_picture_time().
Signed-off-by: Jan Darowski <jan.darowski@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Sequentially parses a file, expected to be a Datatrak/WLog divelog, and
converts the dive info into Subsurface's dive structure.
As my first DC, back in 90s, was an Aladin Air X, the obvious choice of log
software was DTrak (Win version). After using it for some time we moved to WLog
(shareware software more user friendly than Dtrak, printing capable, and still
better, it runs under wine, which, as linux user, was definitive for me). Then,
some years later, my last Aladin died and I moved to an OSTC, forcing me to
look for a software that support this DC.
I found JDivelog which was capable of import Dtrak logs and used it for some
time until discovered Subsurface existence and devoted to it.
The fact was that importing Dtrak dives in JDivelog and then re-importing them
in Subsurface caused a significant data loss (mainly in the profile events and
alarms) and weird location of some other info in the dive notes (mostly tag
items in the original Dtrak software). This situation can't actually be solved
with tools like divelogs.de which causes similar if no greater data loss.
Although this won't be a core feature for Subsurface, I expect it can be useful
for some other divers as has been for me.
Comments and issues:
Datatrak/Wlog files include a lot of diving data which are not directly
supported in Subsurface, in these cases we choose mostly to use "tags".
The lack of some important info in Datatrak archives (e.g. tank's initial
pressure) forces us to do some arbitrary assumptions (e.g. initial pressure =
200 bar).
There might be archives coming directly from old DOS days, as first versions
of Datatrak run on that OS; they were coded CP437 or CP850, while dive logs
coming from Win versions seems to be coded CP1252. Finally, Wlog seems to use a
mixed confusing style. Program directly converts some of the old encoded chars
to iso8859 but is expected there be some issues with non alphabetic chars, e.g.
"ª".
There are two text fields: "Other activities" and "Dive notes", both limited to
256 char size. We have merged them in Subsurface's "Dive Notes" although the
first one could be "tagged", but we're unsure that the user had filled it in
a tag friendly way.
WLog adds some information to the dive and lets the user to write more than
256 chars notes. This is achieved, while keeping compatibility with DTrak
divelogs, by adding a complementary file named equally as the .log file and
with .add extension where all this info is stored. We have, still, not worked
with this complementary files.
This work is based on the paper referenced in butracker #194 which has some
errors (e.g. beginning of log and beginning of dive are changed) and a lot of
bytes of unknown meaning. Example.log shows, at least, one more byte than those
referred in the paper for the O2 Aladin computer, this could be a byte referred
to the use of SCR but the lack of an OC dive with O2 computer makes impossible
for us to compare.
The only way we have figured out to distinguish a priori between SCR and non
SCR dives with O2 computers is that the dives are tagged with a "rebreather"
tag. Obviously this is not a very trusty way of doing things. In SCR dives,
the O2% in mix means, probably, the maximum O2% in the circuit, not the O2%
of the EAN mix in the tanks, which would be unknown in this case.
The list of DCs related in bug #194 paper seems incomplete, we have added
one or two from WLog and discarded those which are known to exist but whose
model is unknown, grouping them under the imaginative name of "unknown". The
list can easily be increased in the future if we ever know the models
identifiers.
BTW, in Example.log, 0x00 identifier is used for some DC dives and from my own
divelogs is inferred that 0x00 is used for manually entered dives, this could
easily be an error in Example.log coming from a preproduction DC model.
Example.log which is shipped in datatrak package is included in dives
directory for testing pourposes.
[Dirk Hohndel: some small cleanups, merged with latest master, support
divesites, remove the pointless memset() before free() calls
add to cmake build]
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Upon successfull reading an image file, this computes a SHA1 hash of the
image and saves it with the picture tag in the log file. When a file is
not successfully loaded (for example because the log was created on a
different computer) we look up the hash in a dictionary that maps hashes
to local file names.
That dictionary (actually two for both directions), is loaded on startup
and saved upon destruction of the main window.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
And hook things up when double clicking the globe.
The user experience isn't consistent with what we do on the main tab
(i.e., no coloring of fields that are changed), but it seems to work.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch adds tests if the computer is actually a Predator and
if the value is close to what we would have computed anyway.
[Dirk Hohndel: minor change to use same_string() ]
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If current_dive and displayed_dive still have the same tags then we
shouldn't touch the tags of other selected dives.
Fixes#826
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
For some dive computers (at least the Shearwater Predator, I haven't
checked if there are others), libdivecomputer used to return setpoint
values in each sample even if the dive computer was in OC mode. Those
setpoint values are redundant and confuse our algorithm that tries to
detect if a dive is OC or CCR. So when manually switching from CCR to
OC we make sure that there are no setpoint values in the samples.
This is a destructive change - if the user switches to OC by mistake and
accepts that change, even when switching back to CCR the setpoint changes
during the dive are lost.
I rewrote the code dealing with the events as it was rather confused.
Looping over the events that way didn't make any sense since
get_next_event() is guaranteed to give you the first (if any) event of the
requested name.
See #826
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When one has 2 pictures with the same timestamp, these pictures are
reordered on saving (requires something changed on the specific dive for
the save to actually trigger). This patch keeps the picture order intact
in the XML log.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
No oxygen at all is not quite enough for the average diver but this prevents
inifintie loops/extremely long deco when subsurface is used in pscr mode
with unrealistic values for pscr preferences.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Reported-by: Anton "glance" Lundin <glance@acc.umu.se>
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The ratio between SAC and oxygen metabolism rate can be assumed constant
but not the metabolism rate. So we better base our calculation on the ratio
that uses the SAC from the preferences as that pairs well with the O2
consumption from the preferences.
Hence we ran remove the sac parameter from fill_pressures().
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When changing to a CCR dive, add a setpoint change to the default setpoint
at the beginning of the dive. Otherwise add an explicit setpoint change to 0
.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Calculations for passive semi-closed rebreathers are pretty much like OC except
the pO2 is lower bey a certain (SAC dependent) factor. This patch introduces the
corresponding calculations in case dctype == PSCR which is so far never set and
there is currently no UI for these calculations. As pO2 is SAC dependent it takes
a certain attempt at getting it and drops to defaults from the prefs otherwise.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
A value of zero (which is the normal legacy one) remains "unknown", but
the divecomputer backend can now give both gasmix and cylinder number
this way.
Currently only the EON Steel backend does that, but it should be easy
enough to extend others too.
Also, fix the user-visible cylinder numbering in the cylinder change
tooltip to use a human-friendlier one-based numbering (ie first cylinder
is "cyl 1", not "cyl 0")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Commit 6cf3787a0e ("Remove code that zeroes out duplicate oxygen sensor
and temperature values") incorrectly removes the code that zeroes out the
temperatures.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This ensures that the list of configured cylinders is consecutive.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
No, this causes double free problems.
And they show up in the weirdest ways, e.g. when just closing your data
file. Or when making changes and re-saving a modified dive.
Fixes#794
This reverts commit 0bdd8f41e0.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When we load multiple files, we call set_userid multiple times. Then we
need to free the old user id before we set a new.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The previous taglist_free didn't free the divetag, it only freed the
tag.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
