Use type 25 on UDDF import when the gas mix contains some helium.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The wrong O2 percentage was due to not referring to the correct cylinder
but grabbing any o2 value in any cylinder. While fixing this, I noticed
that also end pressure was missing the cylinder selection, so fixing
that as well.
Fixes#626
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Most dive computers start always with the first cylinder (and if not,
they give a gas switch in the first few seconds). Trying to follow this
principle on divelogs.de export.
The bug 626 sounds like a bit different case so a sample XML log would
be needed to ensure that use case gets fixed.
See #626
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Gas changes generated by Xdeep can contain mix information with decimal
comma instead of decimal point. Thus we need to convert the comma to
point before performing mathematical operations on it.
Fixes#612
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch adds image support to UDDF export. As far as I can tell from
the UDDF specification, the images can only be added to notes field, so
that is what we do here. (buddy and owner can also have image tags, but
that is irrelevant for Subsurface currently.)
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This speeds up Aquadivelog imports marginally by doing lookup to used
equipments only instead of all equipment ever used.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Writing the samples/waypoints to UDDF export in a totally new way.
Previously the preparations for approximating waypoints was done for
every sample, now only for the events.
A few days ago it took 36 seconds for my test set of 8 dives to export
to UDDF. This optimization round took it down from 0m4.745s to
0m0.253s.
Fixes#508
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Duration is better term than time for the length of the dive. Also
remove the repetitive word dive from some of the header fields.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
UDDF export used all samples in the export when detecting previous
sample. This resulted in a very slow processing as we had to go through
massive amount of data for each event that's time was to be
approximated. The detection of previous sample was also erroneous
resulting in incorrect depth samples for the events in some occasions.
This patch should address these issues. And along with patch that
included pressure data on import this will fix the bug #499.
Well, the performance after the patch is still not stellar, but still
quite a difference with such a small change. The sample set of 8 dives
(with one event each) takes now less than 5 seconds instead of the
original 36 seconds (measured by doing the conversion with xsltproc).
Fixes#499
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Importing pressure samples on UDDF import was missing. This patch adds
that bit of information to our import.
See #499
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Let's make the export consistent by quoting the header line the same way
the sample rows are quoted.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Crafting depth samples produced a negative depth for the last sample.
This patch checks that negative last sample is ignored. (This occurred
when last manual sample is not full minutes.)
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch creates artificial depth samples for manually entered dives
every 60 seconds. This is used for export to divelogs.de.
Fixes#486
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Sensus dataloggers report depth in (absolute) millibar. This simple
fix corrects the CSV import. The depth is now computed correctly.
Signed-off-by: Jan Mulder <jlmulder@planet.nl>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This implements divelog export into CSV format. This currently supports
only metric units and lacks user interface.
See #434
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Empty values included the rest of the line in the value, instead of
leaving it empty. This patch leaves empty values empty.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Sensus divides temperature in one or two fields (depending on locale),
so it cannot be parsed with our generic CSV import XSLT. And e.g. depth
is in millimeters.
This supports only importing of single dive at a time. Multiple dives
would be in same file with first column being dive number.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We want to skip duplicate lines on CSV import. For this we need to use
time field, that is not necessarily the first field of the input. This
patch uses the configured time field instead of first one.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Nowadays cylinder tags are generated based on actual usage. This
includes all the required information on cylinders. Therefore generating
incomplete cylinder tags from all listed cylinders is not needed
anymore.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch adds support for cylinders for UDDF version 3.2.1.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Adding support for multiple cylinders on UDDF export. This also adds a
cylinder reference to the tankdata to enable mix lookups.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch generates a waypoint for events occurring between profile
samples. It approximates the depth of the crafted waypoint linearly
between the recorded sample depths.
Fixes#385
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
UDDF export saved events only if they occurred at exactly same time as a
profile sample. This patch will include events between samples as well.
This assumes that profile samples are recorded in even intervals.
Fixes#385
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Seems there is a different way to state O2 percentage in DivingLog.
This patch includes that format along with the previously seen one.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Seems that we can detect the unit of samples from DivingLog import when
comparing the maximum dive depth to the maximum recorded sample depth.
If sample is bigger, it is recorded in imperial. Of course this relies
on the assumption that the max depth of dive is recorded in metric unit,
but that seems to be the case, at least in the sample received.
It really does not make any sense to have both units in use for one dive
without any info on the unit used. But since that is the case, let's try
to cope with it.
Fixes#316
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
After change to include XSLT files in resources, the import quit
working (even though the files and templates were found properly). This
patch uses include statement instead as it seems to work in our case.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The separator selector in the CSV import dialog was unused. This passes
the value into the xslt and adds ',' as possible value.
I'm sure this could be done much better (pass the actual character instead
of the index), but I couldn't get that to work and this does seem to do
the trick.
Also added a test dive to test this feature.
Fixes#321
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Use if around the attribute, instead if a choose inside. Its less
horrible on the eyes.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I managed to mess up and grab parts of my unfinished otu field code and
squash it into the wrong patch. Here is the code for stopdepth, and now
the code sets the in_deco flag to.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This adds a checkbox to be able to import csv files without temperature.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Basic information like buddy, location and visibility is included in
UDDF export. Also attempt to grab temperature information from both
temperature and divetemperature tags.
This partially fixes bug #224. However, at least divemaster and tags are
not exported (as the UDDF spec either does not have these or has hidden
the information of these fields somewhere between so many seemingly
unnecessary other fields).
See #224
Signed-off-by: Miika Turkia <miika.turkia@nixu.fi>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Some improvements on importing different versions of UDDF logs. Also
include rating, gps coordinates and gas mixture in the import.
Signed-off-by: Miika Turkia <miika.turkia@nixu.fi>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since CSV import does not include date/time stamp, we need to generate
one for ourselves. This patch uses current time of the import as dive
time.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
As some CSV log files seem to contain multiple sample lines for a single
time, we'll just grab the first one of them to avoid duplicates.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Field variables for the CSV XSLT import have disappeared at some point
during developing GUI for the CSV import. So adding them to the XSLT for
the field selections to have effect.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When we convert time in seconds to mm:ss format, we do not want to round
but get floor instead.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
