subsurface/planner.c

1366 lines
38 KiB
C
Raw Normal View History

First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
/* planner.c
*
* code that allows us to plan future dives
*
* (c) Dirk Hohndel 2013
*/
#include <libintl.h>
#include <glib/gi18n.h>
#include <unistd.h>
#include <ctype.h>
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
#include "dive.h"
#include "divelist.h"
#include "display-gtk.h"
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
int decostoplevels[] = { 0, 3000, 6000, 9000, 12000, 15000, 18000, 21000, 24000, 27000,
30000, 33000, 36000, 39000, 42000, 45000, 48000, 51000, 54000, 57000,
60000, 63000, 66000, 69000, 72000, 75000, 78000, 81000, 84000, 87000,
90000, 100000, 110000, 120000, 130000, 140000, 150000, 160000, 170000,
180000, 190000, 200000, 220000, 240000, 260000, 280000, 300000,
320000, 340000, 360000, 380000
};
double plangflow, plangfhigh;
char *disclaimer;
GtkWidget *planner, *planner_error_bar, *error_label;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
#if DEBUG_PLAN
void dump_plan(struct diveplan *diveplan)
{
struct divedatapoint *dp;
struct tm tm;
if (!diveplan) {
printf ("Diveplan NULL\n");
return;
}
utc_mkdate(diveplan->when, &tm);
printf("\nDiveplan @ %04d-%02d-%02d %02d:%02d:%02d (surfpres %dmbar):\n",
tm.tm_year+1900, tm.tm_mon+1, tm.tm_mday,
tm.tm_hour, tm.tm_min, tm.tm_sec,
diveplan->surface_pressure);
dp = diveplan->dp;
while (dp) {
printf("\t%3u:%02u: %dmm gas: %d o2 %d h2\n", FRACTION(dp->time, 60), dp->depth, dp->o2, dp->he);
dp = dp->next;
}
}
#endif
static void on_error_bar_response(GtkWidget *widget, gint response, gpointer data)
{
if (response == GTK_RESPONSE_OK)
{
gtk_widget_destroy(widget);
planner_error_bar = NULL;
error_label = NULL;
}
}
static void show_error(const char *fmt, ...)
{
va_list args;
GError *error;
GtkWidget *box, *container;
gboolean bar_is_visible = TRUE;
va_start(args, fmt);
error = g_error_new_valist(g_quark_from_string("subsurface"), DIVE_ERROR_PLAN, fmt, args);
va_end(args);
if (!planner_error_bar) {
planner_error_bar = gtk_info_bar_new_with_buttons(GTK_STOCK_OK, GTK_RESPONSE_OK, NULL);
g_signal_connect(planner_error_bar, "response", G_CALLBACK(on_error_bar_response), NULL);
gtk_info_bar_set_message_type(GTK_INFO_BAR(planner_error_bar), GTK_MESSAGE_ERROR);
bar_is_visible = FALSE;
}
container = gtk_info_bar_get_content_area(GTK_INFO_BAR(planner_error_bar));
if (error_label)
gtk_container_remove(GTK_CONTAINER(container), error_label);
error_label = gtk_label_new(error->message);
gtk_container_add(GTK_CONTAINER(container), error_label);
box = gtk_dialog_get_content_area(GTK_DIALOG(planner));
if (!bar_is_visible)
gtk_box_pack_start(GTK_BOX(box), planner_error_bar, FALSE, FALSE, 0);
gtk_widget_show_all(box);
/* make sure this actually gets shown BEFORE the calculations run */
while (gtk_events_pending())
gtk_main_iteration_do(FALSE);
}
void get_gas_from_events(struct divecomputer *dc, int time, int *o2, int *he)
{
struct event *event = dc->events;
while (event && event->time.seconds <= time) {
if (!strcmp(event->name, "gaschange")) {
*o2 = 10 * event->value & 0xffff;
*he = 10 * event->value >> 16;
}
event = event->next;
}
}
/* simple helper function to compare two permille values with
* (rounded) percent granularity */
static inline gboolean match_percent(int a, int b)
{
return (a + 5) / 10 == (b + 5) / 10;
}
static int get_gasidx(struct dive *dive, int o2, int he)
{
int gasidx = -1;
/* we treat air as 0/0 because it is special */
if (is_air(o2, he))
o2 = 0;
while (++gasidx < MAX_CYLINDERS)
if (match_percent(dive->cylinder[gasidx].gasmix.o2.permille, o2) &&
match_percent(dive->cylinder[gasidx].gasmix.he.permille, he))
return gasidx;
return -1;
}
static void get_gas_string(int o2, int he, char *text, int len)
{
if (is_air(o2, he))
snprintf(text, len, _("air"));
else if (he == 0)
snprintf(text, len, _("EAN%d"), (o2 + 5) / 10);
else
snprintf(text, len, "(%d/%d)", (o2 + 5) / 10, (he + 5) / 10);
}
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
/* returns the tissue tolerance at the end of this (partial) dive */
double tissue_at_end(struct dive *dive, char **cached_datap)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
{
struct divecomputer *dc;
struct sample *sample, *psample;
int i, j, t0, t1, gasidx, lastdepth;
int o2, he;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
double tissue_tolerance;
if (!dive)
return 0.0;
if (*cached_datap) {
tissue_tolerance = restore_deco_state(*cached_datap);
} else {
tissue_tolerance = init_decompression(dive);
cache_deco_state(tissue_tolerance, cached_datap);
}
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dc = &dive->dc;
if (!dc->samples)
return tissue_tolerance;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
psample = sample = dc->sample;
lastdepth = t0 = 0;
/* we always start with gas 0 (unless an event tells us otherwise) */
o2 = dive->cylinder[0].gasmix.o2.permille;
he = dive->cylinder[0].gasmix.he.permille;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
for (i = 0; i < dc->samples; i++, sample++) {
t1 = sample->time.seconds;
get_gas_from_events(&dive->dc, t0, &o2, &he);
if ((gasidx = get_gasidx(dive, o2, he)) == -1) {
show_error(_("Can't find gas %d/%d"), (o2 + 5) / 10, (he + 5) / 10);
gasidx = 0;
}
if (i > 0)
lastdepth = psample->depth.mm;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
for (j = t0; j < t1; j++) {
int depth = interpolate(lastdepth, sample->depth.mm, j - t0, t1 - t0);
tissue_tolerance = add_segment(depth_to_mbar(depth, dive) / 1000.0,
&dive->cylinder[gasidx].gasmix, 1, sample->po2, dive);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
}
psample = sample;
t0 = t1;
}
return tissue_tolerance;
}
/* how many seconds until we can ascend to the next stop? */
int time_at_last_depth(struct dive *dive, int o2, int he, int next_stop, char **cached_data_p)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
{
int depth, gasidx;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
double surface_pressure, tissue_tolerance;
int wait = 0;
struct sample *sample;
if (!dive)
return 0;
surface_pressure = dive->dc.surface_pressure.mbar / 1000.0;
tissue_tolerance = tissue_at_end(dive, cached_data_p);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
sample = &dive->dc.sample[dive->dc.samples - 1];
depth = sample->depth.mm;
gasidx = get_gasidx(dive, o2, he);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
while (deco_allowed_depth(tissue_tolerance, surface_pressure, dive, 1) > next_stop) {
wait++;
tissue_tolerance = add_segment(depth_to_mbar(depth, dive) / 1000.0,
&dive->cylinder[gasidx].gasmix, 1, sample->po2, dive);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
}
return wait;
}
int add_gas(struct dive *dive, int o2, int he)
{
int i;
struct gasmix *mix;
cylinder_t *cyl;
for (i = 0; i < MAX_CYLINDERS; i++) {
cyl = dive->cylinder + i;
mix = &cyl->gasmix;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
if (cylinder_nodata(cyl))
break;
if (match_percent(o2, mix->o2.permille) && match_percent(he, mix->he.permille))
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
return i;
}
if (i == MAX_CYLINDERS) {
show_error(_("Too many gas mixes"));
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
return -1;
}
mix->o2.permille = o2;
mix->he.permille = he;
/* since air is stored as 0/0 we need to set a name or an air cylinder
* would be seen as unset (by cylinder_nodata()) */
cyl->type.description = strdup(_("Cylinder for planning"));
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
return i;
}
struct dive *create_dive_from_plan(struct diveplan *diveplan)
{
struct dive *dive;
struct divedatapoint *dp;
struct divecomputer *dc;
struct sample *sample;
int oldo2 = O2_IN_AIR, oldhe = 0;
int oldpo2 = 0;
int lasttime = 0;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
if (!diveplan || !diveplan->dp)
return NULL;
#if DEBUG_PLAN & 4
printf("in create_dive_from_plan\n");
dump_plan(diveplan);
#endif
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dive = alloc_dive();
dive->when = diveplan->when;
dive->dc.surface_pressure.mbar = diveplan->surface_pressure;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dc = &dive->dc;
dc->model = strdup(_("Simulated Dive"));
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dp = diveplan->dp;
/* let's start with the gas given on the first segment */
if (dp->o2 || dp->he) {
oldo2 = dp->o2;
oldhe = dp->he;
}
sample = prepare_sample(dc);
sample->po2 = dp->po2;
finish_sample(dc);
add_gas(dive, oldo2, oldhe);
while (dp) {
int o2 = dp->o2, he = dp->he;
int po2 = dp->po2;
int time = dp->time;
int depth = dp->depth;
if (time == 0) {
/* special entries that just inform the algorithm about
* additional gases that are available */
add_gas(dive, o2, he);
dp = dp->next;
continue;
}
if (!o2 && !he) {
o2 = oldo2;
he = oldhe;
}
/* Check for SetPoint change */
if (oldpo2 != po2) {
if (lasttime)
add_event(dc, lasttime, 20, 0, po2, "SP change"); // SAMPLE_EVENT_PO2
oldpo2 = po2;
}
/* Create new gas, and gas change event if necessary;
* Sadly, we inherited our gaschange event from libdivecomputer which only
* support percentage values, so round the entries */
if (o2 != oldo2 || he != oldhe) {
int plano2 = (o2 + 5) / 10 * 10;
int planhe = (he + 5) / 10 * 10;
int value;
add_gas(dive, plano2, planhe);
value = (plano2 / 10) | ((planhe / 10) << 16);
add_event(dc, lasttime, 25, 0, value, "gaschange"); // SAMPLE_EVENT_GASCHANGE2
oldo2 = o2; oldhe = he;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
}
/* Create sample */
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
sample = prepare_sample(dc);
/* set po2 at beginning of this segment */
/* and keep it valid for last sample - where it likely doesn't matter */
sample[-1].po2 = po2;
sample->po2 = po2;
sample->time.seconds = time;
sample->depth.mm = depth;
finish_sample(dc);
lasttime = time;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dp = dp->next;
}
if (dc->samples <= 1) {
/* not enough there yet to create a dive - most likely the first time is missing */
free(dive);
dive = NULL;
}
#if DEBUG_PLAN & 32
if (dive)
save_dive(stdout, dive);
#endif
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
return dive;
}
void free_dps(struct divedatapoint *dp)
{
while (dp) {
struct divedatapoint *ndp = dp->next;
free(dp);
dp = ndp;
}
}
struct divedatapoint *create_dp(int time_incr, int depth, int o2, int he, int po2)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
{
struct divedatapoint *dp;
dp = malloc(sizeof(struct divedatapoint));
dp->time = time_incr;
dp->depth = depth;
dp->o2 = o2;
dp->he = he;
dp->po2 = po2;
dp->entered = FALSE;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
dp->next = NULL;
return dp;
}
struct divedatapoint *get_nth_dp(struct diveplan *diveplan, int idx)
{
struct divedatapoint **ldpp, *dp = diveplan->dp;
int i = 0;
ldpp = &diveplan->dp;
while (dp && i++ < idx) {
ldpp = &dp->next;
dp = dp->next;
}
while (i++ <= idx) {
*ldpp = dp = create_dp(0, 0, 0, 0, 0);
ldpp = &((*ldpp)->next);
}
return dp;
}
void add_duration_to_nth_dp(struct diveplan *diveplan, int idx, int duration, gboolean is_rel)
{
struct divedatapoint *pdp, *dp = get_nth_dp(diveplan, idx);
if (idx > 0) {
pdp = get_nth_dp(diveplan, idx - 1);
if (duration && (is_rel || duration <= pdp->time))
duration += pdp->time;
}
dp->time = duration;
if (duration > 180 * 60)
show_error(_("Warning - extremely long dives can cause long calculation time"));
}
/* this function is ONLY called from the dialog callback - so it
* marks this entry as 'entered'.
* Do NOT call from other parts of the planning code without changing
* that logic */
void add_depth_to_nth_dp(struct diveplan *diveplan, int idx, int depth)
{
struct divedatapoint *dp = get_nth_dp(diveplan, idx);
dp->depth = depth;
dp->entered = TRUE;
}
void add_gas_to_nth_dp(struct diveplan *diveplan, int idx, int o2, int he)
{
struct divedatapoint *dp = get_nth_dp(diveplan, idx);
dp->o2 = o2;
dp->he = he;
}
void add_po2_to_nth_dp(struct diveplan *diveplan, int idx, int po2)
{
struct divedatapoint *dp = get_nth_dp(diveplan, idx);
dp->po2 = po2;
}
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
void add_to_end_of_diveplan(struct diveplan *diveplan, struct divedatapoint *dp)
{
struct divedatapoint **lastdp = &diveplan->dp;
struct divedatapoint *ldp = *lastdp;
int lasttime = 0;
while (*lastdp) {
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
ldp = *lastdp;
if (ldp->time > lasttime)
lasttime = ldp->time;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
lastdp = &(*lastdp)->next;
}
*lastdp = dp;
if (ldp)
dp->time += lasttime;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
}
void plan_add_segment(struct diveplan *diveplan, int duration, int depth, int o2, int he, int po2)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
{
struct divedatapoint *dp = create_dp(duration, depth, o2, he, po2);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
add_to_end_of_diveplan(diveplan, dp);
}
struct gaschanges {
int depth;
int gasidx;
};
static struct gaschanges *analyze_gaslist(struct diveplan *diveplan, struct dive *dive, int *gaschangenr)
{
int nr = 0;
struct gaschanges *gaschanges = NULL;
struct divedatapoint *dp = diveplan->dp;
while (dp) {
if (dp->time == 0) {
int i = 0, j = 0;
nr++;
gaschanges = realloc(gaschanges, nr * sizeof(struct gaschanges));
while (i < nr - 1) {
if (dp->depth < gaschanges[i].depth) {
memmove(gaschanges + i + 1, gaschanges + i, (nr - i - 1) * sizeof(struct gaschanges));
break;
}
i++;
}
gaschanges[i].depth = dp->depth;
do {
if (dive->cylinder[j].gasmix.o2.permille == dp->o2 &&
dive->cylinder[j].gasmix.he.permille == dp->he) {
gaschanges[i].gasidx = j;
break;
}
j++;
} while (j < MAX_CYLINDERS);
}
dp = dp->next;
}
*gaschangenr = nr;
#if DEBUG_PLAN & 16
for (nr = 0; nr < *gaschangenr; nr++)
printf("gaschange nr %d: @ %5.2lfm gasidx %d (%d/%d)\n", nr, gaschanges[nr].depth / 1000.0,
gaschanges[nr].gasidx, (dive->cylinder[gaschanges[nr].gasidx].gasmix.o2.permille + 5) / 10,
(dive->cylinder[gaschanges[nr].gasidx].gasmix.he.permille + 5) / 10);
#endif
return gaschanges;
}
/* sort all the stops into one ordered list */
static int *sort_stops(int *dstops, int dnr, struct gaschanges *gstops, int gnr)
{
int i, gi, di;
int total = dnr + gnr;
int *stoplevels = malloc(total * sizeof(int));
/* no gaschanges */
if (gnr == 0) {
memcpy(stoplevels, dstops, dnr * sizeof(int));
return stoplevels;
}
i = total - 1;
gi = gnr - 1;
di = dnr - 1;
while (i >= 0) {
if (dstops[di] > gstops[gi].depth) {
stoplevels[i] = dstops[di];
di--;
} else if (dstops[di] == gstops[gi].depth) {
stoplevels[i] = dstops[di];
di--;
gi--;
} else {
stoplevels[i] = gstops[gi].depth;
gi--;
}
i--;
if (di < 0) {
while (gi >= 0)
stoplevels[i--] = gstops[gi--].depth;
break;
}
if (gi < 0) {
while (di >= 0)
stoplevels[i--] = dstops[di--];
break;
}
}
while (i >= 0)
stoplevels[i--] = 0;
#if DEBUG_PLAN & 16
int k;
for (k = gnr + dnr -1; k >= 0; k--) {
printf("stoplevel[%d]: %5.2lfm\n", k, stoplevels[k]/1000.0);
if (stoplevels[k] == 0)
break;
}
#endif
return stoplevels;
}
static void add_plan_to_notes(struct diveplan *diveplan, struct dive *dive)
{
char buffer[20000];
int consumption[MAX_CYLINDERS] = { 0, };
int len, gasidx, lastdepth = 0, lasttime = 0;
struct divedatapoint *dp = diveplan->dp;
int o2, he;
if (!dp)
return;
snprintf(buffer, sizeof(buffer), _("%s\nSubsurface dive plan\nbased on GFlow = %.0f and GFhigh = %.0f\n\n"),
disclaimer, plangflow * 100, plangfhigh * 100);
/* we start with gas 0, then check if that was changed */
o2 = dive->cylinder[0].gasmix.o2.permille;
he = dive->cylinder[0].gasmix.he.permille;
do {
const char *depth_unit;
char gas[64];
double depthvalue;
int decimals;
double used;
int newo2 = o2, newhe = he;
struct divedatapoint *nextdp;
if (dp->time == 0)
continue;
depthvalue = get_depth_units(dp->depth, &decimals, &depth_unit);
/* do we change gas after this segment? We need to look at the gas
* for the next segment (that isn't just a record of available gas !!)
* to find out */
nextdp = dp->next;
while (nextdp && nextdp->time == 0)
nextdp = nextdp->next;
if (nextdp) {
newo2 = nextdp->o2;
newhe = nextdp->he;
if (newhe == 0 && newo2 == 0) {
/* same as last segment */
newo2 = o2;
newhe = he;
}
}
/* do we want to skip this leg as it is devoid of anything useful? */
if (!dp->entered && o2 == newo2 && he == newhe && nextdp && dp->depth != lastdepth && nextdp->depth != dp->depth)
continue;
get_gas_string(o2, he, gas, sizeof(gas));
gasidx = get_gasidx(dive, o2, he);
len = strlen(buffer);
if (dp->depth != lastdepth) {
used = diveplan->bottomsac / 1000.0 * depth_to_mbar((dp->depth + lastdepth) / 2, dive) *
(dp->time - lasttime) / 60;
snprintf(buffer + len, sizeof(buffer) - len, _("Transition to %.*f %s in %d:%02d min - runtime %d:%02u on %s\n"),
decimals, depthvalue, depth_unit,
FRACTION(dp->time - lasttime, 60),
FRACTION(dp->time, 60),
gas);
} else {
/* we use deco SAC rate during the calculated deco stops, bottom SAC rate everywhere else */
int sac = dp->entered ? diveplan->bottomsac : diveplan->decosac;
used = sac / 1000.0 * depth_to_mbar(dp->depth, dive) * (dp->time - lasttime) / 60;
snprintf(buffer + len, sizeof(buffer) - len, _("Stay at %.*f %s for %d:%02d min - runtime %d:%02u on %s\n"),
decimals, depthvalue, depth_unit,
FRACTION(dp->time - lasttime, 60),
FRACTION(dp->time, 60),
gas);
}
if (gasidx != -1)
consumption[gasidx] += used;
get_gas_string(newo2, newhe, gas, sizeof(gas));
if (o2 != newo2 || he != newhe) {
len = strlen(buffer);
snprintf(buffer + len, sizeof(buffer) - len, _("Switch gas to %s\n"), gas);
}
o2 = newo2;
he = newhe;
lasttime = dp->time;
lastdepth = dp->depth;
} while ((dp = dp->next) != NULL);
len = strlen(buffer);
snprintf(buffer + len, sizeof(buffer) - len, _("Gas consumption:\n"));
for (gasidx = 0; gasidx < MAX_CYLINDERS; gasidx++) {
double volume;
const char *unit;
char gas[64];
if (consumption[gasidx] == 0)
continue;
len = strlen(buffer);
volume = get_volume_units(consumption[gasidx], NULL, &unit);
get_gas_string(dive->cylinder[gasidx].gasmix.o2.permille,
dive->cylinder[gasidx].gasmix.he.permille, gas, sizeof(gas));
snprintf(buffer + len, sizeof(buffer) - len, _("%.0f%s of %s\n"), volume, unit, gas);
}
dive->notes = strdup(buffer);
}
void plan(struct diveplan *diveplan, char **cached_datap, struct dive **divep)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
{
struct dive *dive;
struct sample *sample;
int wait_time, o2, he, po2;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
int ceiling, depth, transitiontime;
int stopidx, gi;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
double tissue_tolerance;
struct gaschanges *gaschanges;
int gaschangenr;
int *stoplevels;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
set_gf(plangflow, plangfhigh);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
if (!diveplan->surface_pressure)
diveplan->surface_pressure = SURFACE_PRESSURE;
if (*divep)
delete_single_dive(dive_table.nr - 1);
*divep = dive = create_dive_from_plan(diveplan);
if (!dive)
return;
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
record_dive(dive);
sample = &dive->dc.sample[dive->dc.samples - 1];
/* we start with gas 0, then check if that was changed */
o2 = dive->cylinder[0].gasmix.o2.permille;
he = dive->cylinder[0].gasmix.he.permille;
get_gas_from_events(&dive->dc, sample->time.seconds, &o2, &he);
po2 = dive->dc.sample[dive->dc.samples - 1].po2;
depth = dive->dc.sample[dive->dc.samples - 1].depth.mm;
tissue_tolerance = tissue_at_end(dive, cached_datap);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
ceiling = deco_allowed_depth(tissue_tolerance, diveplan->surface_pressure / 1000.0, dive, 1);
#if DEBUG_PLAN & 4
printf("gas %d/%d\n", o2, he);
printf("depth %5.2lfm ceiling %5.2lfm\n", depth / 1000.0, ceiling / 1000.0);
#endif
if (depth < ceiling) /* that's not good... */
depth = ceiling;
for (stopidx = 0; stopidx < sizeof(decostoplevels) / sizeof(int); stopidx++)
if (decostoplevels[stopidx] >= depth)
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
break;
stopidx--;
/* so now we know the first decostop level above us
* NOTE, this could be the surface or a long list of potential stops
* we do NOT start only at the ceiling, as the ceiling may come down
* further during the ascent.
* Next we need to figure out if there are better gases available
* and at which depths we are supposed to switch to them */
gaschanges = analyze_gaslist(diveplan, dive, &gaschangenr);
stoplevels = sort_stops(decostoplevels, stopidx + 1, gaschanges, gaschangenr);
gi = gaschangenr - 1;
stopidx += gaschangenr;
if (depth > stoplevels[stopidx]) {
transitiontime = (depth - stoplevels[stopidx]) / 150;
#if DEBUG_PLAN & 2
printf("transitiontime %d:%02d to depth %5.2lfm\n", FRACTION(transitiontime, 60), stoplevels[stopidx] / 1000.0);
#endif
plan_add_segment(diveplan, transitiontime, stoplevels[stopidx], o2, he, po2);
/* re-create the dive */
delete_single_dive(dive_table.nr - 1);
*divep = dive = create_dive_from_plan(diveplan);
record_dive(dive);
}
while (stopidx > 0) { /* this indicates that we aren't surfacing directly */
if (gi >= 0 && stoplevels[stopidx] == gaschanges[gi].depth) {
o2 = dive->cylinder[gaschanges[gi].gasidx].gasmix.o2.permille;
he = dive->cylinder[gaschanges[gi].gasidx].gasmix.he.permille;
#if DEBUG_PLAN & 16
printf("switch to gas %d (%d/%d) @ %5.2lfm\n", gaschanges[gi].gasidx,
(o2 + 5) / 10, (he + 5) / 10, gaschanges[gi].depth / 1000.0);
#endif
gi--;
}
wait_time = time_at_last_depth(dive, o2, he, stoplevels[stopidx - 1], cached_datap);
/* typically deco plans are done in one minute increments; we may want to
* make this configurable at some point */
wait_time = ((wait_time + 59) / 60) * 60;
#if DEBUG_PLAN & 2
tissue_tolerance = tissue_at_end(dive, cached_datap);
ceiling = deco_allowed_depth(tissue_tolerance, diveplan->surface_pressure / 1000.0, dive, 1);
printf("waittime %d:%02d at depth %5.2lfm; ceiling %5.2lfm\n", FRACTION(wait_time, 60),
stoplevels[stopidx] / 1000.0, ceiling / 1000.0);
#endif
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
if (wait_time)
plan_add_segment(diveplan, wait_time, stoplevels[stopidx], o2, he, po2);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
transitiontime = (stoplevels[stopidx] - stoplevels[stopidx - 1]) / 150;
#if DEBUG_PLAN & 2
printf("transitiontime %d:%02d to depth %5.2lfm\n", FRACTION(transitiontime, 60), stoplevels[stopidx - 1] / 1000.0);
#endif
plan_add_segment(diveplan, transitiontime, stoplevels[stopidx - 1], o2, he, po2);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
/* re-create the dive */
delete_single_dive(dive_table.nr - 1);
*divep = dive = create_dive_from_plan(diveplan);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
record_dive(dive);
stopidx--;
}
add_plan_to_notes(diveplan, dive);
/* now make the dive visible in the dive list */
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
report_dives(FALSE, FALSE);
show_and_select_dive(dive);
free(stoplevels);
free(gaschanges);
First stab at simplistic dive planning This comes with absolutely no gui - so the plan literally needs to be compiled into Subsurface. Not exactly a feature, but this allowed me to focus on the planning part instead of spending time on tedious UI work. A new menu "Planner" with entry "Test Planner" calls into the hard-coded function in planner.c. There a simple dive plan can be constructed with calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2) Calling plan(&diveplan) does the deco calculations and creates deco stops that keep us below the ceiling (with the GFlow/high values currently configured). The stop levels used are defined at the top of planner.c in the stoplevels array - there is no need to do the traditional multiples of 3m or anything like that. The dive including the ascents and deco stops all the way to the surface is completed and then added as simulated dive to the end of the divelist (I guess we could automatically select it later) and can be viewed. This is crude but shows the direction we can go with this. Envision a nice UI that allows you to simply enter the segments and pick the desired stops. What is missing is the ability to give the algorithm additional gases that it can use during the deco phase - right now it simply keeps using the last gas used in the diveplan. All that said, there are clear bugs here - and sadly they seem to be in the deco calculations, as with the example given the ceiling that is calculated makes no sense. When displayed in smooth mode it has very strange jumps up and down that I wouldn't expect. For example with GF 35/75 (the default) the deco ceiling when looking at the simulated dive jumps from 16m back up to 13m around 14:10 into the dive. That seems very odd. Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2013-01-05 07:11:42 +00:00
}
/* and now the UI for all this */
/*
* Get a value in tenths (so "10.2" == 102, "9" = 90)
*
* Return negative for errors.
*/
static int get_tenths(const char *begin, const char **endp)
{
char *end;
int value = strtol(begin, &end, 10);
if (begin == end)
return -1;
value *= 10;
/* Fraction? We only look at the first digit */
if (*end == '.') {
end++;
if (!isdigit(*end))
return -1;
value += *end - '0';
do {
end++;
} while (isdigit(*end));
}
*endp = end;
return value;
}
static int get_thousandths(const char *begin, const char **endp)
{
char *end;
int value = strtol(begin, &end, 10);
if (begin == end)
return -1;
value *= 1000;
/* now deal with up to three more digits after decimal point */
if (*end == '.') {
int factor = 100;
do {
++end;
if (!isdigit(*end))
break;
value += (*end - '0') * factor;
factor /= 10;
} while (factor);
}
*endp = end;
return value;
}
static int get_permille(const char *begin, const char **end)
{
int value = get_tenths(begin, end);
if (value >= 0) {
/* Allow a percentage sign */
if (**end == '%')
++*end;
}
return value;
}
static int validate_gas(const char *text, int *o2_p, int *he_p)
{
int o2, he;
if (!text)
return 0;
while (g_ascii_isspace(*text))
text++;
if (!*text)
return 0;
if (!strcasecmp(text, _("air"))) {
o2 = O2_IN_AIR; he = 0; text += strlen(_("air"));
} else if (!strncasecmp(text, _("ean"), 3)) {
o2 = get_permille(text+3, &text); he = 0;
} else {
o2 = get_permille(text, &text); he = 0;
if (*text == '/')
he = get_permille(text+1, &text);
}
/* We don't want any extra crud */
while (g_ascii_isspace(*text))
text++;
if (*text)
return 0;
/* Validate the gas mix */
if (*text || o2 < 1 || o2 > 1000 || he < 0 || o2+he > 1000)
return 0;
/* Let it rip */
*o2_p = o2;
*he_p = he;
return 1;
}
static int validate_time(const char *text, int *sec_p, int *rel_p)
{
int min, sec, rel;
char *end;
if (!text)
return 0;
while (g_ascii_isspace(*text))
text++;
rel = 0;
if (*text == '+') {
rel = 1;
text++;
while (g_ascii_isspace(*text))
text++;
}
min = strtol(text, &end, 10);
if (text == end)
return 0;
if (min < 0 || min > 1000)
return 0;
/* Ok, minutes look ok */
text = end;
sec = 0;
if (*text == ':') {
text++;
sec = strtol(text, &end, 10);
if (end != text+2)
return 0;
if (sec < 0)
return 0;
text = end;
if (*text == ':') {
if (sec >= 60)
return 0;
min = min*60 + sec;
text++;
sec = strtol(text, &end, 10);
if (end != text+2)
return 0;
if (sec < 0)
return 0;
text = end;
}
}
/* Maybe we should accept 'min' at the end? */
if (g_ascii_isspace(*text))
text++;
if (*text)
return 0;
*sec_p = min*60 + sec;
*rel_p = rel;
return 1;
}
static int validate_depth(const char *text, int *mm_p)
{
int depth, imperial;
if (!text)
return 0;
depth = get_tenths(text, &text);
if (depth < 0)
return 0;
while (g_ascii_isspace(*text))
text++;
imperial = get_units()->length == FEET;
if (*text == 'm') {
imperial = 0;
text++;
} else if (!strcasecmp(text, _("ft"))) {
imperial = 1;
text += 2;
}
while (g_ascii_isspace(*text))
text++;
if (*text)
return 0;
if (imperial) {
depth = feet_to_mm(depth / 10.0);
} else {
depth *= 100;
}
*mm_p = depth;
/* we don't support extreme depths */
if (depth > 400000)
return 0;
return 1;
}
static int validate_po2(const char *text, int *mbar_po2)
{
int po2;
if (!text)
return 0;
po2 = get_tenths(text, &text);
if (po2 < 0)
return 0;
while (g_ascii_isspace(*text))
text++;
while (g_ascii_isspace(*text))
text++;
if (*text)
return 0;
*mbar_po2 = po2 * 100;
return 1;
}
static int validate_volume(const char *text, int *sac)
{
int volume, imperial;
if (!text)
return 0;
volume = get_thousandths(text, &text);
if (volume < 0)
return 0;
while (g_ascii_isspace(*text))
text++;
imperial = get_units()->volume == CUFT;
if (*text == 'l') {
imperial = 0;
text++;
} else if (!strncasecmp(text, _("cuft"), 4)) {
imperial = 1;
text += 4;
}
while (g_ascii_isspace(*text) || *text == '/')
text++;
if (!strncasecmp(text, _("min"), 3))
text += 3;
while (g_ascii_isspace(*text))
text++;
if (*text)
return 0;
if (imperial)
volume = cuft_to_l(volume) + 0.5; /* correct for mcuft -> ml */
*sac = volume;
return 1;
}
static GtkWidget *add_entry_to_box(GtkWidget *box, const char *label)
{
GtkWidget *entry, *frame;
entry = gtk_entry_new();
gtk_entry_set_max_length(GTK_ENTRY(entry), 16);
if (label) {
frame = gtk_frame_new(label);
gtk_container_add(GTK_CONTAINER(frame), entry);
gtk_box_pack_start(GTK_BOX(box), frame, FALSE, FALSE, 0);
} else {
gtk_box_pack_start(GTK_BOX(box), entry, FALSE, FALSE, 2);
}
return entry;
}
#define MAX_WAYPOINTS 12
GtkWidget *entry_depth[MAX_WAYPOINTS], *entry_duration[MAX_WAYPOINTS], *entry_gas[MAX_WAYPOINTS], *entry_po2[MAX_WAYPOINTS];
int nr_waypoints = 0;
static GtkListStore *gas_model = NULL;
struct diveplan diveplan = {};
char *cache_data = NULL;
struct dive *planned_dive = NULL;
/* make a copy of the diveplan so far and display the corresponding dive */
void show_planned_dive(void)
{
struct diveplan tempplan;
struct divedatapoint *dp, **dpp;
memcpy(&tempplan, &diveplan, sizeof(struct diveplan));
dpp = &tempplan.dp;
dp = diveplan.dp;
while (dp && *dpp) {
*dpp = malloc(sizeof(struct divedatapoint));
memcpy(*dpp, dp, sizeof(struct divedatapoint));
dp = dp->next;
dpp = &(*dpp)->next;
}
#if DEBUG_PLAN & 1
printf("in show_planned_dive:\n");
dump_plan(&tempplan);
#endif
plan(&tempplan, &cache_data, &planned_dive);
free_dps(tempplan.dp);
}
static gboolean gas_focus_out_cb(GtkWidget *entry, GdkEvent *event, gpointer data)
{
const char *gastext;
int o2, he;
int idx = data - NULL;
gastext = gtk_entry_get_text(GTK_ENTRY(entry));
o2 = he = 0;
if (validate_gas(gastext, &o2, &he))
add_string_list_entry(gastext, gas_model);
add_gas_to_nth_dp(&diveplan, idx, o2, he);
show_planned_dive();
return FALSE;
}
static void gas_changed_cb(GtkWidget *combo, gpointer data)
{
const char *gastext;
int o2, he;
int idx = data - NULL;
gastext = get_active_text(GTK_COMBO_BOX(combo));
/* stupidly this gets called for two reasons:
* a) any keystroke into the entry field
* b) mouse selection of a dropdown
* we only care about b) (a) is handled much better with the focus-out event)
* so let's check that the text returned is actually in our model before going on
*/
if (match_list(gas_model, gastext) != MATCH_EXACT)
return;
o2 = he = 0;
if (!validate_gas(gastext, &o2, &he)) {
/* this should never happen as only validated texts should be
* in the dropdown */
show_error(_("Invalid gas for row %d"),idx);
}
add_gas_to_nth_dp(&diveplan, idx, o2, he);
show_planned_dive();
}
static gboolean depth_focus_out_cb(GtkWidget *entry, GdkEvent *event, gpointer data)
{
const char *depthtext;
int depth = -1;
int idx = data - NULL;
depthtext = gtk_entry_get_text(GTK_ENTRY(entry));
if (validate_depth(depthtext, &depth)) {
if (depth > 150000)
show_error(_("Warning - planning very deep dives can take excessive amounts of time"));
add_depth_to_nth_dp(&diveplan, idx, depth);
show_planned_dive();
} else {
/* it might be better to instead change the color of the input field or something */
if (depth == -1)
show_error(_("Invalid depth - could not parse \"%s\""), depthtext);
else
show_error(_("Invalid depth - values deeper than 400m not supported"));
}
return FALSE;
}
static gboolean duration_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *durationtext;
int duration, is_rel;
int idx = data - NULL;
durationtext = gtk_entry_get_text(GTK_ENTRY(entry));
if (validate_time(durationtext, &duration, &is_rel))
add_duration_to_nth_dp(&diveplan, idx, duration, is_rel);
show_planned_dive();
return FALSE;
}
static gboolean po2_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *po2text;
int po2;
int idx = data - NULL;
po2text = gtk_entry_get_text(GTK_ENTRY(entry));
if (validate_po2(po2text, &po2))
add_po2_to_nth_dp(&diveplan, idx, po2);
show_planned_dive();
return FALSE;
}
/* Subsurface follows the lead of most divecomputers to use times
* without timezone - so all times are implicitly assumed to be
* local time of the dive location; so in order to give the current
* time in that way we actually need to add the timezone offset */
static timestamp_t current_time_notz(void)
{
time_t now = time(NULL);
struct tm *local = localtime(&now);
return utc_mktime(local);
}
static gboolean starttime_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *starttimetext;
int starttime, is_rel;
starttimetext = gtk_entry_get_text(GTK_ENTRY(entry));
if (validate_time(starttimetext, &starttime, &is_rel)) {
/* we alway make this relative - either from the current time or from the
* end of the last dive, whichever is later */
timestamp_t cur = current_time_notz();
if (diveplan.lastdive_nr >= 0) {
struct dive *last_dive = get_dive(diveplan.lastdive_nr);
if (last_dive && last_dive->when + last_dive->dc.duration.seconds > cur)
cur = last_dive->when + last_dive->dc.duration.seconds;
}
diveplan.when = cur + starttime;
show_planned_dive();
} else {
/* it might be better to instead change the color of the input field or something */
show_error(_("Invalid starttime"));
}
return FALSE;
}
static gboolean surfpres_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *surfprestext;
surfprestext = gtk_entry_get_text(GTK_ENTRY(entry));
diveplan.surface_pressure = atoi(surfprestext);
show_planned_dive();
return FALSE;
}
static gboolean sac_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *sactext;
sactext = gtk_entry_get_text(GTK_ENTRY(entry));
if (validate_volume(sactext, data))
show_planned_dive();
return FALSE;
}
static gboolean gf_focus_out_cb(GtkWidget *entry, GdkEvent * event, gpointer data)
{
const char *gftext;
int gf;
double *gfp = data;
gftext = gtk_entry_get_text(GTK_ENTRY(entry));
if (sscanf(gftext, "%d", &gf) == 1) {
*gfp = gf / 100.0;
show_planned_dive();
}
return FALSE;
}
static GtkWidget *add_gas_combobox_to_box(GtkWidget *box, const char *label, int idx)
{
GtkWidget *frame, *combo;
if (!gas_model) {
gas_model = gtk_list_store_new(1, G_TYPE_STRING);
add_string_list_entry(_("AIR"), gas_model);
add_string_list_entry(_("EAN32"), gas_model);
add_string_list_entry(_("EAN36"), gas_model);
}
combo = combo_box_with_model_and_entry(gas_model);
gtk_widget_add_events(combo, GDK_FOCUS_CHANGE_MASK);
g_signal_connect(gtk_bin_get_child(GTK_BIN(combo)), "focus-out-event", G_CALLBACK(gas_focus_out_cb), NULL + idx);
g_signal_connect(combo, "changed", G_CALLBACK(gas_changed_cb), NULL + idx);
if (label) {
frame = gtk_frame_new(label);
gtk_box_pack_start(GTK_BOX(box), frame, FALSE, FALSE, 0);
gtk_container_add(GTK_CONTAINER(frame), combo);
} else {
gtk_box_pack_start(GTK_BOX(box), combo, FALSE, FALSE, 2);
}
return combo;
}
static void add_waypoint_widgets(GtkWidget *box, int idx)
{
GtkWidget *hbox;
hbox = gtk_hbox_new(FALSE, 0);
gtk_box_pack_start(GTK_BOX(box), hbox, FALSE, FALSE, 0);
if (idx == 0) {
entry_depth[idx] = add_entry_to_box(hbox, _("Ending Depth"));
entry_duration[idx] = add_entry_to_box(hbox, _("Segment Time"));
entry_gas[idx] = add_gas_combobox_to_box(hbox, C_("Type of","Gas Used"), idx);
entry_po2[idx] = add_entry_to_box(hbox, _("CC SetPoint"));
} else {
entry_depth[idx] = add_entry_to_box(hbox, NULL);
entry_duration[idx] = add_entry_to_box(hbox, NULL);
entry_gas[idx] = add_gas_combobox_to_box(hbox, NULL, idx);
entry_po2[idx] = add_entry_to_box(hbox, NULL);
}
gtk_widget_add_events(entry_depth[idx], GDK_FOCUS_CHANGE_MASK);
g_signal_connect(entry_depth[idx], "focus-out-event", G_CALLBACK(depth_focus_out_cb), NULL + idx);
gtk_widget_add_events(entry_duration[idx], GDK_FOCUS_CHANGE_MASK);
g_signal_connect(entry_duration[idx], "focus-out-event", G_CALLBACK(duration_focus_out_cb), NULL + idx);
gtk_widget_add_events(entry_po2[idx], GDK_FOCUS_CHANGE_MASK);
g_signal_connect(entry_po2[idx], "focus-out-event", G_CALLBACK(po2_focus_out_cb), NULL + idx);
}
static void add_waypoint_cb(GtkButton *button, gpointer _data)
{
GtkWidget *vbox = _data;
if (nr_waypoints < MAX_WAYPOINTS) {
GtkWidget *ovbox, *dialog;
add_waypoint_widgets(vbox, nr_waypoints);
nr_waypoints++;
ovbox = gtk_widget_get_parent(GTK_WIDGET(button));
dialog = gtk_widget_get_parent(ovbox);
gtk_widget_show_all(dialog);
} else {
show_error(_("Too many waypoints"));
}
}
static void add_entry_with_callback(GtkWidget *box, int length, char *label, char *initialtext,
gboolean (*callback)(GtkWidget *, GdkEvent *, gpointer), gpointer data)
{
GtkWidget *entry = add_entry_to_box(box, label);
gtk_entry_set_max_length(GTK_ENTRY(entry), length);
gtk_entry_set_text(GTK_ENTRY(entry), initialtext);
gtk_widget_add_events(entry, GDK_FOCUS_CHANGE_MASK);
g_signal_connect(entry, "focus-out-event", G_CALLBACK(callback), data);
}
/* set up the dialog where the user can input their dive plan */
void input_plan()
{
GtkWidget *content, *vbox, *hbox, *outervbox, *add_row, *label;
char *bottom_sac, *deco_sac, gflowstring[4], gfhighstring[4];
char *explanationtext = _("<small>Add segments below.\nEach line describes part of the planned dive.\n"
"An entry with depth, time and gas describes a segment that ends "
"at the given depth, takes the given time (if relative, e.g. '+3:30') "
"or ends at the given time (if absolute), and uses the given gas.\n"
"An empty gas means 'use previous gas' (or AIR if no gas was specified).\n"
"An entry that has a depth and a gas given but no time is special; it "
"informs the planner that the gas specified is available for the ascent "
"once the depth given has been reached.\n"
"CC SetPoint specifies CC (rebreather) dives, leave empty for OC.</small>\n");
char *labeltext;
int len;
disclaimer = _("DISCLAIMER / WARNING: THIS IS A NEW IMPLEMENTATION OF THE BUHLMANN "
"ALGORITHM AND A DIVE PLANNER IMPLEMENTION BASED ON THAT WHICH HAS "
"RECEIVED ONLY A LIMITED AMOUNT OF TESTING. WE STRONGLY RECOMMEND NOT TO "
"PLAN DIVES SIMPLY BASED ON THE RESULTS GIVEN HERE.");
if (diveplan.dp)
free_dps(diveplan.dp);
memset(&diveplan, 0, sizeof(diveplan));
diveplan.lastdive_nr = dive_table.nr - 1;
free(cache_data);
cache_data = NULL;
planned_dive = NULL;
planner = gtk_dialog_new_with_buttons(_("Dive Plan - THIS IS JUST A SIMULATION; DO NOT USE FOR DIVING"),
GTK_WINDOW(main_window),
GTK_DIALOG_DESTROY_WITH_PARENT,
GTK_STOCK_OK, GTK_RESPONSE_ACCEPT,
GTK_STOCK_CANCEL, GTK_RESPONSE_CANCEL,
NULL);
content = gtk_dialog_get_content_area (GTK_DIALOG (planner));
outervbox = gtk_vbox_new(FALSE, 2);
gtk_container_add (GTK_CONTAINER (content), outervbox);
len = strlen(explanationtext) + strlen(disclaimer) + sizeof("<span foreground='red'></span>");
labeltext = malloc(len);
snprintf(labeltext, len, "%s<span foreground='red'>%s</span>", explanationtext, disclaimer);
label = gtk_label_new(labeltext);
gtk_label_set_line_wrap(GTK_LABEL(label), TRUE);
gtk_label_set_use_markup(GTK_LABEL(label), TRUE);
gtk_label_set_width_chars(GTK_LABEL(label), 60);
gtk_box_pack_start(GTK_BOX(outervbox), label, TRUE, TRUE, 0);
vbox = gtk_vbox_new(FALSE, 0);
gtk_box_pack_start(GTK_BOX(outervbox), vbox, TRUE, TRUE, 0);
hbox = gtk_hbox_new(FALSE, 0);
gtk_box_pack_start(GTK_BOX(vbox), hbox, TRUE, TRUE, 0);
add_entry_with_callback(hbox, 12, _("Dive starts when?"), "+60:00", starttime_focus_out_cb, NULL);
add_entry_with_callback(hbox, 12, _("Surface Pressure (mbar)"), SURFACE_PRESSURE_STRING, surfpres_focus_out_cb, NULL);
hbox = gtk_hbox_new(FALSE, 0);
gtk_box_pack_start(GTK_BOX(vbox), hbox, TRUE, TRUE, 0);
if (get_units()->volume == CUFT) {
bottom_sac = _("0.7 cuft/min");
deco_sac = _("0.6 cuft/min");
diveplan.bottomsac = 1000 * cuft_to_l(0.7);
diveplan.decosac = 1000 * cuft_to_l(0.6);
} else {
bottom_sac = _("20 l/min");
deco_sac = _("17 l/min");
diveplan.bottomsac = 20000;
diveplan.decosac = 17000;
}
add_entry_with_callback(hbox, 12, _("SAC during dive"), bottom_sac, sac_focus_out_cb, &diveplan.bottomsac);
add_entry_with_callback(hbox, 12, _("SAC during decostop"), deco_sac, sac_focus_out_cb, &diveplan.decosac);
plangflow = prefs.gflow;
plangfhigh = prefs.gfhigh;
snprintf(gflowstring, sizeof(gflowstring), "%3.0f", 100 * plangflow);
snprintf(gfhighstring, sizeof(gflowstring), "%3.0f", 100 * plangfhigh);
add_entry_with_callback(hbox, 5, _("GFlow for plan"), gflowstring, gf_focus_out_cb, &plangflow);
add_entry_with_callback(hbox, 5, _("GFhigh for plan"), gfhighstring, gf_focus_out_cb, &plangfhigh);
diveplan.when = current_time_notz() + 3600;
diveplan.surface_pressure = SURFACE_PRESSURE;
nr_waypoints = 4;
add_waypoint_widgets(vbox, 0);
add_waypoint_widgets(vbox, 1);
add_waypoint_widgets(vbox, 2);
add_waypoint_widgets(vbox, 3);
add_row = gtk_button_new_with_label(_("Add waypoint"));
g_signal_connect(G_OBJECT(add_row), "clicked", G_CALLBACK(add_waypoint_cb), vbox);
gtk_box_pack_start(GTK_BOX(outervbox), add_row, FALSE, FALSE, 0);
gtk_widget_show_all(planner);
if (gtk_dialog_run(GTK_DIALOG(planner)) == GTK_RESPONSE_ACCEPT) {
plan(&diveplan, &cache_data, &planned_dive);
mark_divelist_changed(TRUE);
} else {
if (planned_dive) {
/* we have added a dive during the dynamic construction
* in the dialog; get rid of it */
delete_single_dive(dive_table.nr - 1);
report_dives(FALSE, FALSE);
planned_dive = NULL;
}
}
gtk_widget_destroy(planner);
planner_error_bar = NULL;
error_label = NULL;
set_gf(prefs.gflow, prefs.gfhigh);
}