Add interval constraints for orbit models (Rosetta uses P67 after a time)

Goutte
1 parent dc0be992
Showing 4 changed files with 71 additions and 34 deletions Show diff stats
CHANGELOG.md
config.yml
web/run.py
web/view/layout.html.jinja2
 ## Future ?
-- [ ] Optimize CSV generation (numpy vectorization ?)
+- [ ] Optimize data aggregation (numpy vectorization?)
 - [ ] Credit the author of the pixel art planets
 - [ ] Add a README to the download tarball
 - [ ] Set the log level to _error_ in production (see `web/run.py`)
-- [ ] Cache warmup (generate for today's default interval) `/cache/warmup`
 - [ ] CRON statements to call the cache cleanup and warmup
-- [ ] Add a priority for models to support Juno's trajectory (using Jupiter's)
+- [ ] Cache warmup (generate for today's default interval) `/cache/warmup`
+- [ ] Normalize time interval for time series
+- [ ] Give the future data another color
+- [ ] Sort times series by closeness to the sun
+- [ ] Generate a CDF file (not NetCDF)
+- [ ] Make the footer images clickable
+- [ ] Move the link to the source in the footer
+- [ ] Highlight the visits counter
+
+An heliospheric propagation 1D MHD model for solar wind prediction at planets, probes and comets.
+
 ## 1.0.0-rc4
-- [ ] Make the tarball with netcdf files instead of CSVs
+- [x] Add interval constraints for orbit models (Rosetta uses P67 after a time)
+- [x] Make the download with a netcdf file instead of a tarball of CSVs
 - [x] Support having no position to display (for Rosetta in some intervals)
 - [x] Make the local cache more resilient to corrupted downloads
 - [x] Make the local cache more resilient to naming collisions
@@ -156,6 +156,9 @@ targets:
       orbit:
         models:
           - slug: 'ros_orb_cruise'
+            stopped_at: '2014-08-02T00:00:00'
+          - slug: 'p67_orb_all'
+            started_at: '2014-08-02T00:00:00'
       models:
           - slug: 'tao_ros_sw'
       locked: false
@@ -392,7 +392,7 @@ def retrieve_amda_netcdf(orbiter, what, started_at, stopped_at):
             local_netc_files.append(local_netc_file)
             log.debug("Unzipped '%s'." % local_gzip_file)
-    return local_netc_files
+    return list(set(local_netc_files))  # remove possible dupes
 def get_data_for_target(target_config, started_at, stopped_at):
@@ -412,21 +412,25 @@ def get_data_for_target(target_config, started_at, stopped_at):
         abort(500, "Invalid orbit configuration for '%s' : %s"
               % (target_config['slug'], str(e)))
+    # try:
+    #     started_at = datetime.datetime.strptime(started_at, FILE_DATE_FMT)
+    # except:
+    #     abort(400, "Invalid started_at config : '%s'." % started_at)
+    # try:
+    #     stopped_at = datetime.datetime.strptime(stopped_at, FILE_DATE_FMT)
+    # except:
+    #     abort(400, "Invalid stopped_at config : '%s'." % stopped_at)
+
     # Grab the list of netCDF files from Myriam's API
     model_files = []
-    orbit_files = []
     for model in models:
         model_files = model_files + retrieve_amda_netcdf(
             target_config['slug'], model['slug'], started_at, stopped_at
         )
-    for orbit in orbits:
-        orbit_files = orbit_files + retrieve_amda_netcdf(
-            target_config['slug'], orbit['slug'], started_at, stopped_at
-        )
     # Remove possible duplicates
     model_files = set(model_files)
-    orbit_files = set(orbit_files)
+    # Let's return an empty dict when there's no data instead of crashing
     # if not len(model_files):
     #     abort(500, "No model files found for '%s'." % target_config['slug'])
     # if not len(orbit_files):
@@ -434,20 +438,38 @@ def get_data_for_target(target_config, started_at, stopped_at):
     precision = "%Y-%m-%dT%H"  # model and orbits times are only equal-ish
     orbit_data = {}  # keys are datetime as str, values arrays of XY
-    for orbit_file in orbit_files:
-        log.debug("%s: opening orbit NETCDF4 '%s'..." %
-                  (target_config['name'], orbit_file))
-        cdf_handle = Dataset(orbit_file, "r", format="NETCDF4")
-        times = cdf_handle.variables['Time']  # YYYY DOY HH MM SS .ms
-        data_hee = cdf_handle.variables['HEE']
-        log.debug("%s: aggregating data from '%s'..." %
-                  (target_config['name'], orbit_file))
-        for time, datum_hee in zip(times, data_hee):
-            dtime = datetime_from_list(time)
-            if started_at <= dtime <= stopped_at:
-                dkey = dtime.strftime(precision)
-                orbit_data[dkey] = datum_hee
-        cdf_handle.close()
+
+    for orbit in orbits:
+        if 'started_at' in orbit:
+            s0 = datetime.datetime.strptime(orbit['started_at'], FILE_DATE_FMT)
+        else:
+            s0 = started_at
+        if 'stopped_at' in orbit:
+            s1 = datetime.datetime.strptime(orbit['stopped_at'], FILE_DATE_FMT)
+        else:
+            s1 = stopped_at
+
+        orbit_files = retrieve_amda_netcdf(
+            target_config['slug'], orbit['slug'], s0, s1
+        )
+        for orbit_file in orbit_files:
+            log.debug("%s: opening orbit NETCDF4 '%s'..." %
+                      (target_config['name'], orbit_file))
+            cdf_handle = Dataset(orbit_file, "r", format="NETCDF4")
+            times = cdf_handle.variables['Time']  # YYYY DOY HH MM SS .ms
+            try:
+                data_hee = cdf_handle.variables['HEE']
+            except KeyError:
+                data_hee = cdf_handle.variables['XYZ_HEE']  # p67 uses this
+
+            log.debug("%s: aggregating data from '%s'..." %
+                      (target_config['name'], orbit_file))
+            for time, datum_hee in zip(times, data_hee):
+                dtime = datetime_from_list(time)
+                if s0 <= dtime <= s1:
+                    dkey = dtime.strftime(precision)
+                    orbit_data[dkey] = datum_hee
+            cdf_handle.close()
     all_data = {}  # keys are datetime as str, values tuples of data
     for model_file in model_files:
@@ -681,6 +703,8 @@ def download_targets_tarball(targets, started_at, stopped_at):
     in their own CSV file, and make a tarball of them.
     `started_at` and `stopped_at` should be UTC strings.
+    Note: we do not use this route anymore, but let's keep it shelved for now.
+
     targets: string list of targets' slugs, separated by `-`.
     """
     separator = '-'
@@ -694,17 +718,16 @@ def download_targets_tarball(targets, started_at, stopped_at):
     if 0 == len(targets_configs):
         abort(400, "No valid targets specified. What are you doing?")
-    date_fmt = FILE_DATE_FMT
     try:
-        started_at = datetime.datetime.strptime(started_at, date_fmt)
+        started_at = datetime.datetime.strptime(started_at, FILE_DATE_FMT)
     except:
         abort(400, "Invalid started_at parameter : '%s'." % started_at)
     try:
-        stopped_at = datetime.datetime.strptime(stopped_at, date_fmt)
+        stopped_at = datetime.datetime.strptime(stopped_at, FILE_DATE_FMT)
     except:
         abort(400, "Invalid stopped_at parameter : '%s'." % stopped_at)
-    sta = started_at.strftime(date_fmt)
-    sto = stopped_at.strftime(date_fmt)
+    sta = started_at.strftime(FILE_DATE_FMT)
+    sto = stopped_at.strftime(FILE_DATE_FMT)
     gzip_filename = "%s_%s_%s.tar.gz" % (separator.join(targets), sta, sto)
     local_gzip_file = join(CACHE_DIR, gzip_filename)
@@ -780,9 +803,9 @@ def download_targets_netcdf(targets, params, started_at, stopped_at):
         log.debug("Creating the NetCDF file '%s'..." % nc_filename)
         nc_handle = Dataset(nc_path, "w", format="NETCDF4")
         try:
-            nc_handle.description = "TODO"  # todo
+            nc_handle.description = "Model and orbit data for targets"  # todo
             nc_handle.history = "Created " + time.ctime(time.time())
-            nc_handle.source = "Transplanet (CDDP)"
+            nc_handle.source = "Heliopropa (CDDP)"
             available_params = list(PROPERTIES)
             for target in targets_configs:
                 target_slug = target['slug']
@@ -42,8 +42,9 @@
           <!-- Navigation. We hide it in small screens. -->
           <nav class="mdl-navigation mdl-layout--large-screen-only">
             <a class="mdl-navigation__link" href="https://gitlab.irap.omp.eu/CDPP/SPACEWEATHERONLINE">v{{ version }}</a>
-            <a class="mdl-navigation__link" href="https://onlinelibrary.wiley.com/doi/10.1029/2004JA010959/abstract">Model</a>
-            <a class="mdl-navigation__link" href="https://ec.europa.eu/programmes/horizon2020/">Horizon 2020</a>
+            <a class="mdl-navigation__link" href="https://onlinelibrary.wiley.com/doi/10.1029/2004JA010959/abstract">Tao's Model</a>
+            <a class="mdl-navigation__link" href="http://planetaryspaceweather-europlanet.irap.omp.eu">Planetary Space Weather Services</a>
+{#            <a class="mdl-navigation__link" href="https://ec.europa.eu/programmes/horizon2020/">Horizon 2020</a>#}
           </nav>
         </div>
       </header>