Merge branch 'plots/sankey' into visualization - early parts

* plots/sankey: (28 commits)
  diffinsights_web: File with Sankey data -> file without transition works
  diffinsights_web: Improve message when Sankey diagram is not possible
  diffinsigths_web: Remove leftover debug-print, fix unlikely case
  diffinsights_web: Fix SankeyPlot not refreshing on file change
  diffinsights_web: Fix LinesStatsDataStore not refreshing on file change
  diffinsights_web: Turn off disabling SANKEY_DIAGRAM for "Contributions:"
  diffinsights_web: Add support for "sankey" plot type to sampling_info()
  diffinsights_web: Add SankeyPlot to list of possible plots
  diffinsights_web: Rename data_store to timeline_data_store in contributors.py
  diffinsights_web: Add views/plots/sankey.py
  diffinsights_web: Fix handling of missing lines-stats file
  diffinsights_web: Add process_sankey()
  diffinsights_web: Handle input of None in count_file_x_line_lines_stats()
  diffinsights_web: Add sankey_plot_from_triplets() for rough preview
  diffinsights_web: Add reduce_sankey_thin_out()
  diffinsights_web: Add reduce_sankey_from_tail()
  diffinsights_web: Add add_dashdash_dirs_to_counter(), not tested
  diffinsights_web: Add path_to_dirs_only_counter()
  diffinsights_web: Add sankey_triples_from_counter() and its reverse
  diffinsights_web: Add limit_count_to_selected_files()
  ...

src/diffinsights_web/apps/contributors.py

@@ -5,12 +5,15 @@
 import panel as pn
 
 import diffinsights_web.utils.notifications as notifications
-from diffinsights_web.datastore.timeline import TimelineDataStore, find_dataset_dir
+from diffinsights_web.datastore.linesstats import LinesStatsDataStore
+from diffinsights_web.datastore.timeline import TimelineDataStore
+from diffinsights_web.datastore import find_dataset_dir
 from diffinsights_web.utils.notifications import onload_callback
 from diffinsights_web.views.authorsgrid import AuthorInfo, AuthorsGrid
 from diffinsights_web.views.dataexplorer import TimelineJSONViewer, TimelinePerspective, \
     TimelineDataFrameEnum, perspective_pane
 from diffinsights_web.views.info import ContributorsHeader, RepoPlotHeader, ContributionsPercHeader
+from diffinsights_web.views.plots.sankey import SankeyPlot
 from diffinsights_web.views.plots.timeseries import TimeseriesPlot
 from diffinsights_web.widgets.caching import ClearCacheButton
 
@@ -27,33 +30,43 @@
 pn.state.onload(onload_callback)
 
 dataset_dir = find_dataset_dir()
-data_store = TimelineDataStore(dataset_dir=dataset_dir)
+timeline_data_store = TimelineDataStore(dataset_dir=dataset_dir)
+lines_stats_data_store = LinesStatsDataStore(
+    dataset_dir=dataset_dir,
+    timeseries_file=timeline_data_store.select_file_widget,
+    repo_name=timeline_data_store.select_repo_widget,
+)
 
 page_header = ContributorsHeader(
-    repo=data_store.select_repo_widget,
-    freq=data_store.resample_frequency_widget,
-    end_date=data_store.timeline_max_date_rx,
+    repo=timeline_data_store.select_repo_widget,
+    freq=timeline_data_store.resample_frequency_widget,
+    end_date=timeline_data_store.timeline_max_date_rx,
+)
+sankey_plot = SankeyPlot(
+    data_store=lines_stats_data_store,
+    from_date_str=page_header.select_period_from_widget,
 )
 timeseries_plot = TimeseriesPlot(
-    data_store=data_store,
+    data_store=timeline_data_store,
     column_name=page_header.select_contribution_type_widget,
     from_date_str=page_header.select_period_from_widget,
+    sankey_plot=sankey_plot,
 )
 timeseries_plot_header = RepoPlotHeader(
-    freq=data_store.resample_frequency_widget,
+    freq=timeline_data_store.resample_frequency_widget,
     column_name=page_header.select_contribution_type_widget,
     plot=timeseries_plot,
 )
 contributions_perc_header = ContributionsPercHeader(
-    data_store=data_store,
+    data_store=timeline_data_store,
     from_date_str=page_header.select_period_from_widget,
 )
 authors_info_panel = AuthorInfo(
-    data_store=data_store,
+    data_store=timeline_data_store,
     authors_info_df=timeseries_plot.authors_info_df_rx,
 )
 authors_grid = AuthorsGrid(
-    data_store=data_store,
+    data_store=timeline_data_store,
     main_plot=timeseries_plot,
     authors_info_df=timeseries_plot.authors_info_df_rx,
     top_n=authors_info_panel.top_n_widget,
@@ -65,7 +78,7 @@
     title="Contributors Graph",  # TODO: make title dynamic
     favicon="favicon.svg",
     sidebar=[
-        data_store,
+        timeline_data_store,
         *authors_info_panel.widgets(),
 
         pn.layout.Divider(),  # - - - - - - - - - - - - -
@@ -88,11 +101,11 @@
         authors_grid,
     ],
 )
-timeline_perspective = TimelinePerspective(data_store=data_store)
+timeline_perspective = TimelinePerspective(data_store=timeline_data_store)
 template.main.extend([
     pn.layout.Divider(),
     pn.Tabs(
-        ('JSON', TimelineJSONViewer(data_store=data_store)),
+        ('JSON', TimelineJSONViewer(data_store=timeline_data_store)),
         ('data', timeline_perspective.panel(TimelineDataFrameEnum.TIMELINE_DATA)),
         ('resampled', timeline_perspective.panel(TimelineDataFrameEnum.RESAMPLED_DATA)),
         ('by author+resampled', timeline_perspective.panel(TimelineDataFrameEnum.BY_AUTHOR_DATA)),
@@ -101,7 +114,7 @@
             perspective_pane(
                 df=timeseries_plot.authors_info_df_rx,
                 title=pn.rx("Authors info for repo={repo!r}, from={from_date!r}") \
-                    .format(repo=data_store.select_repo_widget,
+                    .format(repo=timeline_data_store.select_repo_widget,
                             from_date=page_header.select_period_from_widget)
             )
         ),

src/diffinsights_web/datastore/__init__.py

@@ -0,0 +1,17 @@
+from pathlib import Path
+from typing import Optional
+
+import panel as pn
+
+DATASET_DIR = 'data/examples/stats'
+
+
+@pn.cache
+def find_dataset_dir() -> Optional[Path]:
+    for TOP_DIR in ['', '..', '../..']:
+        full_dir = Path(TOP_DIR).joinpath(DATASET_DIR)
+
+        if full_dir.is_dir():
+            return full_dir
+
+    return None