scverse / scverse/spatialdata-plot

Custom title when multiple coordinate system present and no coordinate system specified

Offen
#269 1 Kommentar 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen

Dieses Issue hat noch niemand übernommen.

:speaking_head: discussion enhancement priority: medium
Vorherrschende Sprache
Python
Sterne
86
Forks
21
Ø Merge
14 Std. 50 Min.
Gemergte PRs (30 T.)
3

Beschreibung

When there is a object with 2 coordinate systems and pl.show() is called without specifying coordinate_systems, by default two subplots are created. If the user adjust the title=, both titles are changed, leading to a plot that is confusing, especially if it was not clear that the object (or an element of the object) was aligned to multiple coordinate systems.

This is exactly what happened in the following plots (that appeared from a bug reported by a user).

before setting the title

Screenshot 2024-05-30 at 11 55 06

after setting the title

not sure why the order changes but no big deal, not the point of this issue
Screenshot 2024-05-30 at 11 55 11

solution for this case

Solution for the user: here the fix is to specify coordinate_systems='global', which will lead to a single plot.

proposed solution

Discussed also with @melorona. I proposed to change the default behavior as follows:

  • coordinate_systems will have a new default of None.
  • if the object has a single coordinate system and coordinate_system is None, this will be plotted
  • elif the object has multiple coordinate system and coordinate_system is None and one of them is called 'global' (defined in from spatialdata.models._utils import DEFAULT_COORDINATE_SYSTEM), this will be plotted
  • if coordinate_system is not None and coordinate_system is available, this will be plotted
  • else (i.e. if coordinate_system is None, multiple coordinate systems are present and none of them is called 'global', or if coordinate_system is not None and it's not available in the data, then an error will be raised, printing the list of available coordinate systems.

What will change for the user:

  • with this change, by default we won't plot multiple coordinate systems; if this is still the desired behavior, the user can still pass a list of coordinate systems.

Extra comments:

  • with the incoming "transformations refactoring" it will be required to specify which coordinate system to plot because elements will not contain an explicit list of coordinate systems they are aligned to but there will only be a big graph of transformations where a lot of coordinate systems can be aligned to a lot of other coordinate systems. So it will be important to plot only what the user asks for
  • in general, the current default behavior is useful when there are multiple FOVs for a dataset (like in the MIBITOF example dataset); but in other cases where there is the same data in multiple coordinate systems, like in Xenium, or where the number of FOVs is very large, like in CosMx, the currently implemented default behavior is suboptimal, and requiring the proposed new default would be handier.

Beitragsleitfaden

Beitragsleitfaden öffnen

Erste Schritte

  1. Lies das ganze Issue und danach den Beitragsleitfaden des Projekts.
  2. Schreib ins Issue, dass du es übernimmst — das erspart doppelte Arbeit.
  3. Forke das Repository und arbeite in einem Branch.
  4. Öffne einen Pull Request, der die Issue-Nummer nennt.

Rechercherichtung

Beginne beim Einstiegspunkt pl.show() und verfolge, wie die Standardwerte für coordinate_systems ausgewählt werden, wenn ein Objekt über ein oder mehrere Koordinatensysteme verfügt. Als abgeschlossen gilt die Aufgabe, wenn der Standardwert das einzige System oder das globale System auswählt, bei einem ungültigen Standardwert einen Fehler mit den verfügbaren Systemen auslöst und Aufrufern weiterhin erlaubt, explizit eine Liste zu übergeben.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Bewertung

Tech-Stack
python
Bereich
data-visualization
Issue-Typ
Feature
Schwierigkeit
4/5
Geschätzter Aufwand
3-5 Tage
Aktivitätsstatus
Veraltet
Klarheit
Größtenteils klar
Anfängerfreundlichkeit
35/100

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.