style: Wrap chart titles again

[rusackas]

SUMMARY

This PR re-enables line-wrapping. Truncation might be re-enabled as a feature flag or something for those who prefer it, in a subsequent PR.

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

Before:

After:

TEST PLAN

ADDITIONAL INFORMATION

• Has associated issue:
• Changes UI
• Requires DB Migration.
• Confirm DB Migration upgrade and downgrade tested.
• Introduces new feature or API
• Removes existing feature or API

[codecov-io]

Codecov Report

Merging #11602 into master will decrease coverage by 5.12%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #11602      +/-   ##
==========================================
- Coverage   66.57%   61.45%   -5.13%     
==========================================
  Files         876      420     -456     
  Lines       42179    26193   -15986     
  Branches     3954        0    -3954     
==========================================
- Hits        28082    16096   -11986     
+ Misses      13994    10097    -3897     
+ Partials      103        0     -103     

Flag	Coverage Δ
cypress	?
javascript	?
python	61.45% <ø> (-0.71%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
superset/db_engines/hive.py	0.00% <0.00%> (-85.72%)	⬇️
superset/db_engine_specs/hive.py	53.90% <0.00%> (-30.08%)	⬇️
superset/db_engine_specs/presto.py	69.76% <0.00%> (-12.53%)	⬇️
superset/db_engine_specs/mysql.py	79.59% <0.00%> (-12.25%)	⬇️
superset/examples/world_bank.py	97.10% <0.00%> (-2.90%)	⬇️
superset/examples/birth_names.py	96.51% <0.00%> (-2.33%)	⬇️
superset/views/database/mixins.py	80.70% <0.00%> (-1.76%)	⬇️
superset/models/core.py	87.60% <0.00%> (-1.66%)	⬇️
superset/views/core.py	73.77% <0.00%> (-0.49%)	⬇️
superset/connectors/sqla/models.py	90.36% <0.00%> (-0.27%)	⬇️
... and 456 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1490f30...fb38c43. Read the comment docs.

[ktmud]

Could be a dashboard/slice level config, too!

[junlincc]

LGTM, but again, from a product stand point, i don't think wrapping long chart title in dashboard is the best visual solution nor the best user behavior we want to encourage.if that makes sense to the users, we are alright to bring the long title back.

[ktmud]

Another idea is to use dynamic font size based on text length. We already do that for some visualizations and there is a utility function in superset-ui readily available to use.

Personally I feel it makes more sense to wrap the title instead of truncating it because users put in long titles for a reason---if they really feel it looks bad, they would choose a shorter title already.

[junlincc]

Another idea is to use dynamic font size based on text length. We already do that for some visualizations and there is a utility function in superset-ui readily available to use.

Personally I feel it makes more sense to wrap the title instead of truncating it because users put in long titles for a reason---if they really feel it looks bad, they would choose a shorter title already.

Our original design/implementation is to have tooltip showed while user hover truncated title. so that user can view a cleaner dashboard without seeing overloaded information yet being able to access information as they need. My understanding is user have to put in long time is because

1. to include details for specific chart
2. there are no other easy places to store or display these detailed information
   We certainly shouldn't ask user to cut the necessary details associate with a chart, but we can certainly come up w better solution for the second point.
   Design input and user testing are needed for this topic. though this is a small change in the entire product, we try not to make product decision base on individual preference or a single use case.

[graceguo-supercat]

I think offer the ability to show full title is very important. Chart title is part of data accuracy. If user prefer single line title to descriptive title, they already have other choices.

Here is a real example (before vs after we release new filter status)

[suddjian]

I think offer the ability to show full title is very important.

The full title is shown on hover right now via a title attribute.

[graceguo-supercat]

I think offer the ability to show full title is very important.

The full title is shown on hover right now via a title attribute.

Force user to hover chart one by one?
If the design is always asking user to do extra instead of naturally display it...probably not a good enough.

[rusackas]

@graceguo-supercat this PR gives you exactly what you're asking for... is there a reason not to approve it?

[graceguo-supercat]

@graceguo-supercat this PR gives you exactly what you're asking for... is there a reason not to approve it?

haha No. Sorry i just forgot :)

Thanks for make it happen!!

[graceguo-supercat]

LGTM

[suddjian]

Junlin's outline of the finer points of this design challenge is very insightful. We should look at better ways to address the true underlying user needs driving this discussion. But for now I think removing title truncation makes sense.

[suddjian]

Fixes #11581

[rusackas]

Obviously this is one contentious line of CSS! Some of us like shorter chart headers... some of us don't. I can see a few ways to tackle things like it in the future:

1. More feature flags... though many of us are not fans of adding a bunch of feature flags for tiny CSS changes, the option exists
2. Adding a custom style sheet somewhere - think of it as a config file, where you can just override some things like this. This could get fragile quickly, so it would be a "use at your own risk" thing.
3. Adding a user preferences area... though not all that different from feature flags, at least the user gets the choice.
4. Bringing this sort of thing back to the community as DISCUSS and VOTE threads. We have a governance model for a good reason :)

The important thing (to me anyway) is that Superset is going to have a lot of changes... we need to get better at:
• Considering design proposals before approving, so we don't have to backtrack later
• Surfacing feedback from users in a constructive way, as early as possible
• Iterating on features when we can, to prevent big reverts over small details.
• Staying friends through all of this! ❤️ All of your hard work and input is valued here. Thanks, everyone.

... but again, much ado about just one line... merging!

[graceguo-supercat]

Before implementation, a design document which is clear, accurate, highlight all the changes, public accessible to open source community, will help us.

[mistercrunch]

Before implementation, a design document which is clear, accurate, highlight all the changes, public accessible to open source community, will help us.

It's quite a luxury and ideal to have that when we can, and we're committed to get better at this. But realistically won't always be possible depending on the work at hand. Personally what I think is important is the ability to quickly get to a state that a majority is happy with, and sometimes it'll require some back and forth. It's hard to know ahead of time which areas might be contentious too.