Flex-table-card

[kbrown01]

To answer my own question (well at least half of it), I can do cells like this:

type: custom:flex-table-card
title: Russian River Level
entities:
  include: sensor.russian_river_flood_data
columns:
  - name: Date/Time
    data: flooddata
    modify: x.valid
  - name: Level
    data: flooddata
    modify: x.level
    align: right
  - name: Trend
    data: flooddata
    modify: x.trend
    align: center
  - name: Status
    data: flooddata
    modify: |-
      switch(x.status){
        case 'Above Monitor':
          '<div style="background-color: #FFE033;">' + x.status + '</div>';
          break;
        case 'Above Flood':
          '<div style="background-color: #F08080;">' + x.status + '</div>';
          break;
        default:
          x.status;
        }
  - name: Mode
    data: flooddata
    modify: x.mode

[kbrown01]

And to re-answer my question with enhancements:

type: custom:flex-table-card
entities:
  include: sensor.russian_river_flood_info
columns:
  - name: Date/Time
    data: flooddata
    modify: x.valid
  - name: Level
    data: flooddata
    modify: |-
      switch(x.status){
        case 'Above Monitor':
          '<div style="background-color:  #e67300;">' + parseFloat(x.level).toFixed(1) + '&nbsp;ft</div>';
          break;
        case 'Above Flood':
          '<div style="background-color: #cc0000;">' + parseFloat(x.level).toFixed(1) + '&nbsp;ft</div>';
          break;
        default:
          parseFloat(x.level).toFixed(1) + '&nbsp;ft';
        }
    align: center
  - name: Trend
    data: flooddata
    modify: |-
      switch(x.trend){
        case '+':
          '<ha-icon icon="mdi:arrow-up" style="color:  #e67300;">';
          break;
        case '-':
          '<ha-icon icon="mdi:arrow-down" style="color:  #3399ff;">';
          break;
        default:
          '<ha-icon icon="mdi:arrow-all">';
        }
    align: center

Now, move onto a better solution than scrolling the whole table which is terrible. The scrolling posted above in the author’s samples will not suffice. A table should scroll the body and not the whole table. Should be able to do that in pure CSS

[EdwardTFN (Edward Firmo)]

I’m also using the modify to deal with colors per cell, but in my case the color to be used is part of the data:

type: custom:flex-table-card
title: Västtrafik - Svingeln
strict: true
max_rows: 20
sort_by:
  - Departure+
entities:
  include: sensor.vasttrafik_departure_table
columns:
  - name: Sort column
    data: Departure
    align: center
    hidden: true
    modify: >-
      Math.round((Date.parse((x.rtDate ? x.rtDate : x.Date) + " " + (x.rtTime ?
      x.rtTime : x.Time)) - Date.now())/60000)
  - name: Line
    data: Departure
    modify: x.sname
    align: center
    hidden: true
    style: '''background-color: '' + x.bgColor + '';'' + ''color: '' + x.fgColor + '';'''
  - name: Line
    data: Departure
    modify: >-
      '<div style="padding: 1px 1px 4px 1px; height: 16px; color: ' + x.fgColor
      + '; background-color: ' + x.bgColor + '">&nbsp;' + x.sname +
      '&nbsp;</div>'
    align: center
  - name: Direction
    data: Departure
    modify: x.direction
  - name: Departure
    data: Departure
    align: center
    modify: |-
      if (typeof(x.rtTime) == "string" & typeof(x.rtDate) == "string") 
        { 
          var minutes = Math.round(Math.max(0,(Date.parse(x.rtDate + " " + x.rtTime) - Date.now())/60000));
          if (minutes>=60 || isNaN(minutes))
            {{ x.rtTime }}
          else if (minutes == 0)
            "Now"
          else
            {{ minutes }}
        }
      else
        { 
          var minutes = Math.round(Math.max(0,(Date.parse(x.date + " " + x.time) - Date.now())/60000));
          if (minutes>=60 || isNaN(minutes))
            {{ x.time }}
          else if (minutes == 0)
            "Now"
          else
            {{ minutes }}
        }
  - name: Platform
    data: Departure
    align: center
    modify: >-
      x.track + '&nbsp;<ha-icon icon="mdi:' +
      x.type.toLowerCase().replace("tram", "tram-side") +  '"></ha-icon>'
css:
  tbody tr:nth-child(odd): 'background-color: #535B64;'
  tbody tr:nth-child(even): 'background-color: #3C4650;'
style: |
  ha-card {
    --mdc-icon-size: 16px;
    background-color: #3C4650;
  }

[kbrown01]

@Ildar_Gabdullin … please take a look here and tell me where I am wrong.

Real-Time Sports Scores w/ TeamTracker and TeamTracker-Card (Beta)

Ok so if I do this… - platform: rest name: MLS_Standings_test scan_interval: 7200 resource: https://site.web.api.espn.com/apis/v2/sports/soccer/usa.1/standings?type=0&level=0 value_template: "{{ now() }}" json_attributes: - children Looks like this pulls all MLS teams broken up by conferences…but now how do I pull the data into the flex-table? children: - uid: s:600~l:770~g:1 id: '1' name: Eastern Conference abbreviation: East standings: id…

I am trying to use flex-table to grab stats that are multiple layers deep. I do not see how this is possible as to me, nothing beyond data: myarray works. data: myarray.myarray does not nor anything I have tried to iterate over things.

[Ildar_Gabdullin]

Split data into 2 sensors (western & eastern conferences):

template:
  - sensor:
      - name: test_sport_eastern_conf
        state: true  ###could be anything
        attributes:
          standings: >-
            {{ state_attr('sensor.mls_standings_test','children')[0].standings.entries }}

      - name: test_sport_western_conf
        state: true
        attributes:
          standings: >-
            {{ state_attr('sensor.mls_standings_test','children')[1].standings.entries }}

Then create smth like this for each conference:

type: custom:flex-table-card
title: Standings
entities:
  include: sensor.test_sport_western_conf
columns:
  - name: team
    data: standings
    modify: x.team.name
  - name: Games Played
    data: standings
    modify: |
      x.stats[0].displayValue
  - name: Losses
    data: standings
    modify: |
      x.stats[1].displayValue
  - name: Points
    data: standings
    modify: |
      x.stats[2].displayValue
  - name: Goals Against
    data: standings
    modify: |
      x.stats[3].displayValue
  - name: Goals For
    data: standings
    modify: |
      x.stats[4].displayValue
  - name: Draws
    data: standings
    modify: |
      x.stats[5].displayValue
  - name: Wins
    data: standings
    modify: |
      x.stats[6].displayValue
  - name: Point Deductions
    data: standings
    modify: |
      x.stats[7].value
  - name: Goal Difference
    data: standings
    modify: |
      x.stats[8].displayValue
  - name: Points Per Game
    data: standings
    modify: |
      x.stats[9].displayValue
  - name: Rank
    data: standings
    modify: |
      x.stats[10].displayValue
  - name: Rank Change
    data: standings
    modify: |
      x.stats[11].displayValue
  - name: Overall
    data: standings
    modify: |
      x.stats[12].displayValue

This is just an example.
Hope this will help you.

[kbrown01]

I guess I will live with the four sensor, four card approach …

Results:

[kbrown01]

Is is possible to use a column in which the data uses a “modify” as a “sort_by”? I have tried but it appears it is not possible. I was trying to extend the four sensors into a different view that combines them back into one card with all teams. I get the data, but I cannot sort this:

entities:
  include:
    - sensor.nhl_east_atlantic
    - sensor.nhl_east_metropolitan
    - sensor.nhl_west_central
    - sensor.nhl_west_pacific
sort_by: x.stats[6].displayValue+
columns:
  - name: Team
    data: entries
    modify: x.team.displayName
  - name: GP
    data: entries
    modify: x.stats[3].displayValue
  - name: W
    data: entries
    modify: x.stats[10].displayValue
  - name: L
    data: entries
    modify: x.stats[4].displayValue
  - name: OTL
    data: entries
    modify: x.stats[11].displayValue
  - name: PTS
    data: entries
    modify: x.stats[6].displayValue

The sort appears as if it can only take the data field and that is generic in this case because it uses modify to grab the exact data needed,

[EdwardTFN (Edward Firmo)]

When more than 1 column have the same source, it will sort by the first one, so in order to get to what you want, copy that column to the first column and then add hidden: true, so the column will be there, will be used for sorting, but won’t be visible.

Something like this:

entities:
  include:
    - sensor.nhl_east_atlantic
    - sensor.nhl_east_metropolitan
    - sensor.nhl_west_central
    - sensor.nhl_west_pacific
sort_by:
  - entries
columns:
  - hidden: true
    data: entries
    modify: x.stats[6].displayValue
  - name: Team
    data: entries
    modify: x.team.displayName
  - name: GP
    data: entries
    modify: x.stats[3].displayValue
  - name: W
    data: entries
    modify: x.stats[10].displayValue
  - name: L
    data: entries
    modify: x.stats[4].displayValue
  - name: OTL
    data: entries
    modify: x.stats[11].displayValue
  - name: PTS
    data: entries
    modify: x.stats[6].displayValue

[kbrown01]

Tried that, no bueno:

Even with the column not hidden it doesn;t work I can tell that even in the original because it would assume that the sort would be team name. Tried you sample too with no name, none of those work. Nothing sorts as far as I can see unless it is sorted by a direct attribute without “modify”

  - hidden: true
    data: entries
    modify: x.stats[6].displayValue

Now of course I could create a new sensor but no. It already is bad enough that NHL needs 5 sensors (and NFL 7, and MLB 7) just to have conference and division breakdowns.

EDIT: I did not put sort_by: entries!!!

That works!

That is huge. So now for my major sports card, I can have Divisional, Conference and Overall stats. Then onto what logic to apply for possibly implementing Wild Card

[Ildar_Gabdullin]

@kbrown01

did not put sort_by: entries

entries is a list, each element is a dictionary, isn’t it?
Then how this “sort_by” is supposed to work?
What you should have done and you did not - specifying an ID for a particular column and selecting this ID for sorting.

[kbrown01]

Exacty. Understood. At first I had tried with the wrong “sort_by” and thught it over. The missing link is from @EdwardTFN. I would not have “guessed” that given a frankly not-sortable entry yet “sort_by” is provided , it sorts by the first column

Right now working perfectly for Divisional, Conference and Overall. Just need to think through the logic of wildcard.

Division:

Conference:

Overall:

[Ildar_Gabdullin]

Nice)
Hope your home has become much smarter with this sport info)))
Seriously, so far I have found only one application in my HA for this card - some tech info about network switches, that is all…

[kbrown01]

LOL!

I have a simplistic view. I hate going to other Apps/Websites. I like in all in one thing. Especially one that does not track everything I do.

Flex table has help me with River Flood status and now Sports scores and standings.
All lights, music, tv remotes, internet status, cell phones, calendars … and yes even Pokemon for the kids.

I post most all the results and YAML and such on GIT for others.

[Ildar_Gabdullin]

Probably the main reason of not using table in my setup is that for almost each sensor I have Entities card + graphs + buttons…
It will be great if you share some ideas of using this card.

For me - even only screenshots with description will be fine, w/o code.

[Plouf34]

How to reduce decimal accuracy as stated in other cards and the new HA version 2023.3.0?

[Plouf34]

Do you know how to decrease decimal accuracy for my senros in this card.

type: custom:flex-table-card
sort_by: state-
strict: true
max_rows: 10
clickable: true
entities:
  include:
    - sensor.conso_*_aujourdhui
    - sensor.pool_pump_energy_daily
    - sensor.polaris_energy_daily
    - sensor.all_lights_energy_daily
  exclude:
    - sensor.conso_pac_piscine_aujourdhui
    - sensor.atome_daily
    - sensor.conso_prod_solaire_aujourdhui
    - sensor.conso_zlinky_today
    - sensor.conso_zlinky_aujourdhui
    - sensor.spot_piscine_electric_consumption_kwh
columns:
  - data: friendly_name
    align: left
    name: Consommation Aujourd'hui
  - name: kW
    data: state
    align: center
    suffix: ' kW'
css:
  tbody tr:nth-child(1): 'color: #ff9800'
card_mod:
  style:
    .: |
      ha-card ha-markdown {
        padding:0px
      }
      ha-card ha-markdown.no-header {
        padding:0px
      }
    ha-markdown:
      $:
        ha-markdown-element: |
          h1 {
              font-weight: normal;
              font-size: 24px;
          }
          div {
              background-color:rgb(100, 100, 100);
              padding: 12px 12px;
              color:white;
              font-weight:normal;
              font-size:1.2em;
              border-top-left-radius: 5px; 
              border-top-right-radius: 5px; 
          }
          table{
            border-collapse: collapse;
            font-size: 0.9em;
            font-family: Roboto;
            width: auto;
            outline: 0px solid #393c3d;
            margin-top: 10px;
          } caption {
              text-align: center;
              font-weight: bold;
              font-size: 1.2em;
          } td {
              padding: 5px 5px 5px 5px;
              text-align: left;
              border-bottom: 0px solid #1c2020;
          }
          tr {
              border-bottom: 0px solid #1c2020;
          }
          tr:nth-of-type(even) {
              background-color: rgb(54, 54, 54, 0.3);
          }
          tr:last-of-type {
              border-bottom: transparent;
          }
          mark {
              background: #009767;
              color: #222627;
              border-radius: 10px;
              padding: 5px;
          }
          span {
              background: #EC4B34;
              color: #222627;
              border-radius: 10px;
              padding: 5px;
          }
          span {
              padding: 5px;
          }
          tr:nth-child(n+2) > td:nth-child(2) {
            text-align: left;
          }

For info, with new HA release 2023.3, we can change it, but it doesn’t work for this card.

[Ildar_Gabdullin]

Imho, there is TOO much hype & expectations about this “set precision” feature.
May be I am a special person - but I do not use this feature at all.
In many integrations (template, snmp, …) a user already may define any precision.
And this “set precision” feature only affects Frontend - internally states are intact (as it should be).
Again imho - cards should show “raw” intact data, not damaged by any next untested Frontend changes.
Check Github issues - currently people have problems with displaying data with changed precision.
If you really need to change a precision for some data in your flex-table - I recommend to use “modify”.

P.S. Please avoid posting a large code to demonstrate ONE issue.
MWE

[Plouf34]

Thanks
How to make that with « modify »?

[Ildar_Gabdullin]

  - type: custom:flex-table-card
    entities:
      include:
        - input_number.test_number_very_small
    columns:
      - name: value
        data: state
      - name: value .0
        data: state
        modify: Number(x).toFixed(0)
      - name: value .x
        data: state
        modify: Number(x).toFixed(1)
      - name: value .xx
        data: state
        modify: Number(x).toFixed(2)
      - name: value .xxx
        data: state
        modify: Number(x).toFixed(3)
      - name: value .xxxx
        data: state
        modify: Number(x).toFixed(4)
      - name: value .xxxxx
        data: state
        modify: Number(x).toFixed(5)
      - name: value .xxxxxx
        data: state
        modify: Number(x).toFixed(6)

[Plouf34]

Thank you very much