This project is a Kibana plugin that provides two visualizations:
- Enhanced Table: same than Data Table, but with enhanced features like computed columns, filter bar and pivot table.
- Document Table (NEW): same thing than 'Enhanced Table' visualization, but for single documents (not aggregations). It especially allows to have enhanced features compared to a saved search (custom column labels, custom hits size, custom pagination, computed columns and filter bar).
- Add computed columns, based on other columns :
- Computed formula based on expr-eval expressions (more documentation here)
- Support for numeric columns (ex:
col0 + col1
orcol[0] + col[1]
) - Support for string columns, including HTML (ex:
col0 > 0 ? 'OK' : 'KO'
) - Support for date columns
- Ability to reference total hits count matched by Elasticsearch query (ex:
col0 / total * 100
orcol0 / totalHits * 100
) - Ability to reference a column by its label (ex:
col['Sum of duration'] / col['Count']
) - Ability to reference a column total (ex:
col['Sales by month'] / total['Sales by month']
) - Ability to reference Kibana time range (ex:
timeRange.duration.months
) - Ability to define/reference arrays, do variable assignment and define custom functions in expressions
- Ability to compute column total using formula
- Support for numeric pretty format using Numeral.js (ex:
0,0.00
) - Support for date pretty format using Moment.js (ex:
YYYY-MM-DD
) - Support for duration pretty format using Kibana duration format (with same options than Kibana Duration format)
- Support for column alignment (ex:
left
,right
) - Support for template rendering using Handlebars (ex:
<strong>{{value}}</strong>
) - Template can reference other columns (ex:
<span style="color: {{col0}}">{{value}}</span>
) - Template can reference another column by its label (ex:
<span style="color: {{col['color']}}">{{value}}</span>
) - Template can encode a value to render it as a URL parameter (ex:
<a href="my-dashboard?param={{{encodeURIComponent value}}}">{{value}}</a>
) - Support for cell computed CSS based on a computed formula (ex:
value < 0 ? 'background-color: red' : ''
)- More documentation here
- Support for computed column filtering (Filter for/out value) if formula simply references a column value (ex:
col0
) - Ability to custom column position (to render computed column in a previous position in the table)
- Filter table rows based on a computed formula (ex:
col0 > 0
)- More documentation here
- Set dynamically row CSS based on a computed formula (ex:
col0 < 0 ? 'background-color: red' : ''
)- More documentation here
- Hide some table columns (ex:
0,1,Col2 Label
hides columns 0, 1 and the column labeled 'Col2 Label')- Note that the column label must be written as is (including whitespaces), with no surrounding quotes.
- Column labels containing commas are not supported since the comma is used to separate the columns
- It is recommended to use column labels to hide columns rather than their indices. Using the 'Split cols' feature, in fact, indices might depend on the global timerange that has been set.
- Add a filter bar (ex: when user enters
cat
filter, it will display only rows that contain 'cat')- Works also with numeric and date columns
- Ability to enable case sensitive filter
- Ability to make filter bar hideable
- Ability to filter as you type
- Ability to filter each term separately
- Ability to highlight filter results
- Ability to define filter bar width
- Support for a new bucket type : 'Split Cols'. It lets to create a pivot table
- When combined with computed columns, each computed column can be added per split column or after all split columns
- Ability to sort split columns (by split column value)
- Hide CSV export links (when checked, it will hide 'Raw' and 'Formatted' export links)
- Change CSV export encoding (ex:
iso-8859-1
) - Full CSV export: in Document Table vis, ability to download full data in Elasticsearch, not only displayed data in the table (Kibana 7.7+)
- Add total row to CSV export
- Add a total label on total row first column (ex:
Total:
) - Display striped rows
- Add a row number column
- Ability to add the visualization to a Canvas workpad (Kibana 7.9+)
- Ability to use dashboard drilldowns (Kibana 7.9+)
- Kibana supported versions: all versions from 5.5 to 8.15
- OpenSearch Dashboards supported versions : all versions from 1.x to 2.x
Every release package includes a Plugin version (X.Y.Z) and a Kibana version (A.B.C).
- Go to releases and choose the right one for your Kibana
- launch a shell terminal and go to $KIBANA_HOME folder
- use Kibana CLI to install :
- directly from Internet URL :
./bin/kibana-plugin install https://github.com/fbaligand/kibana-enhanced-table/releases/download/vX.Y.Z/enhanced-table-X.Y.Z_A.B.C.zip
- locally after manual download :
./bin/kibana-plugin install file:https:///path/to/enhanced-table-X.Y.Z_A.B.C.zip
- directly from Internet URL :
- restart Kibana
- Open Kibana URL in your browser (by default: https://localhost:5601)
- Go to "Visualize" app
- Click on "Create visualization" button
- If you use Kibana 7.11 or superior, click on "Aggregation based"
- If you want a table based on aggregated data:
- Choose "Enhanced Table"
- Select your index-pattern, and then your buckets & metrics
- Go to "Options" tab, and enjoy the enhanced features: computed columns, hidden columns, filter bar, and so on...
- If you want a table based on single documents:
- Choose "Document Table"
- Select your index-pattern
- Then in 'Data' tab, define columns to display and indicate "Hits size"
- Go to "Options" tab, and enjoy the enhanced features: computed columns, hidden columns, filter bar, and so on...
If you don't see 'Enhanced Table' (when you go to "Create visualization" screen) or if you see any error, try these actions:
- in your browser, force Kibana page reload: Shift+F5 or Ctrl+F5
- if it still doesn't work, empty your browser cache (using Ctrl+Shift+Del), and then, reload Kibana page
- if it still doesn't work:
- stop Kibana
- delete
$KIBANA_HOME/optimize
folder - start Kibana
- empty again your browser cache
- reload Kibana page in your browser
This is the common documentation for all computed settings:
- Computed Column Formula
- Cell Computed CSS
- Rows Computed Filter
- Rows Computed CSS
- Support for expr-eval expressions
- Support for features brought by expr-eval 2.0:
- Ability to reference arrays. Especially useful to reference a 'Top Hits' metric column:
col1[0]
- New functions for arrays available:
join, map, filter
- Variable assignment:
x = 4
- Custom function definitions:
myfunction(x, y) = x * y
- Evaluate multiple expressions by separating them with
;
- Ability to reference arrays. Especially useful to reference a 'Top Hits' metric column:
- Formula validation, with error notification
col0, col1, ..., colN
: value of a previous column, referenced by its index (0-based index)col['COLUMN_LABEL']
: value of a previous column, referenced by its labelformmattedCol, formmattedCol1, ..., colN
: formatted value of a previous column, referenced by its index (0-based index)formattedCol['COLUMN_LABEL']
: formatted value of a previous column, referenced by its labeltotal0, total1, ..., totalN
: total of a previous column, referenced by its index (0-based index)total['COLUMN_LABEL']
: total of a previous column, referenced by its labeltotal
,totalHits
: total hits count matched by Elasticsearch query (given search bar & filter bar)value
: value of current computed column (only available in "Cell computed CSS" feature)timeRange
: informations about current time range, selected in Kibana time pickerduration
: object containing time range duration, in different unitsyears
: years count in time range (rounded up to the nearest whole number)months
: months count in time range (rounded up to the nearest whole number)weeks
: weeks count in time range (rounded up to the nearest whole number)days
: days count in time range (rounded up to the nearest whole number)hours
: hours count in time range (rounded up to the nearest whole number)minutes
: minutes count in time range (rounded up to the nearest whole number)seconds
: seconds count in time range (rounded up to the nearest whole number)milliseconds
: milliseconds count in time range
from
/to
: object containing all informations onfrom
andto
dates of current time rangefullYear
: result of Date.getFullYear()month
: result of Date.getMonth()date
: result of Date.getDate()day
: result of Date.getDay()hours
: result of Date.getHours()minutes
: result of Date.getMinutes()seconds
: result of Date.getSeconds()milliseconds
: result of Date.getMilliseconds()time
: result of Date.getTime()timezoneOffset
: result of Date.getTimezoneOffset()dateString
: result of Date.toDateString()isoString
: result of Date.toISOString()localeDateString
: result of Date.toLocaleDateString()localeString
: result of Date.toLocaleString()localeTimeString
: result of Date.toLocaleTimeString()string
: result of Date.toString()timeString
: result of Date.toTimeString()utcString
: result of Date.toUTCString()
- All pre-defined functions provided by expr-eval
- Additional custom functions listed in table below (ex:
col['Expiration Date'] > now() ? 'OK' : 'KO'
)
Function | Description |
---|---|
cell(rowRef, colRef, defaultValue) | Returns table cell value referenced by rowRef and colRef (if it exists), or else defaultValue . rowRef is either 'first' (for first row), 'last' (for last row) or a number that is the relative target row position compared to current row (ex: -1 means the previous row). colRef is either the column label (ex: 'Count' ) or the column index (ex: 1 ). |
col(colRef, defaultValue) | Returns column value referenced by colRef (if it exists), or else defaultValue . colRef is either the column label (ex: 'Count' ) or the column index (ex: 1 ). |
countSplitCols() | Returns the count of all split columns (only if 'Split cols' bucket is used). |
dateObject(params) | Given standard Date constructor params (milliseconds since Epoch, ...), builds and returns a date object, same structure than timeRange from/to object. The result can be used in template (ex: {{ rawValue.fullYear }} ). |
durationObject(durationInMillis) | Given a duration in milliseconds, builds and returns a duration object, that breaks down the duration in years, months, weeks, days, hours, minutes, seconds and milliseconds. The result can be used in template (ex: {{ rawValue.hours }} ). |
encodeURIComponent(str) | Encodes the provided string as a Uniform Resource Identifier (URI) component. |
formattedCell(rowRef, colRef, defaultValue) | Returns formatted table cell value referenced by rowRef and colRef (if it exists), or else defaultValue . rowRef is either 'first' (for first row), 'last' (for last row) or a number that is the relative target row position compared to current row (ex: -1 means the previous row). colRef is either the column label (ex: 'Count' ) or the column index (ex: 1 ). |
formattedCol(colRef, defaultValue) | Returns formatted column value referenced by colRef (if it exists), or else defaultValue . colRef is either the column label (ex: 'Count' ) or the column index (ex: 1 ). |
indexOf(strOrArray, searchValue[, fromIndex]) | Returns the index within the calling String or Array object of the first occurrence of the specified value, starting the search at fromIndex. Returns -1 if the value is not found. |
isArray(value) | Determines whether the passed value is an Array. |
lastIndexOf(strOrArray, searchValue[, fromIndex]) | Returns the index within the calling String or Array object of the last occurrence of the specified value, searching backwards from fromIndex. Returns -1 if the value is not found. |
now() | Returns the number of milliseconds elapsed since January 1, 1970 00:00:00 UTC |
parseDate(dateString) | Returns the number of milliseconds elapsed since January 1, 1970 00:00:00 UTC and the date obtained by parsing the given string representation of a date. If the argument doesn't represent a valid date, NaN is returned. Useful to parse date columns in 'Document Table' visualization. |
replace(str, substr, replacement) | Returns a new string with first match of substr replaced by a replacement. Only the first occurrence will be replaced. |
replaceRegexp(str, regexp, replacement) | Returns a new string with all matches of a regexp replaced by a replacement. All the occurrences will be replaced. |
search(str, regexp) | Executes a search for a match between a regular expression on 'str' String. Returns the index of the first match or -1 if not found. |
sort(array[, compareFunction]) | Sorts the elements of an array in place and returns the sorted array. A compare function can be provided to customize the sort order. Example for an array of numbers: comparator(a, b) = a - b; sort(col0, comparator) |
substring(str, indexStart[, indexEnd]) | Returns the part of the string between the start and end indexes, or to the end of the string (if no index end is provided). |
sumSplitCols() | Returns the sum of all split column values (only if 'Split cols' bucket is used). |
toLowerCase(str) | Returns the calling string value converted to lowercase. |
toUpperCase(str) | Returns the calling string value converted to uppercase. |
total(colRef, defaultValue) | Returns column total referenced by colRef (if it exists), or else defaultValue . colRef is either the column label (ex: 'Count' ) or the column index (ex: 1 ). |
trim(str) | Removes whitespace from both ends of a string. |
uniq(array) | Removes duplicates from provided array so that array contains only unique values. |
This is the documentation for "Template" setting in computed columns.
A template is a Handlebars expression.
Examples:
<strong>{{value}} items</strong>
{{timeRange.from.localeDateString}} - {{timeRange.to.localeDateString}}
<a href="my-dashboard?param={{{encodeURIComponent rawValue}}}">{{value}}</a>
{{#if rawValue}}OK{{else}}KO{{/if}}
col0, col1, ..., colN
: raw value of a previous column, referenced by its index (0-based index)col['COLUMN_LABEL']
: raw value of a previous column, referenced by its labelformmattedCol, formmattedCol1, ..., colN
: formatted value of a previous column, referenced by its index (0-based index)formattedCol['COLUMN_LABEL']
: formatted value of a previous column, referenced by its labeltotal0, total1, ..., totalN
: total of a previous column, referenced by its index (0-based index)total['COLUMN_LABEL']
: total of a previous column, referenced by its labeltotal
,totalHits
: total hits count matched by Elasticsearch query (given search bar & filter bar)value
: value of current computed column, formatted using "Format" settingrawValue
: value of current computed column, not formattedtimeRange
: informations about current time range, selected in Kibana time pickerduration
: object containing time range duration, in different unitsyears
: years count in time range (rounded up to the nearest whole number)months
: months count in time range (rounded up to the nearest whole number)weeks
: weeks count in time range (rounded up to the nearest whole number)days
: days count in time range (rounded up to the nearest whole number)hours
: hours count in time range (rounded up to the nearest whole number)minutes
: minutes count in time range (rounded up to the nearest whole number)seconds
: seconds count in time range (rounded up to the nearest whole number)milliseconds
: milliseconds count in time range
from
/to
: object containing all informations onfrom
andto
dates of current time rangefullYear
: result of Date.getFullYear()month
: result of Date.getMonth()date
: result of Date.getDate()day
: result of Date.getDay()hours
: result of Date.getHours()minutes
: result of Date.getMinutes()seconds
: result of Date.getSeconds()milliseconds
: result of Date.getMilliseconds()time
: result of Date.getTime()timezoneOffset
: result of Date.getTimezoneOffset()dateString
: result of Date.toDateString()isoString
: result of Date.toISOString()localeDateString
: result of Date.toLocaleDateString()localeString
: result of Date.toLocaleString()localeTimeString
: result of Date.toLocaleTimeString()string
: result of Date.toString()timeString
: result of Date.toTimeString()utcString
: result of Date.toUTCString()
- All pre-defined helpers provided by Handlebars
- Additional custom helpers listed in table below.
Helper | Description |
---|---|
encodeURIComponent(str) | Encodes the provided string as a Uniform Resource Identifier (URI) component. Example: <a href="my-dashboard?param={{{encodeURIComponent rawValue}}}">{{value}}</a> |
Versions and Release Notes are listed in Releases page
This Kibana plugin is inspired from computed-columns and kbn_searchtables plugins.
Thanks for their great work !
To run enhanced-table plugin in development mode (that enables hot code reload), follow these instructions:
- execute these commands :
git clone --depth 1 -b X.Y https://github.com/elastic/kibana.git # replace 'X.Y' by desired Kibana version
cd kibana/plugins
git clone https://github.com/fbaligand/kibana-enhanced-table.git enhancedTable
cd ..
- install the version of Node.js noted in
kibana/.node-version
file (for instance usingnvm use
command) - ensure that node binary directory is in PATH environment variable
- install the latest version of yarn:
npm install -g yarn
- execute these commands (starting from 'kibana' directory) :
yarn kbn bootstrap
cd plugins/enhancedTable
export NODE_OPTIONS="--openssl-legacy-provider"
yarn install
yarn compile
yarn start
- in your browser, call
https://localhost:5601
and enjoy!
To build a distributable archive, execute this command :
yarn compile-and-build --kibana-version X.Y.Z # replace 'X.Y.Z' by target Kibana version
The zip archive is generated into build
directory.
If this plugin helps you and you want to support it, you can give me a cup of coffee :)