Decision tables
Evaluate business rules written as decision tables, as in DMN 1.4: one column per input, one per output, one row per rule, with the hit policies UNIQUE, ANY, FIRST, PRIORITY, COLLECT (with SUM, MIN, MAX or COUNT), RULE ORDER and OUTPUT ORDER. rules eval gives the result of one case, rules eval_rows adds the outputs to every row of a table, and rules explain shows which entry of each rule matched or failed and why. Typical uses: a discount or pricing grid, an eligibility or approval check, the routing of a request to the right team.
rules evalrules evalrules eval_rowsrules eval_rowsrules explainrules explain

rules eval <table> <input>


Description

    The result of a decision table for one case: the outputs as a JSON object for UNIQUE, ANY, FIRST and PRIORITY, a JSON array of output objects for COLLECT, RULE ORDER and OUTPUT ORDER, and {output: value} for an aggregation

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    input:   The values of one case, as a JSON object {input name: value} - String - required
admin
rules eval "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "{\"customer\": \"Business\", \"orderSize\": 12}"
mentdb
{"discount": 0.15}
No rule matches: the default
admin
rules eval "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "{\"customer\": \"Business\"}"
mentdb
{"discount": 0}
FIRST: the first matching rule, two outputs
admin
rules eval "{\"hit_policy\": \"FIRST\", \"inputs\": [{\"name\": \"country\", \"type\": \"string\"}, {\"name\": \"weight\", \"type\": \"number\"}, {\"name\": \"express\", \"type\": \"boolean\"}], \"outputs\": [{\"name\": \"carrier\", \"type\": \"string\"}, {\"name\": \"price\", \"type\": \"number\"}], \"rules\": [[\"\\"FR\\"\", \"<=2\", \"false\", \"La Poste\", 4.9], [\"\\"FR\\", \\"BE\\", \\"LU\\"\", \"(2..30]\", \"-\", \"Colissimo\", 12.5], [\"not(\\"FR\\")\", \"<=30\", \"true\", \"DHL Express\", 39], [\"-\", \"-\", \"-\", \"Freight\", 120]]}" "{\"country\": \"BE\", \"weight\": 4.2, \"express\": false}"
mentdb
{"carrier": "Colissimo", "price": 12.5}
PRIORITY: DECLINE comes before REFER and ACCEPT
admin
rules eval "{\"hit_policy\": \"PRIORITY\", \"inputs\": [{\"name\": \"age\", \"type\": \"number\", \"integer\": true, \"values\": \"[18..120]\"}, {\"name\": \"score\", \"type\": \"number\"}, {\"name\": \"since\", \"type\": \"date\"}], \"outputs\": [{\"name\": \"decision\", \"type\": \"string\", \"values\": [\"DECLINE\", \"REFER\", \"ACCEPT\"]}], \"rules\": [[\"-\", \">=700\", \"-\", \"ACCEPT\"], [\"-\", \"[600..700)\", \"-\", \"REFER\"], [\"-\", \"<600\", \"-\", \"DECLINE\"], [\"<21\", \"-\", \"-\", \"REFER\"], [\"-\", \"-\", \">date(\\"2026-01-01\\")\", \"REFER\"]]}" "{\"age\": 30, \"score\": 580, \"since\": \"2026-03-01\"}"
mentdb
{"decision": "DECLINE"}
COLLECT SUM: loyalty points
admin
rules eval "{\"hit_policy\": \"COLLECT\", \"aggregation\": \"SUM\", \"inputs\": [\"basket:number\", \"member:boolean\"], \"outputs\": [\"points:number\"], \"rules\": [[\">=50\", \"-\", 5], [\">=100\", \"-\", 10], [\"-\", \"true\", 2]]}" "{\"basket\": 120, \"member\": true}"
mentdb
{"points": 17}
RULE ORDER: every approver
admin
rules eval "{\"hit_policy\": \"RULE ORDER\", \"inputs\": [\"amount:number\", \"category:string\"], \"outputs\": [\"approver:string\"], \"rules\": [[\">1000\", \"-\", \"manager\"], [\">10000\", \"-\", \"director\"], [\"-\", \"\\"IT\\"\", \"it_lead\"]]}" "{\"amount\": 25000, \"category\": \"IT\"}"
mentdb
[{"approver": "manager"}, {"approver": "director"}, {"approver": "it_lead"}]
UNIQUE: two rules match, an error
admin
rules eval "{\"inputs\": [\"age:number\"], \"outputs\": [\"band:string\"], \"rules\": [[\"<=18\", \"A\"], [\">=18\", \"B\"]]}" "{\"age\": 18}"
mentdb
ERROR: Sorry, an error appears (the hit policy UNIQUE is violated: rules 1 and 2 both match this input).

rules eval <table> <input> <json_params>


Description

    The result with parameters: value (false: true gives the value alone for a table with one output: a text without quotes, a number, an empty text for null, a JSON array for several rules), trace (false: true gives {result, matched_rules, used_default})

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    input:   The values of one case, as a JSON object {input name: value} - String - required
    json_params:   The parameters, as a JSON object - String - required
admin
rules eval "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "{\"customer\": \"Business\", \"orderSize\": 3}" "{\"value\": true}"
mentdb
0.1
Which rule gave the result
admin
rules eval "{\"hit_policy\": \"FIRST\", \"inputs\": [{\"name\": \"country\", \"type\": \"string\"}, {\"name\": \"weight\", \"type\": \"number\"}, {\"name\": \"express\", \"type\": \"boolean\"}], \"outputs\": [{\"name\": \"carrier\", \"type\": \"string\"}, {\"name\": \"price\", \"type\": \"number\"}], \"rules\": [[\"\\"FR\\"\", \"<=2\", \"false\", \"La Poste\", 4.9], [\"\\"FR\\", \\"BE\\", \\"LU\\"\", \"(2..30]\", \"-\", \"Colissimo\", 12.5], [\"not(\\"FR\\")\", \"<=30\", \"true\", \"DHL Express\", 39], [\"-\", \"-\", \"-\", \"Freight\", 120]]}" "{\"country\": \"US\", \"weight\": 1, \"express\": true}" "{\"trace\": true}"
mentdb
{"result": {"carrier": "DHL Express", "price": 39}, "matched_rules": ["3"], "used_default": false}
Used in use cases (1 recipe)

    Financial analyst › Pre-approve a loan with a priority table and keep the reasons in the file
Often used with

    rules check  ·  rules explain

rules eval_rows <table> <rows>


Description

    The rows of a table with the outputs added as columns (100000 rows by default): for COLLECT, RULE ORDER and OUTPUT ORDER each output column is a list. The first case that fails stops everything, with the number of its row. An output replaces a column of the same name

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    rows:   The rows, as a JSON array of objects (the result of sql to json, csv read...) - String - required
admin
rules eval_rows "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "[{\"order\": \"A-1\", \"customer\": \"Business\", \"orderSize\": 4}, {\"order\": \"A-2\", \"customer\": \"Private\", \"orderSize\": 40}]"
mentdb
[{"order": "A-1", "customer": "Business", "orderSize": 4, "discount": 0.1}, {"order": "A-2", "customer": "Private", "orderSize": 40, "discount": 0.05}]

rules eval_rows <table> <rows> <json_params>


Description

    The rows with parameters: matched_column (a column to add with the ids of the matched rules), on_error (stop, or column: the row gets null outputs and the message in error_column), error_column (rules_error), max_rows (100000, 1000000 at most). JSON rows of 64 MB at most, and rows x tests of the table of 1 billion at most (each entry is one test, each value of a list one more: about 3 seconds of work); split bigger jobs

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    rows:   The rows, as a JSON array of objects - String - required
    json_params:   The parameters, as a JSON object - String - required
admin
rules eval_rows "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "[{\"customer\": \"Business\", \"orderSize\": 4}, {\"customer\": \"Partner\", \"orderSize\": 4}]" "{\"on_error\": \"column\", \"matched_column\": \"rule\"}"
mentdb
[{"customer": "Business", "orderSize": 4, "discount": 0.1, "rule": ["1"], "rules_error": null}, {"customer": "Partner", "orderSize": 4, "discount": null, "rule": [], "rules_error": "the value \"Partner\" of input 'customer' is not allowed (\"Business\", \"Private\")"}]
Used in use cases (2 recipes)

    Financial analyst › Apply a discount grid that the sales team keeps in a spreadsheet
    Logistics & supply chain › Choose the carrier of each parcel with a table checked before use
Often used with

    rules check  ·  rules from_csv  ·  string char  ·  table group_by

rules explain <table> <input>


Description

    Why a case gives its result: JSON {hit_policy, input (the converted values), result, matched_rules, used_default, rules: [{rule, matched, entries: [{input, test, matched, why}], outputs}]}. A violation of UNIQUE or ANY is given in error, not thrown. For a table of more than 100000 rules x inputs, use matched_only

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    input:   The values of one case, as a JSON object {input name: value} - String - required
admin
rules explain "{\"name\": \"discount\", \"hit_policy\": \"UNIQUE\", \"inputs\": [{\"name\": \"customer\", \"type\": \"string\", \"values\": [\"Business\", \"Private\"]}, {\"name\": \"orderSize\", \"type\": \"number\"}], \"outputs\": [{\"name\": \"discount\", \"type\": \"number\", \"default\": 0}], \"rules\": [{\"when\": [\"\\"Business\\"\", \"<10\"], \"then\": 0.1, \"description\": \"small business order\"}, {\"when\": [\"\\"Business\\"\", \">=10\"], \"then\": 0.15}, {\"when\": [\"\\"Private\\"\", \"-\"], \"then\": 0.05}]}" "{\"customer\": \"Business\", \"orderSize\": 12}"
mentdb
{"hit_policy": "UNIQUE", "input": {"customer": "Business", "orderSize": 12}, "result": {"discount": 0.15}, "matched_rules": ["2"], "used_default": false, "rules": [{"rule": "1", "description": "small business order", "matched": false, "entries": [{"input": "customer", "test": "\"Business\"", "matched": true, "why": "\"Business\" satisfies \"Business\""}, {"input": "orderSize", "test": "<10", "matched": false, "why": "12 does not satisfy <10"}], "outputs": {"discount": 0.1}}, {"rule": "2", "matched": true, "entries": [{"input": "customer", "test": "\"Business\"", "matched": true, "why": "\"Business\" satisfies \"Business\""}, {"input": "orderSize", "test": ">=10", "matched": true, "why": "12 satisfies >=10"}], "outputs": {"discount": 0.15}}, {"rule": "3", "matched": false, "entries": [{"input": "customer", "test": "\"Private\"", "matched": false, "why": "\"Business\" does not satisfy \"Private\""}, {"input": "orderSize", "test": "-", "matched": true, "why": "- accepts any value"}], "outputs": {"discount": 0.05}}]}

rules explain <table> <input> <json_params>


Description

    The explanation with parameters: matched_only (false: true lists the matched rules only)

Parameters

    table:   The decision table, as JSON {"hit_policy", "inputs", "outputs", "rules"} (4 MB at most), for example the result of rules from_csv - String - required
    input:   The values of one case, as a JSON object {input name: value} - String - required
    json_params:   The parameters, as a JSON object - String - required
admin
rules explain "{\"hit_policy\": \"RULE ORDER\", \"inputs\": [\"amount:number\", \"category:string\"], \"outputs\": [\"approver:string\"], \"rules\": [[\">1000\", \"-\", \"manager\"], [\">10000\", \"-\", \"director\"], [\"-\", \"\\"IT\\"\", \"it_lead\"]]}" "{\"amount\": 25000, \"category\": \"HR\"}" "{\"matched_only\": true}"
mentdb
{"hit_policy": "RULE ORDER", "input": {"amount": 25000, "category": "HR"}, "result": [{"approver": "manager"}, {"approver": "director"}], "matched_rules": ["1", "2"], "used_default": false, "rules": [{"rule": "1", "matched": true, "entries": [{"input": "amount", "test": ">1000", "matched": true, "why": "25000 satisfies >1000"}, {"input": "category", "test": "-", "matched": true, "why": "- accepts any value"}], "outputs": {"approver": "manager"}}, {"rule": "2", "matched": true, "entries": [{"input": "amount", "test": ">10000", "matched": true, "why": "25000 satisfies >10000"}, {"input": "category", "test": "-", "matched": true, "why": "- accepts any value"}], "outputs": {"approver": "director"}}]}
Used in use cases (1 recipe)

    Financial analyst › Pre-approve a loan with a priority table and keep the reasons in the file
Often used with

    rules check  ·  rules eval