Validation Rules Reference¶

This page documents every rule checked during the Validate & Fix workflow. Rules are loaded from config/rules.stix.default.json at runtime.

Severity Levels¶

Rule Catalogue¶

Rule: xml-wellformed¶

Severity: error Field: (structural)

Checks that the uploaded file is valid XML. This is the first check — if it fails, no other rules run.

Pass condition: fast-xml-parser can parse the file without throwing. Auto-fix: No. A malformed XML file must be corrected externally.

Rule: root-element¶

Severity: error Field: (structural)

Checks that the root element is ns1:SchoolUpload.

Pass condition: The parsed document has a root key of ns1:SchoolUpload. Auto-fix: No.

Rule: school-structure¶

Severity: error Field: SchoolNumber

Each ns1:School element must have a SchoolNumber attribute or child element.

Pass condition: SchoolNumber is present and non-empty on the school element. Auto-fix: No.

Rule: required-fields¶

Severity: error Fields: FirstName, LastName, BirthDate, Grade, SchoolNumber

Each student record must have all required fields present and non-empty.

Required fields (from rules.stix.default.json):

FirstName, LastName, BirthDate, Grade, SchoolNumber

Pass condition: Each required field exists and contains at least one non-whitespace character. Auto-fix: No. Missing required fields must be supplied manually.

Rule: grade-value¶

Severity: error Field: Grade

The Grade field must be one of the allowed values.

Allowed values:

JK, SK, GR1, GR2, GR3, GR4, GR5, GR6, GR7, GR8, GR9, GR10, GR11, GR12

Grade aliases (auto-corrected):

Auto-fix: Yes, when the value matches a known alias. The suggested fix is the canonical form.

Rule: gender-value¶

Severity: error Field: Gender

The Gender field must be one of the allowed values.

Allowed values:

M, F, X, U

Gender aliases (auto-corrected):

Auto-fix: Yes, when the value matches a known alias.

Rule: province-value¶

Severity: warning Field: Province

The Province field must be a valid Canadian province or territory code.

Allowed values:

AB, BC, MB, NB, NL, NS, NT, NU, ON, PE, QC, SK, YT

Auto-fix: No. Province codes must be corrected manually.

Rule: BIRTHDATE_FORMAT¶

Severity: error Field: BirthDate

The BirthDate field must be in YYYY-MM-DD format and represent a parseable calendar date.

Pass condition: Value matches YYYY-MM-DD and the date is valid (e.g. month 1–12, day within month).

Auto-fix: Only for real, unambiguous numeric dates no later than today. XML validation suggests YYYY-MM-DD; workbook import uses the same calendar rules to normalize birth-date text before validation.

Use a four-digit year and consistent slash or hyphen separators. Month names, two-digit years, numeric serial text, and timestamps require source confirmation; PanoReady does not use locale guessing or timezone conversion. Workbook dates are evaluated as formatted cell text, so export them as YYYY-MM-DD to avoid ambiguous displays. Invalid or ambiguous values are preserved for review. Future birth dates still fail validation and receive no automatic suggestion.

Metadata CreateDate continues to require a real YYYY-MM-DD date no later than today; it is not automatically rewritten. Age reports and filters consume only confirmed, valid ISO birth dates. Unresolved dates have unknown age and do not contribute to age-group counts.

OEN identity findings¶

OEN_DUPLICATE blocks same-school duplicates; OEN_DUAL_ENROLLMENT warns about cross-school occurrences. Neither finding offers a replacement OEN input. Review the records in the source system and upload a corrected file if needed. A cross-school occurrence does not hide later same-school duplicates. Correction of a malformed OEN (OEN_FORMAT) remains available. A separate duplicate resolution workflow is outside the current scope.

Rule: postal-code¶

Severity: info, warning, or error Field: PostalCode

The built-in rule validates Canadian postal-code structure and produces the canonical six-character STIX value A1A1A1 (uppercase, no separator). Canada Post normally displays the same code as A1A 1A1; PanoReady accepts that form as input and suggests the unspaced machine representation used in its output.

Allowed characters:

• Position 1: A B C E G H J K L M N P R S T V X Y
• Positions 3 and 5: A B C E G H J K L M N P R S T V W X Y Z
• Positions 2, 4, and 6: digits 0–9; zero in position 2 is valid and identifies a rural forward sortation area
• Letters D F I O Q U are never valid; W Z are additionally excluded from position 1

Built-in canonical pattern:

^[ABCEGHJKLMNPRSTVXY]\d[ABCEGHJKLMNPRSTVWXYZ]\d[ABCEGHJKLMNPRSTVWXYZ]\d$

The following input connectors are accepted only between the first and last three characters: one or more whitespace characters, -, en dash (–), em dash (—), /, \, _, and .. Whitespace may surround a punctuation connector. For example, N1G / 2W1 safely normalizes to N1G2W1. A connector in any other position, repeated punctuation, or arbitrary punctuation remains invalid; PanoReady does not globally strip punctuation.

The validator reports one of four issue outcomes:

A five-digit numeric value or a five-digit value followed by a hyphen and four digits matches the standard U.S. ZIP / ZIP+4 shape. These values produce a blocking POSTAL_CODE_US_ZIP error and must be replaced with a Canadian postal code. PanoReady does not convert U.S. ZIP codes automatically.

Only O → 0, I → 1, and L → 1 are repaired, and only in positions that must be numeric. PanoReady does not truncate, stitch, reverse a digit into a letter, or make speculative substitutions such as S → 5.

These checks prove syntactic structure only. They do not prove that Canada Post currently assigns the postal code or that it belongs to the supplied address.

Rule sources:

• Canada Post — Addressing guidelines: Postal codes
• Canada Post — Addressing guidelines: Important information
• Statistics Canada — Postal Code Conversion File Reference Guide

Rules: PHONE_FORMAT, PHONE_EXTENSION_NORMALIZE, PHONE_EXTENSION_FORMAT, PHONE_NPA_STRUCTURE, PHONE_NXX_STRUCTURE, PHONE_PLACEHOLDER¶

Severity: error Fields: metadata ContactPhone, student Phone, and each guardian Phone

The built-in phone rule expects the canonical North American Numbering Plan (NANP) form NPA-NXX-XXXX, optionally followed immediately by a lowercase x and an extension of 1–5 digits: NPA-NXX-XXXXx12345. Both the area code (NPA) and central-office or exchange code (NXX) have the form NXX, where N is a digit from 2 through 9 and each X is a digit from 0 through 9.

For example, 519-824-1234 is structurally valid. 019-824-1234 has an invalid NPA, and 519-124-1234 has an invalid NXX. PanoReady never guesses a replacement digit for either structural failure.

Common numeric presentation variants are interpreted deterministically. Ten digits, parentheses and spaces, and an optional leading NANP country code 1 (including +1) can be normalized to XXX-XXX-XXXX. The resulting suggestion uses the existing apply-and-revalidate workflow. Common unambiguous extension markers—uppercase X, spaced x, ext, ext., extension, and #—are similarly normalized. For example, 519-824-1234 ext. 12 becomes 519-824-1234x12.

An extension marker with no digits, more than five digits, non-numeric extension text, too few or too many base-number digits, appended notes, and multiple numbers require manual correction. PanoReady does not guess or truncate those values. Numbers listed in phoneConfig.placeholderNumbers are also errors; a placeholder may receive both the applicable structural issue and the placeholder issue.

Auto-fix: Base-number formatting and unambiguous extension-marker normalization only. Missing, overlong, or non-numeric extensions, NPA/NXX structural failures, multiple numbers, appended notes, and placeholders are not auto-fixable.

Rule sources:

• Canadian Numbering Administrator — NPA Code (Area Code)
• NANPA — CO Codes/Thousands-Blocks
• NANPA — About the North American Numbering Plan

Rule: PHONE_CANADIAN_AREA_CODE¶

Severity: off, info, or warning Fields: metadata ContactPhone, student Phone, and each guardian Phone

After a number passes NANP structural validation, this optional policy rule checks whether its NPA is a currently active Canadian geographic area code. It does not classify a structurally valid US or other NANP number as invalid. The built-in STIX ruleset uses warning; custom rulesets can choose off, info, or warning. The rule can never produce an error and therefore never changes a READY gate to BLOCKED.

PanoReady bundles a static CNAC-derived set rather than making a runtime network request. The set was verified on 2026-09-01 against CNAC's current CO Code Status data and relief notices. It excludes future or reserved relief NPAs such as 273 and 851, and Canadian non-geographic numbering resources such as 600. Because area codes change, that source date is part of the rule and the set must be reviewed over time.

No Canadian policy issue is emitted unless the number first passes both the NPA and NXX structural checks. Policy findings have no suggested fix.

Passing the structural and Canadian-geographic checks proves neither that the subscriber number exists nor that it is active or belongs to the stated person.

Rule sources:

• Canadian Numbering Administrator — CO Code Status
• Canadian Numbering Administrator — Non-Geographic (6YY) Codes
• Canadian Numbering Administrator — 782/902 relief planning for future NPA 851

Rule: oen-format¶

Severity: error Field: OEN

The Ontario Education Number (OEN) must be exactly 9 digits.

Pass condition: Value contains exactly 9 digit characters and no other characters.

Auto-fix: No.

Rule: field-length¶

Severity: warning Field: varies

Field values must not exceed the maximum length for that field.

Maximum field lengths:

Auto-fix: No. Values exceeding the limit must be shortened manually.

PostalCode and PoBoxNumber are handled by their dedicated format rules rather than this generic length rule. Invalid postal codes and box numbers are never auto-truncated.

Rule: whitespace¶

Severity: info Field: varies

Fields with leading or trailing whitespace are flagged.

PostalCode is intentionally excluded: its dedicated rule supplies the full canonical value so the same input does not also receive a generic whitespace issue.

Auto-fix: Yes. The suggested fix is the trimmed value (whitespace removed from both ends).

Rule: ALTERNATE_DELIVERY_IN_STREET_FIELD¶

Severity: warning

Fields: StreetName, StreetNumber

Detects complete numeric PO Box references entered in a street field, including PO Box 42, P.O. Box #42, Post Office Box 42, and Box 42. Detection runs before the generic character policy.

Auto-fix: Yes when PoBoxNumber is empty or identifies the same number and the proposed values meet their field rules. The fix stores only the digits in PoBoxNumber. Leading zeroes are preserved. A complete box reference clears the source field. A leading reference followed by whitespace or a hyphen and a civic-address fragment (street number followed by street text) preserves that fragment in the source field. For example, PO BOX 57 918 Example Grove becomes box number 57 with 918 Example Grove retained in StreetName, and P.O Box 68-427 Sample Way becomes box number 68 with 427 Sample Way retained. The retained fragment can then be reviewed as part of the street address. Conflicting destinations, references in both street fields, letters in the box number, additional delivery references, station information, and unrecognized trailing text remain manual review. A numeric range without street text, such as PO Box 57-918, is not automatically split.

Rule: PO_BOX_NUMBER_FORMAT¶

Severity: warning; error when the configured length limit is exceeded

Field: PoBoxNumber

A populated field must contain only the box number in digits, such as 42 or 0042. Do not include PO Box, a #, spaces, letters, or other address text. The optional field may be empty. Manual field editing uses the same validation and marks an invalid draft Check value. The formatted address preview adds PO Box for readability; the field and exported value contain only the number.

Auto-fix: Yes for a complete recognizable numeric value: PO Box 0042, P.O. Box #0042, Post Office Box 0042, Box 0042, and #0042 become 0042. Numbers are never truncated or converted to integers. Values containing other text or numbers exceeding the configured limit remain manual review. This rule owns the field, so generic character cleanup cannot remove punctuation to turn an ambiguous value into a different box number.

Rule: RURAL_ROUTE_IN_STREET_FIELD¶

Severity: warning Fields: StreetName, StreetNumber

Detects rural-route delivery text entered in a street field. Detection runs before the generic character policy so punctuation can contribute to recognition. For example, R.R. #01 is accepted as input even though that punctuation is not retained in the canonical rural-route value.

Auto-fix: Yes for an unambiguous route number when RuralRoute is empty or already identifies the same route. The fix writes the normalized value, such as RR 1, to RuralRoute. If the route is the complete source value, that source field is cleared. If it is a leading token followed by a hyphen or whitespace, the token and separator are removed while the remaining street text is preserved. A conflicting existing RuralRoute, or another rural-route marker in the remaining text, remains manual review.

Rule: RURAL_ROUTE_FORMAT¶

Severity: warning

Field: RuralRoute

A populated rural-route field must contain the uppercase identifier RR, one space, and a route number containing one to four digits—for example, RR 4. A #, periods, expanded wording, station information, additional spaces, five or more route-number digits, and other text are not accepted in this canonical field. An empty value remains valid because the field is optional. This is a data-quality finding and does not block submission.

The same rule is applied while editing an address during manual review, so an invalid draft is marked Check value before it is applied.

Auto-fix: Yes when a one-to-four digit route number can be parsed unambiguously, including a bare number in the RuralRoute field (for example, 2 becomes RR 2 and R.R. #01 becomes RR 1). Longer route numbers and values containing other text remain manual review.

Rule source: Canada Post — Addressing guidelines: Civic address

Rules: free-text characters¶

Severity: info Fields: configured independently in freeTextCharacterChecks

Four separate rules identify characters that are accepted in STIX but may make downstream reporting or matching less reliable:

Each category has its own field list. The built-in profile permits apostrophes, periods, and ampersands in SchoolName, while the other enabled checks still apply there. Periods and slashes are also preserved in the street-address fields Unit, StreetNumber, StreetNumberSuffix, and StreetName. Periods, slashes, and ampersands are preserved in City as well. Profiles can enable or disable every category for every supported free-text field. When one field contains multiple categories, the findings remain separate but share the same composed suggestion so applying any or all of them produces the complete safe value.

Balanced parenthetical spans are excluded entirely: the brackets and their contents are preserved for possible future handling as aliases or former names. An unmatched round bracket is allowed but does not shield the remaining text.

Character checks run last and act only as a fallback. If another validator, including rural-route detection, already reports the same record field, its specialized correction owns that field and no character finding is added.

Auto-fix: Yes. These findings never change the validation gate.

Rule: duplicate-oen¶

Severity: error Field: OEN

Each OEN must appear only once in the file. A duplicate OEN indicates two records share the same student identifier.

Pass condition: No two student records have the same non-empty OEN value. Auto-fix: No. Duplicate OENs require manual investigation to determine which record is correct.

Rule: duplicate-name-dob-school¶

Severity: warning Field: FirstName, LastName, BirthDate, SchoolNumber

Flags student records where the combination of FirstName + LastName + BirthDate + SchoolNumber appears more than once, indicating a possible duplicate enrollment.

Pass condition: No two records share the same name, date of birth, and school. Auto-fix: No.

Custom Rulesets¶

The allowed values, field lengths, patterns, and aliases that drive these rules are configurable. You can create board-specific rulesets without code changes by using the Validation ruleset control in Validate & Fix.

See docs/rulesets.md for a full field-by-field reference, including how each ruleset field maps to the rule IDs above.

Auto-Fix Summary¶

XML structure acceptance¶

STIX XML is checked before values enter the working model. Every element must use the http://ontario.ca namespace and the supported STIX hierarchy. Unknown elements or attributes, duplicate singleton elements, more than two guardians, mixed container text, multiple roots, and DTD/entity declarations are rejected. Namespace aliases, comments, standard XML entities, phone type attributes and root xsi:schemaLocation are supported. Missing values in known fields remain validation findings. These are local structure checks, not certified XSD validation.

Workbook birth-date interpretation¶

Intake shows evidence across all imported student birth dates, with the first five populated values as a preview. Choose day/month/year or month/day/year to resolve ambiguous text; each file in Compare Files has its own choice. Opposing unambiguous day-first and month-first values block conversion even if an order is selected. Choices reset when the file changes and are not stored with workbook metadata.

Typed Excel date cells use their stored numeric value, date format and workbook 1900/1904 date system, not their displayed text. Excel serial 60 in the 1900 system, fractional days, numeric cells without a date format, formulas and invalid calendar dates are rejected. Year-first and uniquely interpretable text dates can normalize without a choice. Output uses YYYY-MM-DD. Unresolved ambiguous, invalid or conflicting dates block intake; correct the source or choose a consistent interpretation before continuing. Other workbook mapping behavior is unchanged.