fix: allow middle-of-text dates when followed by time expressions (#2195)

Reworked matchDateAtBoundary() to use a single regex pass instead of
two-pass start/end anchoring. Middle-of-text matches are now accepted
when followed by a time expression (at/@ prefix), so inputs like
"meeting 9/11 at 10:00" still parse correctly while "The 9/11 Report"
is rejected.
This commit is contained in:
kolaente
2026-02-06 10:35:13 +01:00
parent cee258edc3
commit 3f0bf71d30
2 changed files with 21 additions and 13 deletions

View File

@@ -15,23 +15,29 @@ interface dateFoundResult {
const monthsRegexGroup = '(january|february|march|april|june|july|august|september|october|november|december|jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)'
/**
* Tries matching a date regex against text, but only at the start or end of the string.
* This prevents false positives like "The 9/11 Report" where the date-like pattern
* appears in the middle of the text.
* Matches a date regex against text, rejecting matches that appear in the middle
* of text with non-date content on both sides. This prevents false positives like
* "The 9/11 Report" while still allowing "meeting 9/11 at 10:00".
*
* The pattern is tested in two passes: first anchored to the start, then anchored to the end.
* Matches at the start or end of text are always accepted. Middle matches are
* only accepted when followed by a time expression (at/@ prefix).
*/
function matchDateAtBoundary(text: string, pattern: string): RegExpExecArray | null {
// Pass 1: try matching at the start of the text
const startRegex = new RegExp(`^${pattern}($| )`, 'gi')
const startResult = startRegex.exec(text)
if (startResult !== null) {
return startResult
}
const regex = new RegExp(`(^| )${pattern}($| )`, 'gi')
const result = regex.exec(text)
if (result === null) return null
// Pass 2: try matching at the end of the text
const endRegex = new RegExp(`(^| )${pattern}$`, 'gi')
return endRegex.exec(text)
const matchEnd = result.index + result[0].length
const isAtStart = result.index === 0
const isAtEnd = matchEnd >= text.length
if (isAtStart || isAtEnd) return result
// Allow middle-of-text matches when followed by a time expression
const afterMatch = text.substring(matchEnd)
if (/^(at |@ )/i.test(afterMatch)) return result
return null
}
function matchesDateExpr(text: string, dateExpr: string): boolean {

View File

@@ -392,6 +392,8 @@ describe('Parse Task Text', () => {
const boundaryTests = [
{input: '9/11 meeting', dateStr: '2021-9-11', text: 'meeting'},
{input: 'meeting 9/11', dateStr: '2021-9-11', text: 'meeting'},
{input: 'meeting 9/11 at 10:00', dateStr: '2021-9-11', text: 'meeting'},
{input: 'meeting 9/11 @ 15:00', dateStr: '2021-9-11', text: 'meeting'},
{input: '2021-06-24 Lorem Ipsum', dateStr: '2021-6-24', text: 'Lorem Ipsum'},
{input: 'Lorem Ipsum 06/26/2021', dateStr: '2021-6-26', text: 'Lorem Ipsum'},
{input: '01.02 Lorem Ipsum', dateStr: '2022-2-1', text: 'Lorem Ipsum'},