@morscherlab/mint-sdk 1.0.0-rc.4 → 1.0.0-rc.6

package/src/__tests__/composables/autoGroup/integration.test.ts

@@ -0,0 +1,79 @@
+import { describe, it, expect } from 'vitest'
+import { readFileSync } from 'node:fs'
+import { resolve } from 'node:path'
+import {
+  pickPrimaryDelimiter,
+  splitMulti,
+  detectClass,
+  buildClassSchema,
+  composeGroups,
+} from '../../../composables/autoGroup'
+
+const FIXTURE = resolve(__dirname, '../../fixtures/auto-group/mixed-lc-ms-batch.txt')
+
+describe('mixed LC-MS batch integration', () => {
+  const samples = readFileSync(FIXTURE, 'utf-8').trim().split('\n')
+  const delim = pickPrimaryDelimiter(samples)
+  const tokenized = samples.map(s => splitMulti(s, delim))
+  const classes = detectClass(tokenized, {})
+
+  it('detects exactly 81 Biological/Tissues samples', () => {
+    const c = classes.find(c => c.kind === 'biological' && c.subKind === 'Tissues')
+    expect(c?.members.length).toBe(81)
+  })
+
+  it('detects exactly 81 Biological/Plasma samples', () => {
+    const c = classes.find(c => c.kind === 'biological' && c.subKind === 'Plasma')
+    expect(c?.members.length).toBe(81)
+  })
+
+  it('detects exactly 3 IQC and 3 EQC samples', () => {
+    const iqc = classes.find(c => c.kind === 'iqc')
+    const eqc = classes.find(c => c.kind === 'eqc')
+    expect(iqc?.members.length).toBe(3)
+    expect(eqc?.members.length).toBe(3)
+  })
+
+  it('totals biological + QC overlay + excluded === 187', () => {
+    const tissues = classes.find(c => c.kind === 'biological' && c.subKind === 'Tissues')
+    const plasma = classes.find(c => c.kind === 'biological' && c.subKind === 'Plasma')
+    const overlayKinds = ['iqc', 'eqc', 'standard', 'spike-in', 'stock', 'blank']
+    const overlayCount = classes
+      .filter(c => overlayKinds.includes(c.kind))
+      .reduce((acc, c) => acc + c.members.length, 0)
+    const unknownCount = classes
+      .filter(c => c.kind === 'unknown')
+      .reduce((acc, c) => acc + c.members.length, 0)
+    expect((tissues?.members.length ?? 0) + (plasma?.members.length ?? 0) + overlayCount + unknownCount).toBe(187)
+  })
+
+  it('default workspace produces 3 Kidney/Liver/Tumor groups of 27 each', () => {
+    const tissuesClass = classes.find(c => c.kind === 'biological' && c.subKind === 'Tissues')!
+    const tissuesTokens = tissuesClass.members.map(i => tokenized[i])
+    const schema = buildClassSchema(
+      tissuesTokens,
+      'biological',
+      'Tissues',
+      tissuesClass.classTagPositions,
+    )
+    // The Tissue column should be a factor with 3 unique values.
+    const tissueCol = schema.columns.find(c => c.uniqueValues.length === 3 && c.role === 'factor')
+    expect(tissueCol).toBeDefined()
+    const result = composeGroups({
+      tokenizedSamples: tokenized,
+      sampleNames: samples,
+      schemas: { 'biological:Tissues': { ...schema, groupBy: [tissueCol!.index] } },
+      classes: [tissuesClass],
+    })
+    expect(result.experimentalGroups!.length).toBe(3)
+    expect(result.experimentalGroups!.every(g => g.samples.length === 27)).toBe(true)
+    const names = result.experimentalGroups!.map(g => g.name).sort()
+    // Class-prefixed: "Biological / Tissues / <organ>" disambiguates from
+    // other classes that might also produce a Kidney/Liver/Tumor factor.
+    expect(names).toEqual([
+      'Biological / Tissues / Kidney',
+      'Biological / Tissues / Liver',
+      'Biological / Tissues / Tumor',
+    ])
+  })
+})

package/src/__tests__/composables/autoGroup/template.test.ts

@@ -0,0 +1,70 @@
+import { describe, it, expect } from 'vitest'
+import { composeTemplate } from '../../../composables/autoGroup/template'
+import type { ClassSchema, SampleClass } from '../../../types/auto-group'
+
+describe('composeTemplate', () => {
+  const samples = ['s_1', 's_2,with_comma', 's_3']
+  const classes: SampleClass[] = [
+    {
+      kind: 'biological',
+      subKind: 'Tissues',
+      label: 'Biological / Tissues',
+      members: [0, 1, 2],
+      classTagPositions: [],
+      disposition: 'group',
+    },
+  ]
+  const schema: ClassSchema = {
+    classKind: 'biological',
+    subKind: 'Tissues',
+    columns: [
+      { index: 0, name: 'Tissue', sourceIndices: [0], uniqueValues: ['Kidney', 'Liver'], cardinality: 2, role: 'factor' },
+    ],
+    groupBy: [0],
+  }
+
+  it('blank mode emits one row per sample with sample_name and header-only suggested columns', () => {
+    const { content } = composeTemplate(samples, null, null, { mode: 'blank', format: 'csv' })
+    const lines = content.split('\n')
+    expect(lines[0]).toBe('sample_name,group,class,replicate,notes')
+    expect(lines[1].startsWith('s_1,')).toBe(true)
+    expect(lines.length).toBe(samples.length + 1)
+  })
+
+  it('quotes values containing commas per RFC 4180', () => {
+    const { content } = composeTemplate(samples, null, null, { mode: 'blank', format: 'csv' })
+    expect(content).toContain('"s_2,with_comma"')
+  })
+
+  it('prefilled mode includes class label and schema columns', () => {
+    const tokenized = [['Kidney'], ['Kidney'], ['Liver']]
+    const { content } = composeTemplate(samples, [schema], classes, {
+      mode: 'prefilled',
+      format: 'csv',
+      tokenizedSamples: tokenized,
+    })
+    const lines = content.split('\n')
+    expect(lines[0]).toBe('sample_name,class,Tissue,group,notes')
+    expect(lines[1]).toContain('Biological / Tissues')
+    expect(lines[1]).toContain('Kidney')
+  })
+
+  it('TSV format uses tab separator and skips quoting commas', () => {
+    const { content } = composeTemplate(samples, null, null, { mode: 'blank', format: 'tsv' })
+    expect(content.split('\n')[2].split('\t')[0]).toBe('s_2,with_comma')
+  })
+
+  it('uses experiment code in filename when provided', () => {
+    const { filename } = composeTemplate(samples, null, null, {
+      mode: 'blank',
+      format: 'csv',
+      experimentCode: 'EXP-042',
+    })
+    expect(filename).toMatch(/^EXP-042_metadata_\d{4}-\d{2}-\d{2}\.csv$/)
+  })
+
+  it('falls back to default filename without experiment code', () => {
+    const { filename } = composeTemplate(samples, null, null, { mode: 'blank', format: 'csv' })
+    expect(filename).toMatch(/^sample_metadata_\d{4}-\d{2}-\d{2}\.csv$/)
+  })
+})

package/src/__tests__/composables/autoGroup/tokenize.test.ts

@@ -0,0 +1,33 @@
+import { describe, it, expect } from 'vitest'
+import { splitMulti, pickPrimaryDelimiter } from '../../../composables/autoGroup/tokenize'
+
+describe('pickPrimaryDelimiter', () => {
+  it('should pick underscore when underscore field counts are most consistent', () => {
+    const lines = ['a_b_c', 'd_e_f', 'g_h_i']
+    expect(pickPrimaryDelimiter(lines)).toBe('_')
+  })
+
+  it('should pick hyphen when underscore is absent but hyphen is consistent', () => {
+    const lines = ['a-b-c', 'd-e-f']
+    expect(pickPrimaryDelimiter(lines)).toBe('-')
+  })
+
+  it('should fall back to underscore when no delimiter splits anything', () => {
+    const lines = ['abc', 'def']
+    expect(pickPrimaryDelimiter(lines)).toBe('_')
+  })
+})
+
+describe('splitMulti', () => {
+  it('should split a single sample on the primary delimiter', () => {
+    expect(splitMulti('a_b_c', '_')).toEqual(['a', 'b', 'c'])
+  })
+
+  it('should not re-split on secondary delimiters', () => {
+    expect(splitMulti('a-b_c-d', '_')).toEqual(['a-b', 'c-d'])
+  })
+
+  it('should handle empty string as a single empty token', () => {
+    expect(splitMulti('', '_')).toEqual([''])
+  })
+})