Initial import
diff --git a/test/test.js b/test/test.js
new file mode 100644
index 0000000..6367049
--- /dev/null
+++ b/test/test.js
@@ -0,0 +1,173 @@
+const { execSync } = require('child_process');
+
+describe('conllu-gender', () => {
+
+ test('Full mode: all gender-sensitive nouns (Genderstern singular)', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+
+ // Genderstern singular noun: Bürger*in → lemma Bürger*in, NOUN NN Gender=NonBin|Number=Sing
+ expect(stdout).toContain('Bürger*in\tBürger*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // Long compound: Jugendpsychiater*in
+ expect(stdout).toContain('Jugendpsychiater*in\tJugendpsychiater*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // Compound with umlaut base: Generalstaatsanwält*in
+ expect(stdout).toContain('Generalstaatsanwält*in\tGeneralstaatsanwält*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // Nachfolger*in
+ expect(stdout).toContain('Nachfolger*in\tNachfolger*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // With umlaut base: Antragssteller*in (no umlaut but long compound)
+ expect(stdout).toContain('Antragssteller*in\tAntragssteller*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ });
+
+ test('Full mode: Genderstern plural nouns → lemma uses singular, Number=Plur', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // Fachärzt*innen → lemma Fachärzt*in
+ expect(stdout).toContain('Fachärzt*innen\tFachärzt*in\tNOUN\tNN\tGender=NonBin|Number=Plur');
+ });
+
+ test('Full mode: Doppelpunkt nouns', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // Doppelpunkt singular
+ expect(stdout).toContain('Anhänger:in\tAnhänger:in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ expect(stdout).toContain('Wirt:in\tWirt:in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // Doppelpunkt plural
+ expect(stdout).toContain('Lehrer:innen\tLehrer:in\tNOUN\tNN\tGender=NonBin|Number=Plur');
+ // Long compound with umlaut base
+ expect(stdout).toContain('Menschenrechtsanwält:innen\tMenschenrechtsanwält:in\tNOUN\tNN\tGender=NonBin|Number=Plur');
+ });
+
+ test('Full mode: Unterstrich plural nouns', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ expect(stdout).toContain('Autor_innen\tAutor_in\tNOUN\tNN\tGender=NonBin|Number=Plur');
+ });
+
+ test('Full mode: Binnen-I nouns (binary intended → Gender=Masc,Fem)', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // Binnen-I singular: ZeugIn → Gender=Masc,Fem|Number=Sing
+ expect(stdout).toContain('ZeugIn\tZeugIn\tNOUN\tNN\tGender=Masc,Fem|Number=Sing');
+ // Binnen-I singular: SchülerIn → Gender=Masc,Fem|Number=Sing
+ expect(stdout).toContain('SchülerIn\tSchülerIn\tNOUN\tNN\tGender=Masc,Fem|Number=Sing');
+ // Binnen-I plural: LehrerInnen → lemma LehrerIn, Number=Plur
+ expect(stdout).toContain('LehrerInnen\tLehrerIn\tNOUN\tNN\tGender=Masc,Fem|Number=Plur');
+ });
+
+ test('Full mode: Klammern plural nouns (binary intended)', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ expect(stdout).toContain('Schüler(innen)\tSchüler(in)\tNOUN\tNN\tGender=Masc,Fem|Number=Plur');
+ });
+
+ test('Full mode: Schrägstrich nouns', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // Simple slash
+ expect(stdout).toContain('Autor/innen\tAutor/in\tNOUN\tNN\tGender=Masc,Fem|Number=Plur');
+ // Slash with Ergänzungsstrich (/-innen)
+ expect(stdout).toContain('Autor/-innen\tAutor/in\tNOUN\tNN\tGender=Masc,Fem|Number=Plur');
+ expect(stdout).toContain('Spieler/-innen\tSpieler/in\tNOUN\tNN\tGender=Masc,Fem|Number=Plur');
+ });
+
+ test('Full mode: gendered determiners/pronouns (non-binary intended)', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // jede*r → DET PIAT Gender=NonBin, lemma jede*r
+ expect(stdout).toContain('jede*r\tjede*r\tDET\tPIAT\tGender=NonBin');
+ const jeder_count = (stdout.match(/jede\*r\tjede\*r/g) || []).length;
+ expect(jeder_count).toBe(2); // appears in two sentences
+ // eine*n → DET ART Gender=NonBin (non-binary via *)
+ expect(stdout).toContain('eine*n\teine*n\tDET\tART\tGender=NonBin');
+ // jede:r → DET PIAT Gender=NonBin
+ expect(stdout).toContain('jede:r\tjede:r\tDET\tPIAT\tGender=NonBin');
+ // die*der → DET ART Gender=NonBin (* marker → NonBin even for merged forms)
+ expect(stdout).toContain('die*der\tdie*der\tDET\tART\tGender=NonBin');
+ });
+
+ test('Full mode: neo-pronoun (sie*er → PRON PPER Gender=NonBin)', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ expect(stdout).toContain('sie*er\tsie*er\tPRON\tPPER\tGender=NonBin');
+ });
+
+ test('Full mode: foundry comment changed to gender', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ const foundry_count = (stdout.match(/# foundry = gender/g) || []).length;
+ expect(foundry_count).toBe(10);
+ });
+
+ test('Full mode: non-gender tokens pass through unchanged', () => {
+ const command = 'node src/index.js < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // Regular nouns are passed through, lemma stays _
+ expect(stdout).toContain('Umbenennung\t_\tNOUN\tNN');
+ expect(stdout).toContain('Ideen\t_\tNOUN\tNN');
+ });
+
+ test('Sparse mode: only annotated tokens are emitted', () => {
+ const command = 'node src/index.js -s < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ const lines = stdout.split('\n');
+ const tokenLines = lines.filter(l => l.match(/^\d+\t/));
+ // Every token line must carry a gender annotation:
+ // either the lemma column (col 2) is non-underscore (noun/pron was annotated)
+ // or the features column (col 5) is non-underscore (det was annotated)
+ tokenLines.forEach(line => {
+ const cols = line.split('\t');
+ const lemmaAnnotated = cols[2] !== '_';
+ const featsAnnotated = cols[5] !== '_';
+ expect(lemmaAnnotated || featsAnnotated).toBe(true);
+ });
+ // Count: 18 NOUN + 5 DET + 1 PRON = 24 annotated tokens
+ expect(tokenLines.length).toBe(24);
+ });
+
+ test('Sparse mode: sentence headers are emitted for sentences with matches', () => {
+ const command = 'node src/index.js -s < test/data/gender.conllu';
+ const stdout = execSync(command).toString();
+ // All 10 test sentences have at least one gender form
+ const text_id_count = (stdout.match(/# text_id = /g) || []).length;
+ expect(text_id_count).toBe(10);
+ });
+
+ test('Inline input: basic Genderstern annotation', () => {
+ const testInput = `# foundry = base
+# text_id = inline-001
+# text = Die Lehrerin und Lehrer*innen kamen
+1\tDie\t_\tDET\tART\t_\t_\t_\t_\t_
+2\tLehrerin\t_\tNOUN\tNN\t_\t_\t_\t_\t_
+3\tund\t_\tCCONJ\tKON\t_\t_\t_\t_\t_
+4\tLehrer*innen\t_\t_\t_\t_\t_\t_\t_\t_
+5\tkamen\t_\tVERB\tVVFIN\t_\t_\t_\t_\t_
+
+`;
+ const stdout = execSync('node src/index.js', { input: testInput }).toString();
+ expect(stdout).toContain('Lehrer*innen\tLehrer*in\tNOUN\tNN\tGender=NonBin|Number=Plur');
+ // Regular noun 'Lehrerin' must not be incorrectly tagged
+ expect(stdout).toContain('Lehrerin\t_\tNOUN\tNN\t_');
+ });
+
+ test('Inline input: existing POS/lemma/feats are replaced for gender forms', () => {
+ const testInput = `# foundry = base
+# text_id = inline-002
+# text = jede Ärztin und jede*r Arzt*in
+1\tjede\t_\tDET\tPIAT\tGender=Fem|Number=Sing\t_\t_\t_\t_
+2\tÄrztin\tÄrztin\tNOUN\tNN\tGender=Fem|Number=Sing\t_\t_\t_\t_
+3\tund\t_\tCCONJ\tKON\t_\t_\t_\t_\t_
+4\tjede*r\t_\tDET\tPIAT\t_\t_\t_\t_\t_
+5\tArzt*in\t_\t_\t_\t_\t_\t_\t_\t_
+
+`;
+ const stdout = execSync('node src/index.js', { input: testInput }).toString();
+ // jede*r: missing feats should be filled in
+ expect(stdout).toContain('jede*r\tjede*r\tDET\tPIAT\tGender=NonBin');
+ // Arzt*in: umlaut base, missing everything
+ expect(stdout).toContain('Arzt*in\tArzt*in\tNOUN\tNN\tGender=NonBin|Number=Sing');
+ // Ärztin (regular moviertes Femininum, no gender marker): unchanged
+ expect(stdout).toContain('Ärztin\tÄrztin\tNOUN\tNN\tGender=Fem|Number=Sing');
+ // jede (without gender marker): unchanged
+ expect(stdout).toContain('jede\t_\tDET\tPIAT\tGender=Fem|Number=Sing');
+ });
+});