From 69778924c8eace4eb0187f1692a007603930dda5 Mon Sep 17 00:00:00 2001
From: Ines Montani <ines@ines.io>
Date: Thu, 12 Jan 2017 01:07:29 +0100
Subject: [PATCH] Modernise and merge parser tests and don't depend on models

---
 spacy/tests/parser/test_base_nps.py           | 41 ----------------
 .../parser/test_initial_actions_parse.py      | 16 ------
 spacy/tests/parser/test_parse.py              | 49 +++++++++++++++++++
 spacy/tests/parser/test_subtree.py            | 17 -------
 4 files changed, 49 insertions(+), 74 deletions(-)
 delete mode 100644 spacy/tests/parser/test_base_nps.py
 delete mode 100644 spacy/tests/parser/test_initial_actions_parse.py
 delete mode 100644 spacy/tests/parser/test_subtree.py

diff --git a/spacy/tests/parser/test_base_nps.py b/spacy/tests/parser/test_base_nps.py
deleted file mode 100644
index b598158d0..000000000
--- a/spacy/tests/parser/test_base_nps.py
+++ /dev/null
@@ -1,41 +0,0 @@
-from __future__ import unicode_literals
-import pytest
-
-
-# @pytest.mark.models
-# def test_nsubj(EN):
-#     sent = EN(u'A base phrase should be recognized.')
-#     base_nps = list(sent.noun_chunks)
-#     assert len(base_nps) == 1
-#     assert base_nps[0].string == 'A base phrase '
-
-
-# @pytest.mark.models
-# def test_coord(EN):
-#     sent = EN(u'A base phrase and a good phrase are often the same.')
-#     base_nps = list(sent.noun_chunks)
-#     assert len(base_nps) == 2
-#     assert base_nps[0].string == 'A base phrase '
-#     assert base_nps[1].string == 'a good phrase '
-
-
-# @pytest.mark.models
-# def test_pp(EN):
-#     sent = EN(u'A phrase with another phrase occurs')
-#     base_nps = list(sent.noun_chunks)
-#     assert len(base_nps) == 2
-#     assert base_nps[0].string == 'A phrase '
-#     assert base_nps[1].string == 'another phrase ' 
-
-
-@pytest.mark.models
-def test_merge_pp(EN):
-    sent = EN(u'A phrase with another phrase occurs')
-    nps = [(np[0].idx, np[-1].idx + len(np[-1]), np.lemma_, np[0].ent_type_) for np in sent.noun_chunks]
-
-    for start, end, lemma, ent_type in nps:
-        sent.merge(start, end, u'NP', lemma, ent_type)
-    assert sent[0].string == 'A phrase '
-    assert sent[1].string == 'with '
-    assert sent[2].string == 'another phrase '
-    assert sent[3].string == 'occurs'
diff --git a/spacy/tests/parser/test_initial_actions_parse.py b/spacy/tests/parser/test_initial_actions_parse.py
deleted file mode 100644
index a4a57e5b2..000000000
--- a/spacy/tests/parser/test_initial_actions_parse.py
+++ /dev/null
@@ -1,16 +0,0 @@
-import pytest
-
-
-@pytest.mark.models
-def test_initial(EN):
-    doc = EN.tokenizer(u'I ate the pizza with anchovies.')
-    EN.tagger(doc)
-    with EN.parser.step_through(doc) as stepwise:
-        stepwise.transition('L-nsubj')
-        stepwise.transition('S')
-        stepwise.transition('L-det')
-    assert doc[0].head.i == 1
-    assert doc[1].head.i == 1
-    assert doc[2].head.i == 3
-    assert doc[3].head.i == 3
-    assert doc
diff --git a/spacy/tests/parser/test_parse.py b/spacy/tests/parser/test_parse.py
index ae6c10021..25d265e09 100644
--- a/spacy/tests/parser/test_parse.py
+++ b/spacy/tests/parser/test_parse.py
@@ -27,6 +27,55 @@ def test_parser_parse_one_word_sentence(en_tokenizer, en_parser, text):
     assert doc[0].dep != 0
 
 
+def test_parser_initial(en_tokenizer, en_parser):
+    text = "I ate the pizza with anchovies."
+    heads = [1, 0, 1, -2, -3, -1, -5]
+    transition = ['L-nsubj', 'S', 'L-det']
+
+    tokens = en_tokenizer(text)
+    apply_transition_sequence(en_parser, tokens, transition)
+
+    assert tokens[0].head.i == 1
+    assert tokens[1].head.i == 1
+    assert tokens[2].head.i == 3
+    assert tokens[3].head.i == 3
+
+
+def test_parser_parse_subtrees(en_tokenizer, en_parser):
+    text = "The four wheels on the bus turned quickly"
+    heads = [2, 1, 4, -1, 1, -2, 0, -1]
+    tokens = en_tokenizer(text)
+    doc = get_doc(tokens.vocab, [t.text for t in tokens], heads=heads)
+
+    assert len(list(doc[2].lefts)) == 2
+    assert len(list(doc[2].rights)) == 1
+    assert len(list(doc[2].children)) == 3
+    assert len(list(doc[5].lefts)) == 1
+    assert len(list(doc[5].rights)) == 0
+    assert len(list(doc[5].children)) == 1
+    assert len(list(doc[2].subtree)) == 6
+
+
+def test_parser_merge_pp(en_tokenizer):
+    text = "A phrase with another phrase occurs"
+    heads = [1, 4, -1, 1, -2, 0]
+    deps = ['det', 'nsubj', 'prep', 'det', 'pobj', 'ROOT']
+    tags = ['DT', 'NN', 'IN', 'DT', 'NN', 'VBZ']
+
+    tokens = en_tokenizer(text)
+    doc = get_doc(tokens.vocab, [t.text for t in tokens], deps=deps, heads=heads)
+    for token in doc:
+        token.tag_ = tags[token.i]
+    nps = [(np[0].idx, np[-1].idx + len(np[-1]), np.lemma_) for np in doc.noun_chunks]
+
+    for start, end, lemma in nps:
+        doc.merge(start, end, label='NP', lemma=lemma)
+    assert doc[0].text == 'A phrase'
+    assert doc[1].text == 'with'
+    assert doc[2].text == 'another phrase'
+    assert doc[3].text == 'occurs'
+
+
 def test_parser_arc_eager_finalize_state(en_tokenizer, en_parser):
     text = "a b c d e"
 
diff --git a/spacy/tests/parser/test_subtree.py b/spacy/tests/parser/test_subtree.py
deleted file mode 100644
index 8502a0ac9..000000000
--- a/spacy/tests/parser/test_subtree.py
+++ /dev/null
@@ -1,17 +0,0 @@
-from __future__ import unicode_literals
-import pytest
-
-
-@pytest.mark.models
-def test_subtrees(EN):
-    sent = EN('The four wheels on the bus turned quickly')
-    wheels = sent[2]
-    bus = sent[5]
-    assert len(list(wheels.lefts)) == 2
-    assert len(list(wheels.rights)) == 1
-    assert len(list(wheels.children)) == 3
-    assert len(list(bus.lefts)) == 1
-    assert len(list(bus.rights)) == 0
-    assert len(list(bus.children)) == 1
-
-    assert len(list(wheels.subtree)) == 6