From 0fb8d437c09234ce913cb84982acfbdaf7b8c61d Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal+gh@gmail.com>
Date: Sun, 14 Feb 2021 13:38:13 +1100
Subject: [PATCH] Fix sentence fragments bug (#7056, #7035) (#7057)

* Add test for #7035

* Update test for issue 7056

* Fix test

* Fix transitions method used in testing

* Fix state eol detection when rebuffer

* Clean up redundant fix
---
 spacy/pipeline/_parser_internals/_state.pxd   |  2 +-
 .../_parser_internals/transition_system.pyx   |  2 --
 spacy/tests/regression/test_issue7056.py      | 27 +++++++++++++++++++
 3 files changed, 28 insertions(+), 3 deletions(-)
 create mode 100644 spacy/tests/regression/test_issue7056.py

diff --git a/spacy/pipeline/_parser_internals/_state.pxd b/spacy/pipeline/_parser_internals/_state.pxd
index a6bf926f9..161f3ca48 100644
--- a/spacy/pipeline/_parser_internals/_state.pxd
+++ b/spacy/pipeline/_parser_internals/_state.pxd
@@ -278,7 +278,7 @@ cdef cppclass StateC:
         return this._stack.size()
 
     int buffer_length() nogil const:
-        return this.length - this._b_i
+        return (this.length - this._b_i) + this._rebuffer.size()
 
     void push() nogil:
         b0 = this.B(0)
diff --git a/spacy/pipeline/_parser_internals/transition_system.pyx b/spacy/pipeline/_parser_internals/transition_system.pyx
index 9bb4f7f5f..9e6f847eb 100644
--- a/spacy/pipeline/_parser_internals/transition_system.pyx
+++ b/spacy/pipeline/_parser_internals/transition_system.pyx
@@ -134,8 +134,6 @@ cdef class TransitionSystem:
 
     def is_valid(self, StateClass stcls, move_name):
         action = self.lookup_transition(move_name)
-        if action.move == 0:
-            return False
         return action.is_valid(stcls.c, action.label)
 
     cdef int set_valid(self, int* is_valid, const StateC* st) nogil:
diff --git a/spacy/tests/regression/test_issue7056.py b/spacy/tests/regression/test_issue7056.py
new file mode 100644
index 000000000..64a420b84
--- /dev/null
+++ b/spacy/tests/regression/test_issue7056.py
@@ -0,0 +1,27 @@
+import pytest
+
+from spacy.tokens.doc import Doc
+from spacy.vocab import Vocab
+from spacy.pipeline._parser_internals.arc_eager import ArcEager
+
+
+def test_issue7056():
+    """Test that the Unshift transition works properly, and doesn't cause
+    sentence segmentation errors."""
+    vocab = Vocab()
+    ae = ArcEager(
+        vocab.strings,
+        ArcEager.get_actions(left_labels=["amod"], right_labels=["pobj"])
+    )
+    doc = Doc(vocab, words="Severe pain , after trauma".split())
+    state = ae.init_batch([doc])[0]
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "L-amod")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "R-pobj")
+    ae.apply_transition(state, "D")
+    ae.apply_transition(state, "D")
+    ae.apply_transition(state, "D")
+    assert not state.eol()