From b4454cf036602c593dbef2099c407101603e4539 Mon Sep 17 00:00:00 2001
From: Matthew Honnibal <honnibal@gmail.com>
Date: Fri, 7 Nov 2014 04:40:36 +1100
Subject: [PATCH] * Add extra context tokens

---
 spacy/context.pxd |  4 ++++
 spacy/context.pyx | 16 ++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/spacy/context.pxd b/spacy/context.pxd
index 5e36aaa8b..f2d901fc7 100644
--- a/spacy/context.pxd
+++ b/spacy/context.pxd
@@ -46,11 +46,15 @@ cdef class Token:
 
 
 cdef class Slots:
+    cdef readonly Token P4
+    cdef readonly Token P3
     cdef readonly Token P2
     cdef readonly Token P1
     cdef readonly Token N0
     cdef readonly Token N1
     cdef readonly Token N2
+    cdef readonly Token N3
+    cdef readonly Token N4
 
 
 cdef int N_FIELDS
diff --git a/spacy/context.pyx b/spacy/context.pyx
index 5413039cc..7f28c352a 100644
--- a/spacy/context.pyx
+++ b/spacy/context.pyx
@@ -4,11 +4,15 @@ from .lexeme cimport *
 
 cdef class Slots:
     def __init__(self):
+        self.P4 = Token()
+        self.P3 = Token()
         self.P2 = Token()
         self.P1 = Token()
         self.N0 = Token()
         self.N1 = Token()
         self.N2 = Token()
+        self.N3 = Token()
+        self.N4 = Token()
 
 
 cdef void _number_token(Token t, int* n_fields):
@@ -137,26 +141,38 @@ cdef int _flatten_token(atom_t* context, Token ids, Token vals) except -1:
 
 
 cdef hash_t fill_slots(Slots s, int i, Tokens tokens) except 0:
+    fill_token(s.P4, tokens.lex[i-4], tokens.pos[i-4], tokens.ner[i-4])
+    fill_token(s.P3, tokens.lex[i-3], tokens.pos[i-3], tokens.ner[i-3])
     fill_token(s.P2, tokens.lex[i-2], tokens.pos[i-2], tokens.ner[i-2])
     fill_token(s.P1, tokens.lex[i-1], tokens.pos[i-1], tokens.ner[i-1])
     fill_token(s.N0, tokens.lex[i], tokens.pos[i], tokens.ner[i])
     fill_token(s.N1, tokens.lex[i+1], tokens.pos[i+1], tokens.ner[i+1])
     fill_token(s.N2, tokens.lex[i+2], tokens.pos[i+2], tokens.ner[i+2])
+    fill_token(s.N3, tokens.lex[i+3], tokens.pos[i+3], tokens.ner[i+3])
+    fill_token(s.N4, tokens.lex[i+4], tokens.pos[i+4], tokens.ner[i+4])
     return 1
 
 
 cdef int fill_flat(atom_t* context, Slots s) except -1:
+    _flatten_token(context, FIELD_IDS.P4, s.P4)
+    _flatten_token(context, FIELD_IDS.P3, s.P3)
     _flatten_token(context, FIELD_IDS.P2, s.P2)
     _flatten_token(context, FIELD_IDS.P1, s.P1)
     _flatten_token(context, FIELD_IDS.N0, s.N0)
     _flatten_token(context, FIELD_IDS.N1, s.N1)
     _flatten_token(context, FIELD_IDS.N2, s.N2)
+    _flatten_token(context, FIELD_IDS.N3, s.N4)
+    _flatten_token(context, FIELD_IDS.N4, s.N4)
 
 
 N_FIELDS = 0
 FIELD_IDS = Slots()
+_number_token(FIELD_IDS.P4, &N_FIELDS)
+_number_token(FIELD_IDS.P3, &N_FIELDS)
 _number_token(FIELD_IDS.P2, &N_FIELDS)
 _number_token(FIELD_IDS.P1, &N_FIELDS)
 _number_token(FIELD_IDS.N0, &N_FIELDS)
 _number_token(FIELD_IDS.N1, &N_FIELDS)
 _number_token(FIELD_IDS.N2, &N_FIELDS)
+_number_token(FIELD_IDS.N3, &N_FIELDS)
+_number_token(FIELD_IDS.N4, &N_FIELDS)