explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-07-13 09:42:26 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Add non-breaking space to specials.json

Browse Source
This commit is contained in:
Matthew Honnibal 2015-10-10 15:54:06 +11:00
parent 9dd2f25c74
commit bdcb8d695c
2 changed files with 45 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lang_data/en/generate_specials.py
View File

@ -341,7 +341,8 @@ hardcoded_specials = {
"E.G.": [{"F": "E.G."}], "E.G.": [{"F": "E.G."}],
"\n": [{"F": "\n", "pos": "SP"}], "\n": [{"F": "\n", "pos": "SP"}],
"\t": [{"F": "\t", "pos": "SP"}], "\t": [{"F": "\t", "pos": "SP"}],
" ": [{"F": " ", "pos": "SP"}] " ": [{"F": " ", "pos": "SP"}],
u"\xa0": [{"F": u"\xa0", "pos": "SP", "L": " "}]
} }

67
lang_data/en/specials.json
View File

@ -605,9 +605,13 @@
"pos": "VB" "pos": "VB"
} }
], ],
":P": [ "11am": [
{ {
"F": ":P" "F": "11"
},
{
"L": "a.m.",
"F": "am"
} }
], ],
"Shan't": [ "Shan't": [
@ -710,6 +714,13 @@
"F": "Kan." "F": "Kan."
} }
], ],
"\u00a0": [
{
"pos": "SP",
"L": " ",
"F": "\u00a0"
}
],
"there'd": [ "there'd": [
{ {
"F": "there" "F": "there"
@ -1624,6 +1635,11 @@
"pos": "RB" "pos": "RB"
} }
], ],
"Wash.": [
{
"F": "Wash."
}
],
"She's": [ "She's": [
{ {
"L": "-PRON-", "L": "-PRON-",
@ -1885,11 +1901,6 @@
"F": "e.g." "F": "e.g."
} }
], ],
":]": [
{
"F": ":]"
}
],
"\t": [ "\t": [
{ {
"pos": "SP", "pos": "SP",
@ -2581,14 +2592,23 @@
"F": "'re" "F": "'re"
} }
], ],
"3a.m.": [
{
"F": "3"
},
{
"F": "a.m."
}
],
"^_^": [ "^_^": [
{ {
"F": "^_^" "F": "^_^"
} }
], ],
"I.e.": [ "\u2018S": [
{ {
"F": "I.e." "L": "'s",
"F": "\u2018S"
} }
], ],
"9p.m.": [ "9p.m.": [
@ -2719,9 +2739,10 @@
"pos": "VB" "pos": "VB"
} }
], ],
"Wash.": [ "\u2018s": [
{ {
"F": "Wash." "L": "'s",
"F": "\u2018s"
} }
], ],
"Couldntve": [ "Couldntve": [
@ -3249,9 +3270,9 @@
"F": "o." "F": "o."
} }
], ],
":')": [ ":]": [
{ {
"F": ":')" "F": ":]"
} }
], ],
"needn't": [ "needn't": [
@ -3535,13 +3556,9 @@
"F": "am" "F": "am"
} }
], ],
"11am": [ ":P": [
{ {
"F": "11" "F": ":P"
},
{
"L": "a.m.",
"F": "am"
} }
], ],
"Why'll": [ "Why'll": [
@ -4363,12 +4380,9 @@
"pos": "MD" "pos": "MD"
} }
], ],
"3a.m.": [ "I.e.": [
{ {
"F": "3" "F": "I.e."
},
{
"F": "a.m."
} }
], ],
"Shes": [ "Shes": [
@ -4406,6 +4420,11 @@
"F": "Apr." "F": "Apr."
} }
], ],
":')": [
{
"F": ":')"
}
],
"Conn.": [ "Conn.": [
{ {
"F": "Conn." "F": "Conn."