node.upostag -> node.upos, node.xpostag

martinpopel · martinpopel · commit 110a2df26b87 · 2017-01-13T10:52:41.000+01:00
Let's follow the agreed names used in udapi-java and udapi-perl. Note that node.upos is used also in https://github.com/UniversalDependencies/tools/blob/master/v2-conversion/depgraph_utils.py The CoNLL-U specification uses UPOSTAG and XPOSTAG as column names, but it uses also ID and HEAD, which are not used (cannot be used) in Udapi.
diff --git a/udapi/block/demo/rehangprepositions.py b/udapi/block/demo/rehangprepositions.py
@@ -4,7 +4,7 @@
 class RehangPrepositions(Block):
 
     def process_node(self, node):
-        if node.upostag == "ADP":
+        if node.upos == "ADP":
             origparent = node.parent
             node.parent = origparent.parent
             origparent.parent = node
diff --git a/udapi/block/read/conllu.py b/udapi/block/read/conllu.py
@@ -16,7 +16,7 @@ def __init__(self, strict=False, **kwargs):
         super().__init__(**kwargs)
 
         # A list of Conllu columns.
-        self.node_attributes = ["ord", "form", "lemma", "upostag", "xpostag",
+        self.node_attributes = ["ord", "form", "lemma", "upos", "xpos",
                                 "feats", "head", "deprel", "deps", "misc"]
 
         # TODO: this should be invoked from the parent class
diff --git a/udapi/block/read/reducedconllu.py b/udapi/block/read/reducedconllu.py
@@ -12,4 +12,4 @@ def __init__(self, args=None):
 
         # Here is the reduced list of the data fields as appear in the Conllu
         # reduced format.
-        self.node_attributes = ["ord", "form", "upostag", "head", "deprel"]
+        self.node_attributes = ["ord", "form", "upos", "head", "deprel"]
diff --git a/udapi/block/write/conllu.py b/udapi/block/write/conllu.py
@@ -11,7 +11,7 @@ def __init__(self, print_sent_id=True, print_text=True, print_empty_trees=True,
         self.print_empty_trees = print_empty_trees
 
         # A list of Conllu columns.
-        self.node_attributes = ["ord", "form", "lemma", "upostag", "xpostag",
+        self.node_attributes = ["ord", "form", "lemma", "upos", "xpos",
                                 "raw_feats", "parent", "deprel", "raw_deps", "misc"]
 
     def process_tree(self, tree):
diff --git a/udapi/block/write/textmodetrees.py b/udapi/block/write/textmodetrees.py
@@ -8,7 +8,7 @@
 COLOR_OF = {
     'form': 'yellow',
     'lemma': 'cyan',
-    'upostag': 'red',
+    'upos': 'red',
     'deprel': 'blue',
     'ord': 'yellow',
 }
@@ -60,7 +60,7 @@ class TextModeTrees(BaseWriter):
     """
 
     def __init__(self, print_sent_id=False, print_sentence=False, add_empty_line=True, indent=1,
-                 minimize_cross=True, color='auto', attributes='form,upostag,deprel',
+                 minimize_cross=True, color='auto', attributes='form,upos,deprel',
                  print_undef_as='', **kwargs):
         '''Create new TextModeTrees block object.
 
@@ -79,16 +79,16 @@ def __init__(self, print_sent_id=False, print_sentence=False, add_empty_line=Tru
                If you plan to pipe the output (e.g. to "less -R") and you want the colors,
                you need to set explicitly color=1, see the example in Synopsis.
         attributes: A comma-separated list of node attributes which should be printed. Possible
-                    values are ord, form, lemma, upostag, xpostag, feats, deprel, deps, misc.
+                    values are ord, form, lemma, upos, xpos, feats, deprel, deps, misc.
         print_undef_as: What should be printed instead of undefined attribute values (if any)?
         '''
         super().__init__(**kwargs)
         self.print_sent_id = print_sent_id
         self.print_sentence = print_sentence
         self.add_empty_line = add_empty_line
-        self.indent  = indent
+        self.indent = indent
         self.minimize_cross = minimize_cross
-        self.color =  color
+        self.color = color
         self.attributes = attributes
         self.print_undef_as = print_undef_as
 
@@ -118,7 +118,7 @@ def _compute_gaps(self, node):
             lm, rm, de = self._compute_gaps(child)
             lmost = min(lm, lmost)
             rmost = max(rm, rmost)
-            descs += de;
+            descs += de
         self._gaps[node.ord] = rmost - lmost - descs
         return lmost, rmost, descs + 1
 
diff --git a/udapi/block/zellig_harris/baseline.py b/udapi/block/zellig_harris/baseline.py
@@ -122,7 +122,7 @@ def process_node(self, node):
 
         """
         # We want to extract contexts only for verbs.
-        if str(node.upostag) not in self.pos:
+        if str(node.upos) not in self.pos:
             return
 
         # Process node's parent.
diff --git a/udapi/block/zellig_harris/csnouns.py b/udapi/block/zellig_harris/csnouns.py
@@ -34,7 +34,7 @@ def process_node(self, node):
 
         """
         # We want to extract contexts only for the .
-        if str(node.upostag) not in self.pos:
+        if str(node.upos) not in self.pos:
             return
 
         if self.verbose:
diff --git a/udapi/block/zellig_harris/csverbs.py b/udapi/block/zellig_harris/csverbs.py
@@ -34,7 +34,7 @@ def process_node(self, node):
 
         """
         # We want to extract contexts only for verbs.
-        if str(node.upostag) not in self.pos:
+        if str(node.upos) not in self.pos:
             return
 
         if self.verbose:
diff --git a/udapi/block/zellig_harris/ennouns.py b/udapi/block/zellig_harris/ennouns.py
@@ -34,7 +34,7 @@ def process_node(self, node):
 
         """
         # We want to extract contexts only for verbs.
-        if str(node.upostag) not in self.pos:
+        if str(node.upos) not in self.pos:
             return
 
         if self.verbose:
diff --git a/udapi/block/zellig_harris/enverbs.py b/udapi/block/zellig_harris/enverbs.py
@@ -34,7 +34,7 @@ def process_node(self, node):
 
         """
         # We want to extract contexts only for verbs.
-        if str(node.upostag) not in self.pos:
+        if str(node.upos) not in self.pos:
             return
 
         if self.verbose:
diff --git a/udapi/block/zellig_harris/queries.py b/udapi/block/zellig_harris/queries.py
@@ -8,7 +8,7 @@ def en_verb_mydobj(node):
     Extract the 'myobj' relation.
 
     """
-    if node.upostag != 'VERB':
+    if node.upos != 'VERB':
         raise ValueError('Is not a verb.')
 
     if node.feats.get('Tense', '') != 'Past':
@@ -26,4 +26,3 @@ def en_verb_mydobj(node):
             triples.append((node, 'dobj', child_node))
 
     return triples
-
diff --git a/udapi/core/node.py b/udapi/core/node.py
@@ -23,8 +23,8 @@ class Node(object):
         'ord',
         'form',    # Word form or punctuation symbol.
         'lemma',   # Lemma of word form.
-        'upostag', # Universal PoS tag
-        'xpostag', # Language-specific part-of-speech tag; underscore if not available.
+        'upos', # Universal PoS tag
+        'xpos', # Language-specific part-of-speech tag; underscore if not available.
         'deprel',  # UD dependency relation to the HEAD (root iff HEAD = 0).
         'misc',    # Any other annotation.
 
@@ -47,8 +47,8 @@ def __init__(self, data=None):
         self.ord = None
         self.form = None
         self.lemma = None
-        self.upostag = None
-        self.xpostag = None
+        self.upos = None
+        self.xpos = None
         self.deprel = None
         self.misc = None
 
@@ -422,7 +422,7 @@ def print_subtree(self, **kwargs):
         so you can use e.g.:
         files: to redirect sys.stdout to a file
         indent: to have wider trees
-        attributes: to override the default list 'form,upostag,deprel'
+        attributes: to override the default list 'form,upos,deprel'
         See TextModeTrees for details and other parameters.
         """
         TextModeTrees(**kwargs).process_tree(self)
diff --git a/udapi/core/root.py b/udapi/core/root.py
@@ -18,8 +18,8 @@ def __init__(self, sent_id=None, zone=None, misc=None, text=None):
         self.ord = 0
         self.form = '<ROOT>'
         self.lemma = '<ROOT>'
-        self.upostag = '<ROOT>'
-        self.xpostag = '<ROOT>'
+        self.upos = '<ROOT>'
+        self.xpos = '<ROOT>'
         self.deprel = '<ROOT>'
         self.misc = misc
         self.text = text
