Skip to content
Open
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
more contractions from PT
  • Loading branch information
arademaker committed Apr 16, 2021
commit 60a9f8214ad4c26b296ae336bfd1bd4c92b1e1d7
39 changes: 39 additions & 0 deletions udapi/block/ud/pt/addmwt.py
Original file line number Diff line number Diff line change
Expand Up @@ -42,11 +42,40 @@
'nesses': {'form': 'em esses', 'lemma': 'em esse'},
'nesta': {'form': 'em esta', 'lemma': 'em este'},
'neste': {'form': 'em este', 'lemma': 'em este'},

'nele': {'form': 'em ele', 'lemma': 'em ele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *'},

'naquilo': {'form': 'em aquilo', 'lemma': 'em aquilo',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Masc|Number=Sing|PronType=Dem'},

'àquela': {'form': 'a aquela', 'lemma': 'a aquilo',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Fem|Number=Sing|PronType=Dem'},

# aquele can be `det` or `pron`
'naquele': {'form': 'em aquele', 'lemma': 'em aquele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Masc|Number=Sing|PronType=Dem'},
'naquela': {'form': 'em aquela', 'lemma': 'em aquele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Fem|Number=Sing|PronType=Dem'},
'naqueles': {'form': 'em aqueles', 'lemma': 'em aquele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Masc|Number=Plur|PronType=Dem'},
'naquelas': {'form': 'em aquelas', 'lemma': 'em aquele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Fem|Number=Plur|PronType=Dem'},

'daquele': {'form': 'de aquele', 'lemma': 'de aquele',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree', 'deprel': 'case *',
'feats': '_ Gender=Masc|Number=Plur|PronType=Dem'},

'nisso': {'form': 'em isso', 'lemma': 'em este'},
'nisto': {'form': 'em isto', 'lemma': 'em este',
'upos': 'ADP PRON', 'main': 1, 'shape': 'subtree'},

'no': {'form': 'em o', 'lemma': 'em o'}, # PRON cases are excluded below
'nos': {'form': 'em os', 'lemma': 'em o'}, # PRON cases are excluded below
'num': {'form': 'em um', 'lemma': 'em um'},
Expand Down Expand Up @@ -80,6 +109,16 @@
'main': 1,
'shape': 'subtree',
}
for pronoun in 'ela ele eles elas'.split():
MWTS['n' + pronoun] = {
'form': 'em ' + pronoun,
'lemma': 'em ele',
'upos': 'ADP PRON',
'deprel': 'case *',
'main': 1,
'feats': '_ *',
'shape': 'subtree',
}


class AddMwt(udapi.block.ud.addmwt.AddMwt):
Expand Down