Merge branch 'develop' of https://github.com/explosion/spaCy into develop

spacy/__init__.py

@@ -8,8 +8,14 @@ from . import util
 
 
 def load(name, **overrides):
-    from .deprecated import resolve_load_name
-    name = resolve_load_name(name, **overrides)
+    depr_path = overrides.get('path')
+    if depr_path not in (True, False, None):
+        util.deprecated(
+            "As of spaCy v2.0, the keyword argument `path=` is deprecated. "
+            "You can now call spacy.load with the path as its first argument, "
+            "and the model's meta.json will be used to determine the language "
+            "to load. For example:\nnlp = spacy.load('{}')".format(depr_path),
+            'error')
     return util.load_model(name, **overrides)
 
 

spacy/deprecated.py

@@ -1,40 +1,4 @@
 # coding: utf8
 from __future__ import unicode_literals
 
-from .util import prints
-from .cli import download
-from . import about
-
-
 PRON_LEMMA = "-PRON-"
-
-
-def depr_model_download(lang):
-    """Replace en/de download modules within, warn and ownload default models.
-
-    lang (unicode): Language shortcut, 'en' or 'de'.
-    """
-    prints("The spacy.%s.download command is now deprecated. Please use "
-           "spacy download [model name or shortcut] instead. For "
-           "more info, see the documentation:" % lang,
-           about.__docs_models__,
-           "Downloading default '%s' model now..." % lang,
-           title="Warning: deprecated command")
-    download(lang)
-
-
-def resolve_load_name(name, **overrides):
-    """Resolve model loading if deprecated path kwarg in overrides.
-
-    name (unicode): Name of model to load.
-    **overrides: Overrides specified in spacy.load().
-    RETURNS: Model name or value of path kwarg.
-    """
-    if overrides.get('path') not in (None, False, True):
-        name = overrides.get('path')
-        prints("To load a model from a path, you can now use the first "
-               "argument. The model meta is used to load the Language class.",
-               "OLD: spacy.load('en', path='/some/path')",
-               "NEW: spacy.load('/some/path')",
-               title="Warning: deprecated argument 'path'")
-    return name

spacy/lang/de/download.py

@@ -1,8 +0,0 @@
-# coding: utf8
-from __future__ import unicode_literals
-
-from ..deprecated import depr_model_download
-
-
-if __name__ == '__main__':
-    depr_model_download('de')

spacy/lang/en/download.py

@@ -1,8 +0,0 @@
-# coding: utf8
-from __future__ import unicode_literals
-
-from ..deprecated import depr_model_download
-
-
-if __name__ == '__main__':
-    depr_model_download('en')

spacy/tests/doc/test_doc_api.py

@@ -127,7 +127,8 @@ def test_doc_api_merge(en_tokenizer):
     # merge 'The Beach Boys'
     doc = en_tokenizer(text)
     assert len(doc) == 9
-    doc.merge(doc[4].idx, doc[6].idx + len(doc[6]), 'NAMED', 'LEMMA', 'TYPE')
+    doc.merge(doc[4].idx, doc[6].idx + len(doc[6]), tag='NAMED', lemma='LEMMA',
+              ent_type='TYPE')
     assert len(doc) == 7
     assert doc[4].text == 'the beach boys'
     assert doc[4].text_with_ws == 'the beach boys '
@@ -136,7 +137,8 @@ def test_doc_api_merge(en_tokenizer):
     # merge 'all night'
     doc = en_tokenizer(text)
     assert len(doc) == 9
-    doc.merge(doc[7].idx, doc[8].idx + len(doc[8]), 'NAMED', 'LEMMA', 'TYPE')
+    doc.merge(doc[7].idx, doc[8].idx + len(doc[8]), tag='NAMED', lemma='LEMMA',
+              ent_type='TYPE')
     assert len(doc) == 8
     assert doc[7].text == 'all night'
     assert doc[7].text_with_ws == 'all night'
@@ -147,7 +149,8 @@ def test_doc_api_merge_children(en_tokenizer):
     text = "WKRO played songs by the beach boys all night"
     doc = en_tokenizer(text)
     assert len(doc) == 9
-    doc.merge(doc[4].idx, doc[6].idx + len(doc[6]), 'NAMED', 'LEMMA', 'TYPE')
+    doc.merge(doc[4].idx, doc[6].idx + len(doc[6]), tag='NAMED', lemma='LEMMA',
+              ent_type='TYPE')
 
     for word in doc:
         if word.i < word.head.i:
@@ -159,8 +162,8 @@ def test_doc_api_merge_children(en_tokenizer):
 def test_doc_api_merge_hang(en_tokenizer):
     text = "through North and South Carolina"
     doc = en_tokenizer(text)
-    doc.merge(18, 32, '', '', 'ORG')
-    doc.merge(8, 32, '', '', 'ORG')
+    doc.merge(18, 32, tag='', lemma='', ent_type='ORG')
+    doc.merge(8, 32, tag='', lemma='', ent_type='ORG')
 
 
 def test_doc_api_sents_empty_string(en_tokenizer):
@@ -188,7 +191,8 @@ def test_doc_api_runtime_error(en_tokenizer):
         if len(np) > 1:
             nps.append((np.start_char, np.end_char, np.root.tag_, np.text, np.root.ent_type_))
     for np in nps:
-        doc.merge(*np)
+        start, end, tag, lemma, ent_type = np
+        doc.merge(start, end, tag=tag, lemma=lemma, ent_type=ent_type)
 
 
 def test_doc_api_right_edge(en_tokenizer):

spacy/tests/doc/test_span_merge.py

@@ -14,7 +14,7 @@ def test_spans_merge_tokens(en_tokenizer):
     assert len(doc) == 4
     assert doc[0].head.text == 'Angeles'
     assert doc[1].head.text == 'start'
-    doc.merge(0, len('Los Angeles'), 'NNP', 'Los Angeles', 'GPE')
+    doc.merge(0, len('Los Angeles'), tag='NNP', lemma='Los Angeles', ent_type='GPE')
     assert len(doc) == 3
     assert doc[0].text == 'Los Angeles'
     assert doc[0].head.text == 'start'
@@ -36,7 +36,8 @@ def test_spans_merge_heads(en_tokenizer):
     doc = get_doc(tokens.vocab, [t.text for t in tokens], heads=heads)
 
     assert len(doc) == 8
-    doc.merge(doc[3].idx, doc[4].idx + len(doc[4]), doc[4].tag_, 'pilates class', 'O')
+    doc.merge(doc[3].idx, doc[4].idx + len(doc[4]), tag=doc[4].tag_,
+              lemma='pilates class', ent_type='O')
     assert len(doc) == 7
     assert doc[0].head.i == 1
     assert doc[1].head.i == 1
@@ -53,7 +54,8 @@ def test_span_np_merges(en_tokenizer):
     doc = get_doc(tokens.vocab, [t.text for t in tokens], heads=heads)
 
     assert doc[4].head.i == 1
-    doc.merge(doc[2].idx, doc[4].idx + len(doc[4]), 'NP', 'tool', 'O')
+    doc.merge(doc[2].idx, doc[4].idx + len(doc[4]), tag='NP', lemma='tool',
+              ent_type='O')
     assert doc[2].head.i == 1
 
     text = "displaCy is a lightweight and modern dependency parse tree visualization tool built with CSS3 and JavaScript."
@@ -63,7 +65,7 @@ def test_span_np_merges(en_tokenizer):
 
     ents = [(e[0].idx, e[-1].idx + len(e[-1]), e.label_, e.lemma_) for e in doc.ents]
     for start, end, label, lemma in ents:
-        merged = doc.merge(start, end, label, lemma, label)
+        merged = doc.merge(start, end, tag=label, lemma=lemma, ent_type=label)
         assert merged != None, (start, end, label, lemma)
 
 
@@ -88,7 +90,7 @@ def test_spans_entity_merge(en_tokenizer):
     assert len(doc) == 17
     for ent in doc.ents:
         label, lemma, type_ = (ent.root.tag_, ent.root.lemma_, max(w.ent_type_ for w in ent))
-        ent.merge(label, lemma, type_)
+        ent.merge(label=label, lemma=lemma, ent_type=type_)
     # check looping is ok
     assert len(doc) == 15
 
@@ -105,8 +107,8 @@ def test_spans_sentence_update_after_merge(en_tokenizer):
     sent1, sent2 = list(doc.sents)
     init_len = len(sent1)
     init_len2 = len(sent2)
-    doc[0:2].merge('none', 'none', 'none')
-    doc[-2:].merge('none', 'none', 'none')
+    doc[0:2].merge(label='none', lemma='none', ent_type='none')
+    doc[-2:].merge(label='none', lemma='none', ent_type='none')
     assert len(sent1) == init_len - 1
     assert len(sent2) == init_len2 - 1
 
@@ -122,5 +124,5 @@ def test_spans_subtree_size_check(en_tokenizer):
     doc = get_doc(tokens.vocab, [t.text for t in tokens], heads=heads, deps=deps)
     sent1 = list(doc.sents)[0]
     init_len = len(list(sent1.root.subtree))
-    doc[0:2].merge('none', 'none', 'none')
-    assert len(list(sent1.root.subtree)) == init_len - 1
+    doc[0:2].merge(label='none', lemma='none', ent_type='none')
+    assert len(list(sent1.root.subtree)) == init_len - 1

spacy/tests/doc/test_token_api.py

@@ -161,7 +161,7 @@ def test_is_sent_start(en_tokenizer):
     doc[5].is_sent_start = True
     assert doc[5].is_sent_start is True
     # Backwards compatibility
-    assert doc[0].sent_start is False
+    with pytest.warns(DeprecationWarning):
+        assert doc[0].sent_start is False
     doc.is_parsed = True
     assert len(list(doc.sents)) == 2
-

spacy/tokenizer.pyx

@@ -62,7 +62,11 @@ cdef class Tokenizer:
         return (self.__class__, args, None, None)
 
     cpdef Doc tokens_from_list(self, list strings):
-        # TODO: deprecation warning
+        util.deprecated(
+            "Tokenizer.from_from list is now deprecated. Create a new Doc "
+            "object instead and pass in the strings as the `words` keyword "
+            "argument, for example:\nfrom spacy.tokens import Doc\n"
+            "doc = Doc(nlp.vocab, words=[...])")
         return Doc(self.vocab, words=strings)
 
     @cython.boundscheck(False)

spacy/tokens/doc.pyx

@@ -842,17 +842,20 @@ cdef class Doc:
         """
         cdef unicode tag, lemma, ent_type
         if len(args) == 3:
-            # TODO: Warn deprecation
+            util.deprecated(
+                "Positional arguments to Doc.merge are deprecated. Instead, "
+                "use the keyword arguments, for example tag=, lemma= or "
+                "ent_type=.")
             tag, lemma, ent_type = args
             attributes[TAG] = tag
             attributes[LEMMA] = lemma
             attributes[ENT_TYPE] = ent_type
         elif not args:
-            if "label" in attributes and 'ent_type' not in attributes:
-                if isinstance(attributes["label"], int):
-                    attributes[ENT_TYPE] = attributes["label"]
+            if 'label' in attributes and 'ent_type' not in attributes:
+                if isinstance(attributes['label'], int):
+                    attributes[ENT_TYPE] = attributes['label']
                 else:
-                    attributes[ENT_TYPE] = self.vocab.strings[attributes["label"]]
+                    attributes[ENT_TYPE] = self.vocab.strings[attributes['label']]
             if 'ent_type' in attributes:
                 attributes[ENT_TYPE] = attributes['ent_type']
         elif args:

spacy/tokens/token.pyx

@@ -19,6 +19,7 @@ from ..attrs cimport IS_OOV, IS_TITLE, IS_UPPER, LIKE_URL, LIKE_NUM, LIKE_EMAIL
 from ..attrs cimport IS_STOP, ID, ORTH, NORM, LOWER, SHAPE, PREFIX, SUFFIX
 from ..attrs cimport LENGTH, CLUSTER, LEMMA, POS, TAG, DEP
 from ..compat import is_config
+from .. import util
 from .. import about
 from .underscore import Underscore
 
@@ -330,8 +331,14 @@ cdef class Token:
             return self.c.r_kids
 
     property sent_start:
-        # TODO deprecation warning
         def __get__(self):
+            util.deprecated(
+                "Token.sent_start is now deprecated. Use Token.is_sent_start "
+                "instead, which returns a boolean value or None if the answer "
+                "is unknown – instead of a misleading 0 for False and 1 for "
+                "True. It also fixes a quirk in the old logic that would "
+                "always set the property to 0 for the first word of the "
+                "document.")
             # Handle broken backwards compatibility case: doc[0].sent_start
             # was False.
             if self.i == 0:

spacy/util.py

@@ -11,6 +11,8 @@ import sys
 import textwrap
 import random
 from collections import OrderedDict
+import inspect
+import warnings
 from thinc.neural._classes.model import Model
 import functools
 
@@ -501,6 +503,18 @@ def from_disk(path, readers, exclude):
     return path
 
 
+def deprecated(message, filter='always'):
+    """Show a deprecation warning.
+
+    message (unicode): The message to display.
+    filter (unicode): Filter value.
+    """
+    stack = inspect.stack()[-1]
+    with warnings.catch_warnings():
+        warnings.simplefilter(filter, DeprecationWarning)
+        warnings.warn_explicit(message, DeprecationWarning, stack[1], stack[2])
+
+
 def print_table(data, title=None):
     """Print data in table format.
 

website/assets/js/models.js

@@ -108,6 +108,7 @@ export class ModelLoader {
         this.renderCompat(tpl, modelId);
         tpl.get('download').setAttribute('href', `${this.repo}/releases/tag/${model}`);
         tpl.get('table').removeAttribute('data-loading');
+        tpl.get('error').style.display = 'none';
     }
 
     renderDetails(tpl, { version, size, description, notes, author, url,
@@ -241,7 +242,7 @@ export class ModelComparer {
     }
 
     showError(err) {
-        console.error(err);
+        console.error(err || 'Error');
         this.tpl.get('result').style.display = 'none';
         this.tpl.get('error').style.display = 'block';
     }
@@ -286,6 +287,8 @@ export class ModelComparer {
         this.chart.update();
         [model1, model2].forEach((model, i) => this.renderTable(metaKeys, i + 1, model));
         this.tpl.get('result').removeAttribute('data-loading');
+        this.tpl.get('error').style.display = 'none';
+        this.tpl.get('result').style.display = 'block';
     }
 
     renderTable(metaKeys, i, { lang, name, version, size, description,

website/usage/_v2/_incompat.jade

@@ -1,6 +1,14 @@
 //- 💫 DOCS > USAGE > WHAT'S NEW IN V2.0 > BACKWARDS INCOMPATIBILITIES
 
+p
+    |  The following modules, classes and methods have changed between v1.x
+    |  and v2.0.
+
 +table(["Old", "New"])
+    +row
+        +cell #[code spacy.download.en], #[code spacy.download.de]
+        +cell #[+api("cli#download") #[code cli.download]]
+
     +row
         +cell
             |  #[code spacy.en] etc.
@@ -136,6 +144,17 @@
         +cell #[code Token.is_ancestor_of]
         +cell #[+api("token#is_ancestor") #[code Token.is_ancestor]]
 
++h(3, "deprecated") Deprecated
+
+p
+    |  The following methods are deprecated. They can still be used,
+    |  but should be replaced.
+
++table(["Old", "New"])
+    +row
+        +cell #[code Tokenizer.tokens_from_list]
+        +cell #[+api("doc") #[code Doc]]
+
     +row
         +cell #[code Span.sent_start]
         +cell #[+api("span#is_sent_start") #[code Span.is_sent_start]]