Merge branch 'master' into spacy.io

spacy/cli/init_config.py

@@ -103,6 +103,10 @@ def fill_config(
     # config result is a valid config
     nlp = util.load_model_from_config(nlp.config)
     filled = nlp.config
+    # If we have sourced components in the base config, those will have been
+    # replaced with their actual config after loading, so we have to re-add them
+    sourced = util.get_sourced_components(config)
+    filled["components"].update(sourced)
     if pretraining:
         validate_config_for_pretrain(filled, msg)
         pretrain_config = util.load_config(DEFAULT_CONFIG_PRETRAIN_PATH)

spacy/pipeline/_parser_internals/_state.pxd

@@ -278,7 +278,7 @@ cdef cppclass StateC:
         return this._stack.size()
 
     int buffer_length() nogil const:
-        return this.length - this._b_i
+        return (this.length - this._b_i) + this._rebuffer.size()
 
     void push() nogil:
         b0 = this.B(0)

spacy/pipeline/_parser_internals/transition_system.pyx

@@ -134,8 +134,6 @@ cdef class TransitionSystem:
 
     def is_valid(self, StateClass stcls, move_name):
         action = self.lookup_transition(move_name)
-        if action.move == 0:
-            return False
         return action.is_valid(stcls.c, action.label)
 
     cdef int set_valid(self, int* is_valid, const StateC* st) nogil:

spacy/tests/regression/test_issue7055.py

@@ -0,0 +1,40 @@
+from spacy.cli.init_config import fill_config
+from spacy.util import load_config
+from spacy.lang.en import English
+from thinc.api import Config
+
+from ..util import make_tempdir
+
+
+def test_issue7055():
+    """Test that fill-config doesn't turn sourced components into factories."""
+    source_cfg = {
+        "nlp": {"lang": "en", "pipeline": ["tok2vec", "tagger"]},
+        "components": {
+            "tok2vec": {"factory": "tok2vec"},
+            "tagger": {"factory": "tagger"},
+        },
+    }
+    source_nlp = English.from_config(source_cfg)
+    with make_tempdir() as dir_path:
+        # We need to create a loadable source pipeline
+        source_path = dir_path / "test_model"
+        source_nlp.to_disk(source_path)
+        base_cfg = {
+            "nlp": {"lang": "en", "pipeline": ["tok2vec", "tagger", "ner"]},
+            "components": {
+                "tok2vec": {"source": str(source_path)},
+                "tagger": {"source": str(source_path)},
+                "ner": {"factory": "ner"},
+            },
+        }
+        base_cfg = Config(base_cfg)
+        base_path = dir_path / "base.cfg"
+        base_cfg.to_disk(base_path)
+        output_path = dir_path / "config.cfg"
+        fill_config(output_path, base_path, silent=True)
+        filled_cfg = load_config(output_path)
+    assert filled_cfg["components"]["tok2vec"]["source"] == str(source_path)
+    assert filled_cfg["components"]["tagger"]["source"] == str(source_path)
+    assert filled_cfg["components"]["ner"]["factory"] == "ner"
+    assert "model" in filled_cfg["components"]["ner"]

spacy/tests/regression/test_issue7056.py

@@ -0,0 +1,27 @@
+import pytest
+
+from spacy.tokens.doc import Doc
+from spacy.vocab import Vocab
+from spacy.pipeline._parser_internals.arc_eager import ArcEager
+
+
+def test_issue7056():
+    """Test that the Unshift transition works properly, and doesn't cause
+    sentence segmentation errors."""
+    vocab = Vocab()
+    ae = ArcEager(
+        vocab.strings,
+        ArcEager.get_actions(left_labels=["amod"], right_labels=["pobj"])
+    )
+    doc = Doc(vocab, words="Severe pain , after trauma".split())
+    state = ae.init_batch([doc])[0]
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "L-amod")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "S")
+    ae.apply_transition(state, "R-pobj")
+    ae.apply_transition(state, "D")
+    ae.apply_transition(state, "D")
+    ae.apply_transition(state, "D")
+    assert not state.eol()

website/meta/universe.json

@@ -1,5 +1,33 @@
 {
     "resources": [
+        {
+            "id": "spacy-dbpedia-spotlight",
+            "title": "DBpedia Spotlight for SpaCy",
+            "slogan": "Use DBpedia Spotlight to link entities inside SpaCy",
+            "description": "This library links SpaCy with [DBpedia Spotlight](https://www.dbpedia-spotlight.org/). You can easily get the DBpedia entities from your documents, using the public web service or by using your own instance of DBpedia Spotlight. The `doc.ents` are populated with the entities and all their details (URI, type, ...).",
+            "github": "MartinoMensio/spacy-dbpedia-spotlight",
+            "pip": "spacy-dbpedia-spotlight",
+            "code_example": [
+                "import spacy_dbpedia_spotlight",
+                "# load your model as usual",
+                "nlp = spacy.load('en_core_web_lg')",
+                "# add the pipeline stage",
+                "nlp.add_pipe('dbpedia_spotlight')",
+                "# get the document",
+                "doc = nlp('The president of USA is calling Boris Johnson to decide what to do about coronavirus')",
+                "# see the entities",
+                "print('Entities', [(ent.text, ent.label_, ent.kb_id_) for ent in doc.ents])",
+                "# inspect the raw data from DBpedia spotlight",
+                "print(doc.ents[0]._.dbpedia_raw_result)"
+            ],
+            "category": ["models", "pipeline"],
+            "author": "Martino Mensio",
+            "author_links": {
+                "twitter": "MartinoMensio",
+                "github": "MartinoMensio",
+                "website": "https://martinomensio.github.io"
+            }
+        },
         {
             "id": "spacy-textblob",
             "title": "spaCyTextBlob",