[BUGFIX] Pruning fix plus CI fixes

.github/workflows/build_test.yml

@@ -22,13 +22,14 @@ jobs:
     outputs:
       error-check: ${{ steps.error-check.conclusion }}
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install linter
-      run: python -m pip install flake8 flake8-broken-line flake8-bugbear flake8-copyright flake8-import-order flake8-quotes
+      # pinned so that new linter releases cannot break CI unannounced
+      run: python -m pip install flake8==7.4.1 flake8-broken-line==1.0.0 flake8-bugbear==26.9.30 flake8-copyright==0.2.4 flake8-import-order==0.19.2 flake8-quotes==3.4.0
     - name: Check for errors
       id: error-check
       run:
@@ -50,17 +51,18 @@ jobs:
       matrix:
         python-version: [ "3.10", "3.11", "3.12" ]
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install base package
       run: pip install .
     - name: Check package import works
       run: python -c 'import lambeq'
     - name: Install extra dependencies and tester
-      run: pip install .[extras,test,experimental]
+      # qiskit-ibm-provider still imports pkg_resources, removed in setuptools 81
+      run: pip install "setuptools<81" .[extras,test,experimental]
     - name: Locate bobcat pre-trained model cache
       id: loc-bobcat-cache
       run: echo "dir=$(python -c 'from lambeq.text2diagram.model_based_reader.model_downloader import ModelDownloader; print(ModelDownloader("bobcat").model_dir)')" >> $GITHUB_OUTPUT
@@ -96,9 +98,9 @@ jobs:
       matrix:
         python-version: [ "3.10", "3.11", "3.12" ]
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies with type hints

lambeq/backend/quantum.py

@@ -362,7 +362,7 @@ def eval(self,
                 if result.shape and post_result.shape:
                     result = np.tensordot(result, post_result, -1)
                 else:
-                    result * post_result
+                    result = result * post_result
             results.append(result)
         return results if len(results) > 1 else results[0]
 

lambeq/backend/snake_removal.py

@@ -30,7 +30,12 @@
 class InterchangerError(Exception):
     """ This is raised when we try to interchange conected boxes. """
     def __init__(self, box0: Box, box1: Box) -> None:
-        super().__init__(f'Boxes {box0} and {box1} do not commute.')
+        super().__init__(box0, box1)
+        self.box0 = box0
+        self.box1 = box1
+
+    def __str__(self) -> str:
+        return f'Boxes {self.box0} and {self.box1} do not commute.'
 
 
 def snake_removal(diagram: Diagram, left: bool = False) -> Iterator[Diagram]:

lambeq/experimental/discocirc/reader.py

@@ -262,38 +262,32 @@ def _tree2sandwiches_rec(self,
 
         return bigdiag, nouns, noun_inds, noun_cursor
 
-    def _get_index(self, s, pnoun):
-        for j, w in enumerate(s):
-            if w == pnoun:
-                return j
-        return -1
-
     def _prune_indices(self,
                        sentences,
-                       corefs,
-                       pruned_nouns: Iterable[str] = ()):
-        initial_ids = []
-
-        pruned_ids = [[] for _ in sentences]
-        # Find first coref of each pruned noun
-        for pnoun in pruned_nouns:
-            for i, s in enumerate(sentences):
-                j = self._get_index(s, pnoun)
-                if j != -1:
-                    initial_ids.append((i, j))
-                    break
-
-        for i, j in initial_ids:
-            pruned_ids[i].append(j)
-            for p in corefs:
-                if i >= len(p):
-                    break
-                elif j in p[i]:
-                    for h, hword in enumerate(p):
-                        if len(hword) > 0 and h != i:
-                            pruned_ids[h].append(hword[0])
-                    break
+                       corefd,
+                       pruned_nouns: Iterable[str] = (),
+                       min_noun_freq: int = 1) -> list[set[int]]:
+        """
+        Entities are pruned if they are referenced fewer times than
+        `min_noun_freq`, or if any of their mentions is a word in
+        `pruned_nouns`. Entities are identified by position, so other
+        entities with the same surface form are unaffected.
 
+        """
+        pruned_nouns = set(pruned_nouns)
+        noun_counts = Counter(corefd.values())
+        pruned_entities = {nid for nid, count in noun_counts.items()
+                           if count < min_noun_freq}
+        pruned_ids: list[set[int]] = [set() for _ in sentences]
+        for i, sentence in enumerate(sentences):
+            for j, word in enumerate(sentence):
+                if word in pruned_nouns:
+                    pruned_ids[i].add(j)
+                    if (i, j) in corefd:
+                        pruned_entities.add(corefd[i, j])
+        for (i, j), nid in corefd.items():
+            if nid in pruned_entities:
+                pruned_ids[i].add(j)
         return pruned_ids
 
     def text2circuit(self,
@@ -323,9 +317,9 @@ def text2circuit(self,
         pruned_nouns : iterable of strings, default: ()
             If any of the nouns in this list are present in the diagram,
             the corresponding state and wire are removed from the
-            diagram.
+            diagram, alongside every noun coreferent with them.
         min_noun_freq: int, default: 1
-            Mininum number of times a noun needs to be referenced to
+            Mininum number of times an entity needs to be referenced to
             appear in the circuit.
         rewrite_rules : list of `TreeRewriteRule` or str
             List of rewrite rules to apply to the pregroup tree
@@ -345,14 +339,8 @@ def text2circuit(self,
         sentences, corefs = self.coref_resolver.tokenise_and_coref(text)
         corefd = self.coref_resolver.dict_from_corefs(corefs)
 
-        noun_counts = Counter(corefd.values())
-        freq_pruned_ids = [nid for nid, count in noun_counts.items()
-                           if count < min_noun_freq]
-
-        pruned_nouns = set(pruned_nouns).union(
-                        {sentences[i][j] for (i, j) in freq_pruned_ids})
-
-        pruned_ids = self._prune_indices(sentences, corefs, pruned_nouns)
+        pruned_ids = self._prune_indices(sentences, corefd, pruned_nouns,
+                                         min_noun_freq)
 
         rewriter = TreeRewriter(rewrite_rules)
 

lambeq/text2diagram/ccg_rule.py

@@ -27,6 +27,7 @@
 class CCGRuleUseError(Exception):
     """Error raised when a :py:class:`CCGRule` is applied incorrectly."""
     def __init__(self, rule: CCGRule, message: str) -> None:
+        super().__init__(rule, message)
         self.rule = rule
         self.message = message
 

lambeq/text2diagram/ccg_type.py

@@ -28,6 +28,7 @@ class CCGParseError(Exception):
     """Error when parsing a CCG type string."""
 
     def __init__(self, cat: str, message: str) -> None:
+        super().__init__(cat, message)
         self.cat = cat
         self.message = message
 

lambeq/text2diagram/ccgbank_parser.py

@@ -52,6 +52,7 @@ class CCGBankParseError(Exception):
     """Error raised if parsing fails in CCGBank."""
 
     def __init__(self, sentence: str = '', message: str = '') -> None:
+        super().__init__(sentence, message)
         if message:
             self.sentence = sentence
             self.message = message

lambeq/text2diagram/depccg_parser.py

@@ -66,6 +66,7 @@ def _import_depccg() -> None:
 
 class DepCCGParseError(Exception):
     def __init__(self, sentence: str) -> None:
+        super().__init__(sentence)
         self.sentence = sentence
 
     def __str__(self) -> str:  # pragma: no cover

lambeq/text2diagram/model_based_reader/bobcat_parser.py

@@ -48,6 +48,7 @@
 
 class BobcatParseError(Exception):
     def __init__(self, sentence: str) -> None:
+        super().__init__(sentence)
         self.sentence = sentence
 
     def __str__(self) -> str:

lambeq/text2diagram/model_based_reader/model_downloader.py

@@ -42,6 +42,7 @@
 
 class ModelDownloaderError(Exception):
     def __init__(self, error_msg: str) -> None:
+        super().__init__(error_msg)
         self.error_msg = error_msg
 
     def __str__(self) -> str:

lambeq/text2diagram/model_based_reader/oncilla_parser.py

@@ -45,6 +45,7 @@
 
 class OncillaParseError(Exception):
     def __init__(self, sentence: str, reason: str = '') -> None:
+        super().__init__(sentence, reason)
         self.sentence = sentence
         self.reason = reason
 

lambeq/text2diagram/pregroup_tree.py

@@ -24,6 +24,7 @@
 
 class PregroupTreeNodeError(Exception):
     def __init__(self, sentence: str) -> None:
+        super().__init__(sentence)
         self.sentence = sentence
 
     def __str__(self) -> str:

lambeq/text2diagram/web_parser.py

@@ -40,6 +40,7 @@
 
 class WebParseError(OSError):
     def __init__(self, sentence: str) -> None:
+        super().__init__(sentence)
         self.sentence = sentence
 
     def __str__(self) -> str:

setup.cfg

@@ -75,6 +75,7 @@ extras =
     discopy >= 1.1.7
     jax < 0.6.0
     jaxlib < 0.6.0
+    autoray < 0.7.0
     pennylane >= 0.29.1, < 0.37.0
     pennylane-honeywell
     pennylane-qiskit >= 0.29.1, < 0.37.0

tests/test_discocirc_reader.py

@@ -98,3 +98,43 @@ def test_discocirc_reader_w_different_parsers(monkeypatch):
     parser.sentence2diagram.assert_called_once_with(
         sentence, tokenised=True,
     )
+
+
+prune_sentences = [['Alice', 'likes', 'it'],
+                   ['it', 'hates', 'it'],
+                   ['Alice', 'sees', 'it']]
+
+# `it` in the first two sentences is one entity, the last `it` is another
+prune_corefs = [[[0], [],     [0]],
+                [[2], [0, 2], []],
+                [[],  [],     [2]]]
+
+
+def _transitive_tree(words):
+    return PregroupTreeNode(words[1], 1, s, children=[
+        PregroupTreeNode(words[0], 0, n),
+        PregroupTreeNode(words[2], 2, n)])
+
+
+@pytest.mark.parametrize('kwargs, expected', [
+    ({'min_noun_freq': 2}, [set(), set(), {2}]),
+    ({'pruned_nouns': ['it']}, [{2}, {0, 2}, {2}]),
+])
+def test_discocirc_reader_pruning(monkeypatch, kwargs, expected):
+    parser = MockOncillaParser()
+    monkeypatch.setattr(
+        parser, '_sentence2pregrouptree',
+        MagicMock(side_effect=[_transitive_tree(w) for w in prune_sentences])
+    )
+    coref_resolver = MockCorefResolver()
+    monkeypatch.setattr(
+        coref_resolver, 'tokenise_and_coref',
+        MagicMock(return_value=(prune_sentences, prune_corefs))
+    )
+
+    r = DisCoCircReader(parser=parser, coref_resolver=coref_resolver)
+    monkeypatch.setattr(r, '_tree2circuital',
+                        MagicMock(wraps=r._tree2circuital))
+
+    r.text2circuit('', **kwargs)
+    assert [set(c.args[2]) for c in r._tree2circuital.call_args_list] == expected