Support text separator customization

.code-samples.meilisearch.yaml

@@ -683,6 +683,18 @@ facet_search_2: |-
   client.index('books').update_faceting_settings({ 'sortFacetValuesBy': { 'genres': 'count' } })
 facet_search_3: |-
   client.index('books').facet_search('genres', 'c')
+get_separator_tokens_1: |-
+  client.index('articles').get_separator_tokens()
+update_separator_tokens_1: |-
+  client.index('articles').update_separator_tokens(["|", "…"])
+reset_separator_tokens_1: |-
+  client.index('articles').reset_separator_tokens()
+get_non_separator_tokens_1: |-
+  client.index('articles').get_non_separator_tokens()
+update_non_separator_tokens_1: |-
+  client.index('articles').update_non_separator_tokens(["@", "#"])
+reset_non_separator_tokens_1: |-
+  client.index('articles').reset_non_separator_tokens()
 get_dictionary_1: |-
   client.index('books').get_dictionary()
 update_dictionary_1: |-

meilisearch/config.py

@@ -34,6 +34,8 @@ class Paths:
         pagination = "pagination"
         faceting = "faceting"
         dictionary = "dictionary"
+        separator_tokens = "separator-tokens"
+        non_separator_tokens = "non-separator-tokens"
         swap = "swap-indexes"
 
     def __init__(

meilisearch/index.py

@@ -1637,6 +1637,124 @@ def reset_dictionary(self) -> TaskInfo:
 
         return TaskInfo(**task)
 
+    # TEXT SEPARATOR SUB-ROUTES
+
+    def get_separator_tokens(self) -> List[str]:
+        """Get the additional text separator tokens set on this index.
+
+        Returns
+        -------
+        settings:
+            List containing the separator tokens of the index.
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        return self.http.get(self.__settings_url_for(self.config.paths.separator_tokens))
+
+    def get_non_separator_tokens(self) -> List[str]:
+        """Get the list of disabled text separator tokens on this index.
+
+        Returns
+        -------
+        settings:
+            List containing the disabled separator tokens of the index.
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        return self.http.get(self.__settings_url_for(self.config.paths.non_separator_tokens))
+
+    def update_separator_tokens(self, body: Union[List[str], None]) -> TaskInfo:
+        """Update the additional separator tokens of the index.
+
+        Parameters
+        ----------
+        body:
+            List of the new separator tokens.
+
+        Returns
+        -------
+        task_info:
+            TaskInfo instance containing information about a task to track the progress of an asynchronous process.
+            https://www.meilisearch.com/docs/reference/api/tasks#get-one-task
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        task = self.http.put(self.__settings_url_for(self.config.paths.separator_tokens), body)
+
+        return TaskInfo(**task)
+
+    def update_non_separator_tokens(self, body: Union[List[str], None]) -> TaskInfo:
+        """Update the disabled separator tokens of the index.
+
+        Parameters
+        ----------
+        body:
+            List of the newly disabled separator tokens.
+
+        Returns
+        -------
+        task_info:
+            TaskInfo instance containing information about a task to track the progress of an asynchronous process.
+            https://www.meilisearch.com/docs/reference/api/tasks#get-one-task
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        task = self.http.put(self.__settings_url_for(self.config.paths.non_separator_tokens), body)
+
+        return TaskInfo(**task)
+
+    def reset_separator_tokens(self) -> TaskInfo:
+        """Clear all additional separator tokens
+
+        Returns
+        -------
+        task_info:
+            TaskInfo instance containing information about a task to track the progress of an asynchronous process.
+            https://www.meilisearch.com/docs/reference/api/tasks#get-one-task
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        task = self.http.delete(
+            self.__settings_url_for(self.config.paths.separator_tokens),
+        )
+
+        return TaskInfo(**task)
+
+    def reset_non_separator_tokens(self) -> TaskInfo:
+        """Clear all disabled separator tokens
+
+        Returns
+        -------
+        task_info:
+            TaskInfo instance containing information about a task to track the progress of an asynchronous process.
+            https://www.meilisearch.com/docs/reference/api/tasks#get-one-task
+
+        Raises
+        ------
+        MeilisearchApiError
+            An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
+        """
+        task = self.http.delete(
+            self.__settings_url_for(self.config.paths.non_separator_tokens),
+        )
+
+        return TaskInfo(**task)
+
     @staticmethod
     def _batch(
         documents: List[Dict[str, Any]], batch_size: int

tests/settings/test_settings_text_separators_meilisearch.py

@@ -0,0 +1,76 @@
+NEW_SEPARATOR_TOKENS = ["|", "…"]
+NEW_NON_SEPARATOR_TOKENS = ["@", "#"]
+
+
+def test_get_separator_tokens_default(empty_index):
+    """Tests getting the default value of separator tokens."""
+    separator_tokens = empty_index().get_separator_tokens()
+    assert separator_tokens == []
+
+
+def test_get_non_separator_tokens_default(empty_index):
+    """Tests getting the default value of separator tokens."""
+    non_separator_tokens = empty_index().get_separator_tokens()
+    assert non_separator_tokens == []
+
+
+def test_update_separator_tokens(empty_index):
+    """Tests updating the separator tokens."""
+    index = empty_index()
+    task = index.update_separator_tokens(NEW_SEPARATOR_TOKENS)
+    task = index.wait_for_task(task.task_uid)
+    assert task.status == "succeeded"
+
+    separator_tokens = index.get_separator_tokens()
+    for token in NEW_SEPARATOR_TOKENS:
+        assert token in separator_tokens
+
+
+def test_update_non_separator_tokens(empty_index):
+    """Tests updating the non separator tokens."""
+    index = empty_index()
+    task = index.update_non_separator_tokens(NEW_NON_SEPARATOR_TOKENS)
+    task = index.wait_for_task(task.task_uid)
+    assert task.status == "succeeded"
+
+    non_separator_tokens = index.get_non_separator_tokens()
+    for token in NEW_NON_SEPARATOR_TOKENS:
+        assert token in non_separator_tokens
+
+
+def test_reset_separator_tokens(empty_index):
+    """Tests resetting the separator tokens to its default empty list."""
+    index = empty_index()
+    task = index.update_separator_tokens(NEW_SEPARATOR_TOKENS)
+    task = index.wait_for_task(task.task_uid)
+    assert task.status == "succeeded"
+
+    separator_tokens = index.get_separator_tokens()
+    for token in NEW_SEPARATOR_TOKENS:
+        assert token in separator_tokens
+
+    reset_task = index.reset_separator_tokens()
+    reset_task = index.wait_for_task(reset_task.task_uid)
+    assert reset_task.status == "succeeded"
+
+    separator_tokens = index.get_separator_tokens()
+    assert separator_tokens == []
+
+
+def test_non_reset_separator_tokens(empty_index):
+    """Tests resetting the separator tokens to its default empty list."""
+    index = empty_index()
+    task = index.update_non_separator_tokens(NEW_NON_SEPARATOR_TOKENS)
+    task = index.wait_for_task(task.task_uid)
+    assert task.status == "succeeded"
+
+    non_separator_tokens = index.get_non_separator_tokens()
+    for token in NEW_NON_SEPARATOR_TOKENS:
+        assert token in non_separator_tokens
+
+    reset_task = index.reset_non_separator_tokens()
+    reset_task = index.wait_for_task(reset_task.task_uid)
+    assert reset_task.status == "succeeded"
+
+    non_separator_tokens = index.get_non_separator_tokens()
+    assert non_separator_tokens == []