feat: add Parse JSON component

pyproject.toml

@@ -107,6 +107,7 @@ spider-client = "^0.0.27"
 nltk = "^3.9.1"
 bson = "^0.5.10"
 lark = "^1.2.2"
+jq = "^1.8.0"
 
 
 [tool.poetry.group.dev.dependencies]

src/backend/base/langflow/components/helpers/ParseJSONData.py

@@ -0,0 +1,74 @@
+import json
+from json import JSONDecodeError
+
+import jq
+from json_repair import repair_json
+
+from langflow.custom import Component
+from langflow.inputs import HandleInput, MessageTextInput
+from langflow.io import Output
+from langflow.schema import Data
+from langflow.schema.message import Message
+
+
+class ParseJSONDataComponent(Component):
+    display_name = "Parse JSON"
+    description = "Convert and extract JSON fields."
+    icon = "braces"
+    name = "ParseJSONData"
+
+    inputs = [
+        HandleInput(
+            name="input_value",
+            display_name="Input",
+            info="Data object to filter.",
+            required=True,
+            input_types=["Message", "Data"],
+        ),
+        MessageTextInput(
+            name="query",
+            display_name="JQ Query",
+            info="JQ Query to filter the data. The input is always a JSON list.",
+            required=True,
+        ),
+    ]
+
+    outputs = [
+        Output(display_name="Filtered Data", name="filtered_data", method="filter_data"),
+    ]
+
+    def _parse_data(self, input_value) -> str:
+        if isinstance(input_value, Message) and isinstance(input_value.text, str):
+            return input_value.text
+        if isinstance(input_value, Data):
+            return json.dumps(input_value.data)
+        return str(input_value)
+
+    def filter_data(self) -> list[Data]:
+        to_filter = self.input_value
+        if not to_filter:
+            return []
+        if isinstance(to_filter, list):
+            to_filter = [self._parse_data(f) for f in to_filter]
+        else:
+            to_filter = [self._parse_data(to_filter)]
+
+        to_filter = [repair_json(f) for f in to_filter]
+        to_filter_as_dict = []
+        for f in to_filter:
+            try:
+                to_filter_as_dict.append(json.loads(f))
+            except JSONDecodeError:
+                try:
+                    to_filter_as_dict.append(json.loads(repair_json(f)))
+                except JSONDecodeError as e:
+                    raise ValueError(f"Invalid JSON: {e}")
+
+        full_filter_str = json.dumps(to_filter_as_dict)
+
+        print("to_filter: ", to_filter)
+
+        results = jq.compile(self.query).input_text(full_filter_str).all()
+        print("results: ", results)
+        docs = [Data(data=value) if isinstance(value, dict) else Data(text=str(value)) for value in results]
+        return docs

src/backend/base/langflow/initial_setup/starter_projects/Agent Flow.json

@@ -414,7 +414,7 @@
                 "show": true,
                 "title_case": false,
                 "type": "code",
-                "value": "from langflow.base.io.chat import ChatComponent\nfrom langflow.inputs import BoolInput\nfrom langflow.io import DropdownInput, MessageTextInput, Output\nfrom langflow.memory import store_message\nfrom langflow.schema.message import Message\nfrom langflow.utils.constants import MESSAGE_SENDER_NAME_AI, MESSAGE_SENDER_USER, MESSAGE_SENDER_AI\n\n\nclass ChatOutput(ChatComponent):\n    display_name = \"Chat Output\"\n    description = \"Display a chat message in the Playground.\"\n    icon = \"ChatOutput\"\n    name = \"ChatOutput\"\n\n    inputs = [\n        MessageTextInput(\n            name=\"input_value\",\n            display_name=\"Text\",\n            info=\"Message to be passed as output.\",\n        ),\n        BoolInput(\n            name=\"should_store_message\",\n            display_name=\"Store Messages\",\n            info=\"Store the message in the history.\",\n            value=True,\n            advanced=True,\n        ),\n        DropdownInput(\n            name=\"sender\",\n            display_name=\"Sender Type\",\n            options=[MESSAGE_SENDER_AI, MESSAGE_SENDER_USER],\n            value=MESSAGE_SENDER_AI,\n            advanced=True,\n            info=\"Type of sender.\",\n        ),\n        MessageTextInput(\n            name=\"sender_name\",\n            display_name=\"Sender Name\",\n            info=\"Name of the sender.\",\n            value=MESSAGE_SENDER_NAME_AI,\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"session_id\",\n            display_name=\"Session ID\",\n            info=\"The session ID of the chat. If empty, the current session ID parameter will be used.\",\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"data_template\",\n            display_name=\"Data Template\",\n            value=\"{text}\",\n            advanced=True,\n            info=\"Template to convert Data to Text. If left empty, it will be dynamically set to the Data's text key.\",\n        ),\n    ]\n    outputs = [\n        Output(display_name=\"Message\", name=\"message\", method=\"message_response\"),\n    ]\n\n    def message_response(self) -> Message:\n        message = Message(\n            text=self.input_value,\n            sender=self.sender,\n            sender_name=self.sender_name,\n            session_id=self.session_id,\n        )\n        if (\n            self.session_id\n            and isinstance(message, Message)\n            and isinstance(message.text, str)\n            and self.should_store_message\n        ):\n            store_message(\n                message,\n                flow_id=self.graph.flow_id,\n            )\n            self.message.value = message\n\n        self.status = message\n        return message\n"
+                "value": "from langflow.base.io.chat import ChatComponent\nfrom langflow.inputs import BoolInput\nfrom langflow.io import DropdownInput, MessageTextInput, Output\nfrom langflow.memory import store_message\nfrom langflow.schema.message import Message\nfrom langflow.utils.constants import MESSAGE_SENDER_AI, MESSAGE_SENDER_NAME_AI, MESSAGE_SENDER_USER\n\n\nclass ChatOutput(ChatComponent):\n    display_name = \"Chat Output\"\n    description = \"Display a chat message in the Playground.\"\n    icon = \"ChatOutput\"\n    name = \"ChatOutput\"\n\n    inputs = [\n        MessageTextInput(\n            name=\"input_value\",\n            display_name=\"Text\",\n            info=\"Message to be passed as output.\",\n        ),\n        BoolInput(\n            name=\"should_store_message\",\n            display_name=\"Store Messages\",\n            info=\"Store the message in the history.\",\n            value=True,\n            advanced=True,\n        ),\n        DropdownInput(\n            name=\"sender\",\n            display_name=\"Sender Type\",\n            options=[MESSAGE_SENDER_AI, MESSAGE_SENDER_USER],\n            value=MESSAGE_SENDER_AI,\n            advanced=True,\n            info=\"Type of sender.\",\n        ),\n        MessageTextInput(\n            name=\"sender_name\",\n            display_name=\"Sender Name\",\n            info=\"Name of the sender.\",\n            value=MESSAGE_SENDER_NAME_AI,\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"session_id\",\n            display_name=\"Session ID\",\n            info=\"The session ID of the chat. If empty, the current session ID parameter will be used.\",\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"data_template\",\n            display_name=\"Data Template\",\n            value=\"{text}\",\n            advanced=True,\n            info=\"Template to convert Data to Text. If left empty, it will be dynamically set to the Data's text key.\",\n        ),\n    ]\n    outputs = [\n        Output(display_name=\"Message\", name=\"message\", method=\"message_response\"),\n    ]\n\n    def message_response(self) -> Message:\n        message = Message(\n            text=self.input_value,\n            sender=self.sender,\n            sender_name=self.sender_name,\n            session_id=self.session_id,\n        )\n        if (\n            self.session_id\n            and isinstance(message, Message)\n            and isinstance(message.text, str)\n            and self.should_store_message\n        ):\n            store_message(\n                message,\n                flow_id=self.graph.flow_id,\n            )\n            self.message.value = message\n\n        self.status = message\n        return message\n"
               },
               "data_template": {
                 "_input_type": "MessageTextInput",

src/backend/base/langflow/initial_setup/starter_projects/Travel Planning Agents.json

@@ -653,7 +653,7 @@
                 "show": true,
                 "title_case": false,
                 "type": "code",
-                "value": "from langflow.base.io.chat import ChatComponent\nfrom langflow.inputs import BoolInput\nfrom langflow.io import DropdownInput, MessageTextInput, Output\nfrom langflow.memory import store_message\nfrom langflow.schema.message import Message\nfrom langflow.utils.constants import MESSAGE_SENDER_NAME_AI, MESSAGE_SENDER_USER, MESSAGE_SENDER_AI\n\n\nclass ChatOutput(ChatComponent):\n    display_name = \"Chat Output\"\n    description = \"Display a chat message in the Playground.\"\n    icon = \"ChatOutput\"\n    name = \"ChatOutput\"\n\n    inputs = [\n        MessageTextInput(\n            name=\"input_value\",\n            display_name=\"Text\",\n            info=\"Message to be passed as output.\",\n        ),\n        BoolInput(\n            name=\"should_store_message\",\n            display_name=\"Store Messages\",\n            info=\"Store the message in the history.\",\n            value=True,\n            advanced=True,\n        ),\n        DropdownInput(\n            name=\"sender\",\n            display_name=\"Sender Type\",\n            options=[MESSAGE_SENDER_AI, MESSAGE_SENDER_USER],\n            value=MESSAGE_SENDER_AI,\n            advanced=True,\n            info=\"Type of sender.\",\n        ),\n        MessageTextInput(\n            name=\"sender_name\",\n            display_name=\"Sender Name\",\n            info=\"Name of the sender.\",\n            value=MESSAGE_SENDER_NAME_AI,\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"session_id\",\n            display_name=\"Session ID\",\n            info=\"The session ID of the chat. If empty, the current session ID parameter will be used.\",\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"data_template\",\n            display_name=\"Data Template\",\n            value=\"{text}\",\n            advanced=True,\n            info=\"Template to convert Data to Text. If left empty, it will be dynamically set to the Data's text key.\",\n        ),\n    ]\n    outputs = [\n        Output(display_name=\"Message\", name=\"message\", method=\"message_response\"),\n    ]\n\n    def message_response(self) -> Message:\n        message = Message(\n            text=self.input_value,\n            sender=self.sender,\n            sender_name=self.sender_name,\n            session_id=self.session_id,\n        )\n        if (\n            self.session_id\n            and isinstance(message, Message)\n            and isinstance(message.text, str)\n            and self.should_store_message\n        ):\n            store_message(\n                message,\n                flow_id=self.graph.flow_id,\n            )\n            self.message.value = message\n\n        self.status = message\n        return message\n"
+                "value": "from langflow.base.io.chat import ChatComponent\nfrom langflow.inputs import BoolInput\nfrom langflow.io import DropdownInput, MessageTextInput, Output\nfrom langflow.memory import store_message\nfrom langflow.schema.message import Message\nfrom langflow.utils.constants import MESSAGE_SENDER_AI, MESSAGE_SENDER_NAME_AI, MESSAGE_SENDER_USER\n\n\nclass ChatOutput(ChatComponent):\n    display_name = \"Chat Output\"\n    description = \"Display a chat message in the Playground.\"\n    icon = \"ChatOutput\"\n    name = \"ChatOutput\"\n\n    inputs = [\n        MessageTextInput(\n            name=\"input_value\",\n            display_name=\"Text\",\n            info=\"Message to be passed as output.\",\n        ),\n        BoolInput(\n            name=\"should_store_message\",\n            display_name=\"Store Messages\",\n            info=\"Store the message in the history.\",\n            value=True,\n            advanced=True,\n        ),\n        DropdownInput(\n            name=\"sender\",\n            display_name=\"Sender Type\",\n            options=[MESSAGE_SENDER_AI, MESSAGE_SENDER_USER],\n            value=MESSAGE_SENDER_AI,\n            advanced=True,\n            info=\"Type of sender.\",\n        ),\n        MessageTextInput(\n            name=\"sender_name\",\n            display_name=\"Sender Name\",\n            info=\"Name of the sender.\",\n            value=MESSAGE_SENDER_NAME_AI,\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"session_id\",\n            display_name=\"Session ID\",\n            info=\"The session ID of the chat. If empty, the current session ID parameter will be used.\",\n            advanced=True,\n        ),\n        MessageTextInput(\n            name=\"data_template\",\n            display_name=\"Data Template\",\n            value=\"{text}\",\n            advanced=True,\n            info=\"Template to convert Data to Text. If left empty, it will be dynamically set to the Data's text key.\",\n        ),\n    ]\n    outputs = [\n        Output(display_name=\"Message\", name=\"message\", method=\"message_response\"),\n    ]\n\n    def message_response(self) -> Message:\n        message = Message(\n            text=self.input_value,\n            sender=self.sender,\n            sender_name=self.sender_name,\n            session_id=self.session_id,\n        )\n        if (\n            self.session_id\n            and isinstance(message, Message)\n            and isinstance(message.text, str)\n            and self.should_store_message\n        ):\n            store_message(\n                message,\n                flow_id=self.graph.flow_id,\n            )\n            self.message.value = message\n\n        self.status = message\n        return message\n"
               },
               "data_template": {
                 "_input_type": "MessageTextInput",

src/backend/tests/integration/components/astra/test_astra_component.py

@@ -1,14 +1,11 @@
 import os
-from typing import List
 
 from astrapy.db import AstraDB
 import pytest
 
 from langflow.components.embeddings import OpenAIEmbeddingsComponent
-from langflow.custom import Component
-from langflow.inputs import StrInput
-from langflow.template import Output
 from tests.api_keys import get_astradb_application_token, get_astradb_api_endpoint, get_openai_api_key
+from tests.integration.components.mock_components import TextToData
 from tests.integration.utils import ComponentInputHandle
 from langchain_core.documents import Document
 
@@ -70,14 +67,6 @@ async def test_base(astradb_client: AstraDB):
     assert astradb_client.collection(BASIC_COLLECTION)
 
 
-class TextToData(Component):
-    inputs = [StrInput(name="text_data", is_list=True)]
-    outputs = [Output(name="data", display_name="Data", method="create_data")]
-
-    def create_data(self) -> List[Data]:
-        return [Data(text=t) for t in self.text_data]
-
-
 @pytest.mark.api_key_required
 @pytest.mark.asyncio
 async def test_astra_embeds_and_search():
@@ -93,7 +82,7 @@ async def test_astra_embeds_and_search():
             "number_of_results": 1,
             "search_input": "test1",
             "ingest_data": ComponentInputHandle(
-                clazz=TextToData, inputs={"text_data": ["test1", "test2"]}, output_name="data"
+                clazz=TextToData, inputs={"text_data": ["test1", "test2"]}, output_name="from_text"
             ),
             "embedding": ComponentInputHandle(
                 clazz=OpenAIEmbeddingsComponent,

src/backend/tests/integration/components/helpers/test_parse_json_data.py

@@ -0,0 +1,57 @@
+import pytest
+
+from langflow.components.helpers.ParseJSONData import ParseJSONDataComponent
+from langflow.components.inputs import ChatInput
+from langflow.schema import Data
+from tests.integration.components.mock_components import TextToData
+from tests.integration.utils import run_single_component, ComponentInputHandle
+
+
+@pytest.mark.asyncio
+async def test_from_data():
+    outputs = await run_single_component(
+        ParseJSONDataComponent,
+        inputs={
+            "input_value": ComponentInputHandle(
+                clazz=TextToData, inputs={"text_data": ['{"key":"value1"}'], "is_json": True}, output_name="from_text"
+            ),
+            "query": ".[0].key",
+        },
+    )
+    assert outputs["filtered_data"] == [Data(text="value1")]
+
+    outputs = await run_single_component(
+        ParseJSONDataComponent,
+        inputs={
+            "input_value": ComponentInputHandle(
+                clazz=TextToData,
+                inputs={"text_data": ['{"key":[{"field1": 1, "field2": 2}]}'], "is_json": True},
+                output_name="from_text",
+            ),
+            "query": ".[0].key.[0].field2",
+        },
+    )
+    assert outputs["filtered_data"] == [Data(text="2")]
+
+
+@pytest.mark.asyncio
+async def test_from_message():
+    outputs = await run_single_component(
+        ParseJSONDataComponent,
+        inputs={
+            "input_value": ComponentInputHandle(clazz=ChatInput, inputs={}, output_name="message"),
+            "query": ".[0].key",
+        },
+        run_input="{'key':'value1'}",
+    )
+    assert outputs["filtered_data"] == [Data(text="value1")]
+
+    outputs = await run_single_component(
+        ParseJSONDataComponent,
+        inputs={
+            "input_value": ComponentInputHandle(clazz=ChatInput, inputs={}, output_name="message"),
+            "query": ".[0].key.[0].field2",
+        },
+        run_input='{"key":[{"field1": 1, "field2": 2}]}',
+    )
+    assert outputs["filtered_data"] == [Data(text="2")]

src/backend/tests/integration/components/mock_components.py

@@ -0,0 +1,25 @@
+import json
+from typing import List
+
+from langflow.custom import Component
+from langflow.inputs import StrInput, BoolInput
+from langflow.schema import Data
+from langflow.template import Output
+
+
+class TextToData(Component):
+    inputs = [
+        StrInput(name="text_data", is_list=True),
+        BoolInput(name="is_json", info="Parse text_data as json and fill the data object."),
+    ]
+    outputs = [
+        Output(name="from_text", display_name="From text", method="create_data"),
+    ]
+
+    def _to_data(self, text: str) -> Data:
+        if self.is_json:
+            return Data(data=json.loads(text))
+        return Data(text=text)
+
+    def create_data(self) -> List[Data]:
+        return [self._to_data(t) for t in self.text_data]