v0.12.23 (#18050)

run-llama · Mar 7, 2025 · 4c8d1d6 · 4c8d1d6
1 parent 54e0a2c
commit 4c8d1d6
Show file tree

Hide file tree

Showing 10 changed files with 313 additions and 310 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,60 @@
 # ChangeLog
 
+## [2025-03-07]
+
+### `llama-index-core` [0.12.23]
+
+- added `merging_separator` argument to allow for specifying chunk merge separator in semantic splitter (#18027)
+- Add support for running single-agent workflows within the BaseWorkflowAgent class (#18038)
+- Fix the error raised when ReactAgent is created without an explicit system message (#18041)
+- add a field keep_whitespaces to TokenTextSplitter (#17998)
+- do not convert raw tool output to string in AgentWorkflow (#18006)
+
+### `llama-index-embeddings-ollama` [0.6.0]
+
+- feat: add client_kwargs Parameter to OllamaEmbedding Class (#18012)
+
+### `llama-index-llms-anthropic` [0.6.10]
+
+- anthropic caching and thinking updates (#18039)
+- allow caching of tool results (#18028)
+- support caching of anthropic system prompt (#18008)
+- Ensure resuming a workflow actually works (#18023)
+- [MarkdownNodeParser] Adding customizable header path separator char (#17964)
+- feat: return event instance from run() when stop event is custom (#18001)
+
+### `llama-index-llms-azure-openai` [0.3.2]
+
+- AzureOpenAI: api_base and azure_endpoint are mutually exclusive (#18037)
+- Add base_url to AzureOpenAI (#17996)
+
+### `llama-index-llms-bedrock-converse` [0.4.8]
+
+- message text is required in boto3 model (#17989)
+
+### `llama-index-llms-ollama` [0.5.3]
+
+- Make request_timeout in Ollama LLM optional (#18007)
+
+### `llama-index-llms-mistralai` [0.4.0]
+
+- MistralAI support for multImodal content blocks (#17997)
+
+### `llama-index-readers-file` [0.4.6]
+
+- Bugfix: Use `torch.no grad()` in inference in ImageVisionLLMReader when PyTorch is installed (#17970)
+
+### `llama-index-storage-chat-store-mongo` [0.1.0]
+
+- Feat/mongo chat store (#17979)
+
+### `llama-index-core` [0.12.23]
+
+- added `merging_separator` argument to allow for specifying chunk merge separator in semantic splitter (#18027)
+- Add support for running single-agent workflows within the BaseWorkflowAgent class (#18038)
+- Fix the error raised when ReactAgent is created without an explicit system message (#18041)
+- add a field keep_whitespaces to TokenTextSplitter (#17998)
+
 ## [2025-02-28]
 
 ### `llama-index-core` [0.12.22]

diff --git a/docs/docs/CHANGELOG.md b/docs/docs/CHANGELOG.md
@@ -1,5 +1,60 @@
 # ChangeLog
 
+## [2025-03-07]
+
+### `llama-index-core` [0.12.23]
+
+- added `merging_separator` argument to allow for specifying chunk merge separator in semantic splitter (#18027)
+- Add support for running single-agent workflows within the BaseWorkflowAgent class (#18038)
+- Fix the error raised when ReactAgent is created without an explicit system message (#18041)
+- add a field keep_whitespaces to TokenTextSplitter (#17998)
+- do not convert raw tool output to string in AgentWorkflow (#18006)
+
+### `llama-index-embeddings-ollama` [0.6.0]
+
+- feat: add client_kwargs Parameter to OllamaEmbedding Class (#18012)
+
+### `llama-index-llms-anthropic` [0.6.10]
+
+- anthropic caching and thinking updates (#18039)
+- allow caching of tool results (#18028)
+- support caching of anthropic system prompt (#18008)
+- Ensure resuming a workflow actually works (#18023)
+- [MarkdownNodeParser] Adding customizable header path separator char (#17964)
+- feat: return event instance from run() when stop event is custom (#18001)
+
+### `llama-index-llms-azure-openai` [0.3.2]
+
+- AzureOpenAI: api_base and azure_endpoint are mutually exclusive (#18037)
+- Add base_url to AzureOpenAI (#17996)
+
+### `llama-index-llms-bedrock-converse` [0.4.8]
+
+- message text is required in boto3 model (#17989)
+
+### `llama-index-llms-ollama` [0.5.3]
+
+- Make request_timeout in Ollama LLM optional (#18007)
+
+### `llama-index-llms-mistralai` [0.4.0]
+
+- MistralAI support for multImodal content blocks (#17997)
+
+### `llama-index-readers-file` [0.4.6]
+
+- Bugfix: Use `torch.no grad()` in inference in ImageVisionLLMReader when PyTorch is installed (#17970)
+
+### `llama-index-storage-chat-store-mongo` [0.1.0]
+
+- Feat/mongo chat store (#17979)
+
+### `llama-index-core` [0.12.23]
+
+- added `merging_separator` argument to allow for specifying chunk merge separator in semantic splitter (#18027)
+- Add support for running single-agent workflows within the BaseWorkflowAgent class (#18038)
+- Fix the error raised when ReactAgent is created without an explicit system message (#18041)
+- add a field keep_whitespaces to TokenTextSplitter (#17998)
+
 ## [2025-02-28]
 
 ### `llama-index-core` [0.12.22]

diff --git a/docs/docs/api_reference/storage/chat_store/mongo.md b/docs/docs/api_reference/storage/chat_store/mongo.md
@@ -0,0 +1,4 @@
+::: llama_index.storage.chat_store.mongo
+    options:
+      members:
+        - MongoChatStore
diff --git a/docs/mkdocs.yml b/docs/mkdocs.yml
@@ -1539,6 +1539,7 @@ nav:
               - ./api_reference/storage/chat_store/azurecosmosnosql.md
               - ./api_reference/storage/chat_store/dynamodb.md
               - ./api_reference/storage/chat_store/index.md
+              - ./api_reference/storage/chat_store/mongo.md
               - ./api_reference/storage/chat_store/postgres.md
               - ./api_reference/storage/chat_store/redis.md
               - ./api_reference/storage/chat_store/simple.md
@@ -2402,6 +2403,8 @@ plugins:
             - ../llama-index-integrations/retrievers/llama-index-retrievers-tldw
             - ../llama-index-integrations/tools/llama-index-tools-valyu
             - ../llama-index-integrations/postprocessor/llama-index-postprocessor-ibm
+            - ../llama-index-integrations/storage/chat_store/llama-index-storage-chat-store-mongo
+            - ../llama-index-integrations/llms/llama-index-llms-google-genai
   - redirects:
       redirect_maps:
         ./api/llama_index.vector_stores.MongoDBAtlasVectorSearch.html: api_reference/storage/vector_store/mongodb.md

diff --git a/llama-index-core/llama_index/core/__init__.py b/llama-index-core/llama_index/core/__init__.py
@@ -1,6 +1,6 @@
 """Init file of LlamaIndex."""
 
-__version__ = "0.12.22"
+__version__ = "0.12.23.post2"
 
 import logging
 from logging import NullHandler

diff --git a/llama-index-core/llama_index/core/readers/file/base.py b/llama-index-core/llama_index/core/readers/file/base.py
@@ -121,7 +121,13 @@ def _format_file_timestamp(
     if timestamp is None:
         return None
 
-    timestamp_dt = datetime.fromtimestamp(timestamp, tz=timezone.utc)
+    # Convert timestamp to UTC
+    # Check if timestamp is already a datetime object
+    if isinstance(timestamp, datetime):
+        timestamp_dt = timestamp.astimezone(timezone.utc)
+    else:
+        timestamp_dt = datetime.fromtimestamp(timestamp, tz=timezone.utc)
+
     if include_time:
         return timestamp_dt.strftime("%Y-%m-%dT%H:%M:%SZ")
     return timestamp_dt.strftime("%Y-%m-%d")

diff --git a/llama-index-core/pyproject.toml b/llama-index-core/pyproject.toml
@@ -46,7 +46,7 @@ name = "llama-index-core"
 packages = [{include = "llama_index"}]
 readme = "README.md"
 repository = "https://github.com/run-llama/llama_index"
-version = "0.12.22"
+version = "0.12.23.post2"
 
 [tool.poetry.dependencies]
 SQLAlchemy = {extras = ["asyncio"], version = ">=1.4.49"}

diff --git a/llama-index-integrations/llms/llama-index-llms-oci-genai/pyproject.toml b/llama-index-integrations/llms/llama-index-llms-oci-genai/pyproject.toml
@@ -34,7 +34,7 @@ readme = "README.md"
 version = "0.4.0"
 
 [tool.poetry.dependencies]
-python = ">=3.8.1,<4.0"
+python = ">=3.9,<4.0"
 oci = "^2.134.0"
 llama-index-core = "^0.12.0"