langchain - ✅(Solved) Fix TypeError in merge_lists when streaming Mistral responses with inline citations [9 pull requests, 5 comments, 5 participants]

GhaziBenDahmane · 2026-02-16T09:20:39Z

[langchain] Model: mistral-medium-2505 but likely affects any Mistral model returning citations The error occurs when: - Using LangGraph's create react agent w… Model: mistral-medium-2505 (but likely affects any Mistral model returning citations) The error occurs when: - Using LangGraph's create_react_agent with Mistral - Streaming mode enabled - Tools like Tavily search that return results the model cites # PR #35265: fix(core): handle non-dict elements in merge_lists index lookup - Repository: langchain-ai/langchain - Author: nileshhadalgi016 - State: open | merged: False - Link: https://github.com/langchain-ai/langchain/pull/35265 ## Description (problem / solution / changelog) ## Description Fixes #35259 When streaming Mistral responses with inline citations, the content list can contain a mix of `str` and `dict` elements. For example: ```python [ "start answer...", {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]}, "other answer..." ] ``` The `merge_lists` function in `_merge.py` assumed all elements in the `merged` list would be dicts when searching for a matching `index`. When it encountered a `str` element, the expression `"index" in e_left` would check for substring containment (valid but wrong), and then `e_left["index"]` would fail with `TypeError: string indices must be integers, not 'str'`. ## Fix Added an `isinstance(e_left, dict)` guard in the list comprehension (line 114) that finds merge targets by index. This ensures only dict elements are checked for matching `index` keys, skipping strings and other non-dict types. ## Tests Added 2 new parametrized test cases to `test_merge_lists`: 1. **Mixed str and dict with matching index** - verifies dicts with the same index are correctly merged while strings are preserved 2. **Mixed str and dict with non-matching index** - verifies new dict elements are appended when no matching index exists ## Issue This affects any Mistral model returning inline citations (e.g., `mistral-medium-2505`) when used with streaming and tools like Tavily search. ## Dependencies None ## Changed files - `libs/core/langchain_core/utils/_merge.py` (modified, +3/-1) - `libs/core/tests/unit_tests/utils/test_utils.py` (modified, +35/-0) --- # PR #35271: fix: TypeError in merge_lists with mixed content types & pretty_repr HTML for non-string content - Repository: langchain-ai/langchain - Author: quantumnic - State: closed | merged: False - Link: https://github.com/langchain-ai/langchain/pull/35271 ## Description (problem / solution / changelog) ## Fixes ### 1. TypeError in `merge_lists` when streaming Mistral responses with inline citations (#35259) **Problem:** When `merge_lists` processes a list containing both strings and dicts with `"index"` keys, the list comprehension tries to access `e_left["index"]` on string elements, causing `TypeError: string indices must be integers, not 'str'`. **Fix:** Added `isinstance(e_left, dict)` guard before checking for `"index"` key in the list comprehension. ### 2. `pretty_repr(html=True)` does not return HTML for non-string content (#34875) **Problem:** When `BaseMessage.content` is a list (valid type), `pretty_repr(html=True)` returns the raw Python repr with no HTML formatting, despite the docstring promising HTML output. **Fix:** - Parse list content blocks (text, image_url, plain strings) into readable text - Apply HTML escaping and line break conversion when `html=True` - Resolves the existing TODO comment in the code ## Changed files - `libs/core/langchain_core/messages/base.py` (modified, +29/-2) - `libs/core/langchain_core/utils/_merge.py` (modified, +3/-1) --- # PR #35290: fix(core): fix TypeError in merge_lists when merged list contains non-dict elements - Repository: langchain-ai/langchain - Author: nightcityblade - State: closed | merged: False - Link: https://github.com/langchain-ai/langchain/pull/35290 ## Description (problem / solution / changelog) ## Description Fixes #35259. When streaming Mistral responses with inline citations, `merge_lists` can encounter a mixed list of strings and dicts. The list comprehension that searches for matching `index` values assumed all elements in `merged` were dicts, causing: ``` TypeError: string indices must be integers, not 'str' ``` ## Changes Added `isinstance(e_left, dict)` check in the list comprehension before accessing dict keys, so string elements are safely skipped. ## Testing Reproduces with the exact example from the issue: ```python from langchain_core.utils._merge import merge_lists merged = ["start answer...", {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]}, "other answer..."] other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}] result = merge_lists(merged, other) # no longer raises TypeError ``` ## Changed files - `libs/core/langchain_core/utils/_merge.py` (modified, +3/-1) --- # PR #35349: fix(core): handle non-dict elements in merge_lists index lookup - Repository: langchain-ai/langchain - Author: hztBUAA - State:

langchain2026-02-16 09:20:39

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

langchain-ai/langchain#35259•Fetched 2026-04-08 00:26:54

View on GitHub

Comments

Participants

Timeline

Reactions

Author

Participants

Assignees

Timeline (top)

cross-referenced ×9referenced ×7commented ×5labeled ×4

Model: mistral-medium-2505 (but likely affects any Mistral model returning citations)

The error occurs when:

Using LangGraph's create_react_agent with Mistral
Streaming mode enabled
Tools like Tavily search that return results the model cites

Error Message

File ".../langchain_core/utils/_merge.py", line 114, in <listcomp> if "index" in e_left and e_left["index"] == e["index"] ~~~~~~~^^^^^^^^^ TypeError: string indices must be integers, not 'str'

Root Cause

Model: mistral-medium-2505 (but likely affects any Mistral model returning citations)

The error occurs when:

Using LangGraph's create_react_agent with Mistral
Streaming mode enabled
Tools like Tavily search that return results the model cites

Fix Action

Fix / Workaround

This is a bug, not a usage question.
I added a clear and descriptive title that summarizes this issue.
I used the GitHub search to find a similar question and didn't find it.
I am sure that this is a bug in LangChain rather than my code.
The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
This is not related to the langchain-community package.
I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Other Dependencies

aiohttp: 3.11.14 dataclasses-json: 0.6.7 elasticsearch: 8.19.3 httpx: 0.28.1 httpx-sse: 0.4.1 jsonpatch: 1.33 langgraph: 1.0.5 mcp: 1.17.0 numpy: 2.2.6 openai: 2.14.0 opentelemetry-api: 1.34.1 opentelemetry-exporter-otlp-proto-http: 1.34.1 opentelemetry-sdk: 1.34.1 orjson: 3.10.18 packaging: 24.2 pydantic: 2.11.7 pydantic-settings: 2.10.1 pytest: 7.4.0 pyyaml: 6.0.3 PyYAML: 6.0.3 requests: 2.32.5 requests-toolbelt: 1.0.0 rich: 14.0.0 SQLAlchemy: 2.0.41 sqlalchemy: 2.0.41 tenacity: 8.2.3 tiktoken: 0.9.0 tokenizers: 0.21.2 typing-extensions: 4.14.0 unstructured: 0.18.21 unstructured-client: 0.39.1 uuid-utils: 0.12.0 zstandard: 0.23.0

PR fix notes

PR #35265: fix(core): handle non-dict elements in merge_lists index lookup

Repository: langchain-ai/langchain
Author: nileshhadalgi016
State: open | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35265

Description (problem / solution / changelog)

Description

Fixes #35259

When streaming Mistral responses with inline citations, the content list can contain a mix of str and dict elements. For example:

[
    "start answer...",
    {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]},
    "other answer..."
]

The merge_lists function in _merge.py assumed all elements in the merged list would be dicts when searching for a matching index. When it encountered a str element, the expression "index" in e_left would check for substring containment (valid but wrong), and then e_left["index"] would fail with TypeError: string indices must be integers, not 'str'.

Fix

Added an isinstance(e_left, dict) guard in the list comprehension (line 114) that finds merge targets by index. This ensures only dict elements are checked for matching index keys, skipping strings and other non-dict types.

Tests

Added 2 new parametrized test cases to test_merge_lists:

Mixed str and dict with matching index - verifies dicts with the same index are correctly merged while strings are preserved
Mixed str and dict with non-matching index - verifies new dict elements are appended when no matching index exists

Issue

This affects any Mistral model returning inline citations (e.g., mistral-medium-2505) when used with streaming and tools like Tavily search.

Dependencies

None

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +3/-1)
libs/core/tests/unit_tests/utils/test_utils.py (modified, +35/-0)

PR #35271: fix: TypeError in merge_lists with mixed content types & pretty_repr HTML for non-string content

Repository: langchain-ai/langchain
Author: quantumnic
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35271

Description (problem / solution / changelog)

Fixes

1. TypeError in `merge_lists` when streaming Mistral responses with inline citations (#35259)

Problem: When merge_lists processes a list containing both strings and dicts with "index" keys, the list comprehension tries to access e_left["index"] on string elements, causing TypeError: string indices must be integers, not 'str'.

Fix: Added isinstance(e_left, dict) guard before checking for "index" key in the list comprehension.

2. `pretty_repr(html=True)` does not return HTML for non-string content (#34875)

Problem: When BaseMessage.content is a list (valid type), pretty_repr(html=True) returns the raw Python repr with no HTML formatting, despite the docstring promising HTML output.

Fix:

Parse list content blocks (text, image_url, plain strings) into readable text
Apply HTML escaping and line break conversion when html=True
Resolves the existing TODO comment in the code

Changed files

libs/core/langchain_core/messages/base.py (modified, +29/-2)
libs/core/langchain_core/utils/_merge.py (modified, +3/-1)

PR #35290: fix(core): fix TypeError in merge_lists when merged list contains non-dict elements

Repository: langchain-ai/langchain
Author: nightcityblade
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35290

Description (problem / solution / changelog)

Description

Fixes #35259.

When streaming Mistral responses with inline citations, merge_lists can encounter a mixed list of strings and dicts. The list comprehension that searches for matching index values assumed all elements in merged were dicts, causing:

TypeError: string indices must be integers, not 'str'

Changes

Added isinstance(e_left, dict) check in the list comprehension before accessing dict keys, so string elements are safely skipped.

Testing

Reproduces with the exact example from the issue:

from langchain_core.utils._merge import merge_lists
merged = ["start answer...", {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]}, "other answer..."]
other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}]
result = merge_lists(merged, other)  # no longer raises TypeError

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +3/-1)

PR #35349: fix(core): handle non-dict elements in merge_lists index lookup

Repository: langchain-ai/langchain
Author: hztBUAA
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35349

Description (problem / solution / changelog)

Adds a type guard in merge_lists to skip non-dict elements when looking up the merge index, preventing AttributeError when streaming Mistral citation objects.

Fixes #35259

Generated with the help of AI tools.

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)
libs/core/tests/unit_tests/utils/test_utils.py (modified, +35/-0)

PR #35481: fix(core): guard merge_lists against non-dict elements when matching by index

Repository: langchain-ai/langchain
Author: giulio-leone
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35481

Description (problem / solution / changelog)

Problem

merge_lists crashes with TypeError: string indices must be integers, not 'str' when the merged list contains a mix of strings and dicts with "index" keys.

This happens with Mistral streaming responses that include inline citations — the content array alternates between text strings and reference dicts:

merged = [
    "start answer...",
    {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]},
    "other answer..."
]
other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}]

merge_lists(merged, other)  # TypeError!

Root Cause

The index-matching list comprehension iterates over all elements in merged and checks "index" in e_left. For string elements, Python's in operator checks for substring containment (which works), but then e_left["index"] tries to index a string with a string key, which raises TypeError.

Fix

Add isinstance(e_left, dict) guard before accessing dict-specific attributes.

Tests

Added regression test with mixed strings and dicts (the exact pattern from the Mistral streaming scenario). 81 tests pass.

Fixes #35259

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)
libs/core/tests/unit_tests/utils/test_utils.py (modified, +21/-0)

PR #35623: fix(core): prevent TypeError in merge_lists when list contains mixed strings and dicts

Repository: langchain-ai/langchain
Author: giulio-leone
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35623

Description (problem / solution / changelog)

Bug

merge_lists raises TypeError: string indices must be integers, not 'str' when the merged list contains a mix of plain strings and dicts with an "index" key. This happens during Mistral streaming with inline citations.

Reported in: #35259

Root cause

The guard "index" in e_left at line 121 of _merge.py performs substring containment when e_left is a string, not dict key lookup. If any string in the list happens to contain the substring "index", the check passes and e_left["index"] fails because you cannot index a string with a string key.

Fix

Add isinstance(e_left, dict) before the "index" in e_left check:

# Before
"index" in e_left

# After
isinstance(e_left, dict) and "index" in e_left

Tests

Added test_merge_lists_mixed_str_and_dict_with_index covering the exact scenario from the issue
All 54 merge tests pass with no regressions

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)
libs/core/tests/unit_tests/utils/test_utils.py (modified, +25/-0)

PR #35625: Fix TypeError in merge_lists when handling mixed type lists

Repository: langchain-ai/langchain
Author: 04cb
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35625

Description (problem / solution / changelog)

Fixes #35259. Add isinstance check before accessing dict keys to prevent TypeError when merged list contains mixed types (strings and dicts).

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)

PR #35655: core: fix merge_lists crash when merged list contains non-dict elements

Repository: langchain-ai/langchain
Author: alvinttang
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35655

Description (problem / solution / changelog)

Description

Fixes a bug in merge_lists() where iterating over merged to find elements with a matching "index" key assumes all elements are dicts. However, merged can contain non-dict elements (strings, ints, etc.):

Using "index" in e_left on a string checks for substring membership (unexpected behavior)
Using "index" in e_left on an int raises TypeError: argument of type 'int' is not iterable

This adds an isinstance(e_left, dict) guard before the "index" in e_left check to prevent both cases.

Related: #35259

Issue

Prevents TypeError and unexpected substring matching when non-dict items exist in merged lists during tool call chunk aggregation.

Dependencies

None.

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)

PR #35876: fix: add type check in merge_lists for non-dict elements

Repository: langchain-ai/langchain
Author: alvinttang
State: closed | merged: False
Link: https://github.com/langchain-ai/langchain/pull/35876

Description (problem / solution / changelog)

Summary

Fixes #35259

merge_lists in langchain_core/utils/_merge.py crashes with TypeError: string indices must be integers when merging streaming chunks that contain a mix of strings and dicts (e.g., Mistral responses with inline citations).
Root cause: The list comprehension checks "index" in e_left without first verifying e_left is a dict. When e_left is a string, Python's in operator performs substring matching instead of dict key lookup. If the check passes, e_left["index"] fails because string indexing requires integers, not strings.
Fix: Add isinstance(e_left, dict) guard before the "index" in e_left check to skip non-dict elements during the index-matching scan.

Reproduction

from langchain_core.utils._merge import merge_lists

merged = [
    "start answer...",
    {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]},
    "other answer..."
]
other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}]

# Before fix: TypeError: string indices must be integers
# After fix: works correctly
result = merge_lists(merged, other)

Test plan

Verify the reproduction case from the issue no longer raises TypeError
Existing merge_lists tests continue to pass
Test with Mistral streaming citations end-to-end (requires Mistral API key)

Changed files

libs/core/langchain_core/utils/_merge.py (modified, +2/-1)

Code Example

from langchain_core.utils._merge import merge_lists

merged = [
    "start answer...",
    {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]},
    "other answer..."
]

# Another reference to merge (from streaming chunk)
other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}]

# This raises TypeError
result = merge_lists(merged, other)

---

File ".../langchain_core/utils/_merge.py", line 114, in <listcomp>
      if "index" in e_left and e_left["index"] == e["index"]
                               ~~~~~~~^^^^^^^^^
  TypeError: string indices must be integers, not 'str'

RAW_BUFFERClick to expand / collapse

Checked other resources

This is a bug, not a usage question.
I added a clear and descriptive title that summarizes this issue.
I used the GitHub search to find a similar question and didn't find it.
I am sure that this is a bug in LangChain rather than my code.
The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
This is not related to the langchain-community package.
I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Package (Required)

Related Issues / PRs

No response

Reproduction Steps / Example Code (Python)

from langchain_core.utils._merge import merge_lists

merged = [
    "start answer...",
    {"type": "reference", "index": 0, "reference_ids": ["iKcb2CAQ7"]},
    "other answer..."
]

# Another reference to merge (from streaming chunk)
other = [{"type": "reference", "index": 0, "reference_ids": ["DGqzzmqc3"]}]

# This raises TypeError
result = merge_lists(merged, other)

Error Message and Stack Trace (if applicable)

File ".../langchain_core/utils/_merge.py", line 114, in <listcomp>
      if "index" in e_left and e_left["index"] == e["index"]
                               ~~~~~~~^^^^^^^^^
  TypeError: string indices must be integers, not 'str'

Description

Model: mistral-medium-2505 (but likely affects any Mistral model returning citations)

The error occurs when:

Using LangGraph's create_react_agent with Mistral
Streaming mode enabled
Tools like Tavily search that return results the model cites

System Info

System Information

OS: Linux OS Version: #1 SMP Thu Oct 5 21:02:42 UTC 2023 Python Version: 3.11.13 (main, Jun 20 2025, 10:35:36) [GCC 11.4.0]

Package Information

langchain_core: 1.2.5 langchain: 1.2.0 langchain_community: 0.4.1 langsmith: 0.3.45 langchain_classic: 1.0.1 langchain_elasticsearch: 1.0.0 langchain_experimental: 0.4.1 langchain_mcp_adapters: 0.2.1 langchain_mistralai: 1.1.1 langchain_openai: 1.1.6 langchain_tavily: 0.2.16 langchain_text_splitters: 1.1.0 langchain_unstructured: 1.0.1 langgraph_sdk: 0.3.1 langgraph_supervisor: 0.0.31

Optional packages not installed

langserve

Other Dependencies

aiohttp: 3.11.14 dataclasses-json: 0.6.7 elasticsearch: 8.19.3 httpx: 0.28.1 httpx-sse: 0.4.1 jsonpatch: 1.33 langgraph: 1.0.5 mcp: 1.17.0 numpy: 2.2.6 openai: 2.14.0 opentelemetry-api: 1.34.1 opentelemetry-exporter-otlp-proto-http: 1.34.1 opentelemetry-sdk: 1.34.1 orjson: 3.10.18 packaging: 24.2 pydantic: 2.11.7 pydantic-settings: 2.10.1 pytest: 7.4.0 pyyaml: 6.0.3 PyYAML: 6.0.3 requests: 2.32.5 requests-toolbelt: 1.0.0 rich: 14.0.0 SQLAlchemy: 2.0.41 sqlalchemy: 2.0.41 tenacity: 8.2.3 tiktoken: 0.9.0 tokenizers: 0.21.2 typing-extensions: 4.14.0 unstructured: 0.18.21 unstructured-client: 0.39.1 uuid-utils: 0.12.0 zstandard: 0.23.0

extent analysis

Problem Summary

The issue is a TypeError when merging lists in LangChain's merge_lists function, specifically when dealing with a string index.

Root Cause Analysis

The root cause is likely due to the fact that the merge_lists function is trying to access a string index as if it were an integer.

Fix Plan

Step 1: Update `merge_lists` function to handle string indices

def merge_lists(left, right):
    if isinstance(left, str):
        return [left] + right
    elif isinstance(right, str):
        return left + [right]
    else:
        # existing logic
        pass

Step 2: Update the `merge_lists` function to handle nested lists

def merge_lists(left, right):
    if isinstance(left, str):
        return [left] + right
    elif isinstance(right, str):
        return left + [right]
    elif isinstance(left, list) and isinstance(right, list):
        merged = []
        for e_left, e_right in zip(left, right):
            if isinstance(e_left, dict) and isinstance(e_right, dict):
                merged.append(merge_dicts(e_left, e_right))
            elif isinstance(e_left, list) and isinstance(e_right, list):
                merged.append(merge_lists(e_left, e_right))
            else:
                merged.append(e_left)
                merged.append(e_right)
        return merged
    else:
        # existing logic
        pass

Step 3: Update the `merge_dicts` function to handle string keys

def merge_dicts(left, right):
    merged = {}
    for key in set(list(left.keys()) + list(right.keys())):
        if key in left and key in right:
            if isinstance(left[key], str) and isinstance(right[key], str):
                merged[key] = left[key] + right[key]
            elif isinstance(left[key], list)

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #ssr #installation #tensor shape #autograd error #dependency error #configuration error #environment variable

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

langchain - ✅(Solved) Fix TypeError in merge_lists when streaming Mistral responses with inline citations [9 pull requests, 5 comments, 5 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Fix Action

Fix / Workaround

Other Dependencies

PR fix notes

PR #35265: fix(core): handle non-dict elements in merge_lists index lookup

Description (problem / solution / changelog)

Description

Fix

Tests

Issue

Dependencies

Changed files

PR #35271: fix: TypeError in merge_lists with mixed content types & pretty_repr HTML for non-string content

Description (problem / solution / changelog)

Fixes

1. TypeError in merge_lists when streaming Mistral responses with inline citations (#35259)

2. pretty_repr(html=True) does not return HTML for non-string content (#34875)

Changed files

PR #35290: fix(core): fix TypeError in merge_lists when merged list contains non-dict elements

Description (problem / solution / changelog)

Description

Changes

Testing

Changed files

PR #35349: fix(core): handle non-dict elements in merge_lists index lookup

Description (problem / solution / changelog)

Changed files

PR #35481: fix(core): guard merge_lists against non-dict elements when matching by index

Description (problem / solution / changelog)

Problem

Root Cause

Fix

Tests

Changed files

PR #35623: fix(core): prevent TypeError in merge_lists when list contains mixed strings and dicts

Description (problem / solution / changelog)

Bug

Root cause

Fix

Tests

Changed files

PR #35625: Fix TypeError in merge_lists when handling mixed type lists

Description (problem / solution / changelog)

Changed files

PR #35655: core: fix merge_lists crash when merged list contains non-dict elements

Description (problem / solution / changelog)

Description

Issue

Dependencies

Changed files

PR #35876: fix: add type check in merge_lists for non-dict elements

Description (problem / solution / changelog)

Summary

Reproduction

Test plan

Changed files

Code Example

Checked other resources

Package (Required)

Related Issues / PRs

Reproduction Steps / Example Code (Python)

Error Message and Stack Trace (if applicable)

Description

System Info

System Information

Package Information

Optional packages not installed

Other Dependencies

extent analysis

Problem Summary

Root Cause Analysis

Fix Plan

Step 1: Update merge_lists function to handle string indices

Step 2: Update the merge_lists function to handle nested lists

Step 3: Update the merge_dicts function to handle string keys

1. TypeError in `merge_lists` when streaming Mistral responses with inline citations (#35259)

2. `pretty_repr(html=True)` does not return HTML for non-string content (#34875)

Step 1: Update `merge_lists` function to handle string indices

Step 2: Update the `merge_lists` function to handle nested lists

Step 3: Update the `merge_dicts` function to handle string keys