[feature] Add StructuredDataReader support for xlsx, csv, json and js… (

#14369)

llama-index-integrations/readers/llama-index-readers-structured-data/.gitignore

@@ -0,0 +1,153 @@
+llama_index/_static
+.DS_Store
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+bin/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+etc/
+include/
+lib/
+lib64/
+parts/
+sdist/
+share/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+.ruff_cache
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+notebooks/
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+pyvenv.cfg
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Jetbrains
+.idea
+modules/
+*.swp
+
+# VsCode
+.vscode
+
+# pipenv
+Pipfile
+Pipfile.lock
+
+# pyright
+pyrightconfig.json

llama-index-integrations/readers/llama-index-readers-structured-data/BUILD

@@ -0,0 +1,3 @@
+poetry_requirements(
+    name="poetry",
+)

llama-index-integrations/readers/llama-index-readers-structured-data/Makefile

@@ -0,0 +1,17 @@
+GIT_ROOT ?= $(shell git rev-parse --show-toplevel)
+
+help:	## Show all Makefile targets.
+	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[33m%-30s\033[0m %s\n", $$1, $$2}'
+
+format:	## Run code autoformatters (black).
+	pre-commit install
+	git ls-files | xargs pre-commit run black --files
+
+lint:	## Run linters: pre-commit (black, ruff, codespell) and mypy
+	pre-commit install && git ls-files | xargs pre-commit run --show-diff-on-failure --files
+
+test:	## Run tests via pytest.
+	pytest tests
+
+watch-docs:	## Build and watch documentation.
+	sphinx-autobuild docs/ docs/_build/html --open-browser --watch $(GIT_ROOT)/llama_index/

llama-index-integrations/readers/llama-index-readers-structured-data/README.md

@@ -0,0 +1,46 @@
+# LlamaIndex Readers Integration: Structured-Data
+
+The function 'StructuredDataReader' supports reading files in JSON, JSONL, CSV, and XLSX formats. It provides parameters 'col_index' and 'col_metadata' to differentiate between columns that should be written into the document's main text and additional metadata.
+
+## Install package
+
+```bash
+pip install llama-index-readers-structured-data
+```
+
+Or install locally:
+
+```bash
+pip install -e llama-index-integrations/readers/llama-index-readers-structured-data
+```
+
+## Usage
+
+1. for single document:
+
+```python
+from pathlib import Path
+from llama_index.readers.structured_data.base import StructuredDataReader
+
+parser = StructuredDataReader(col_index=["col1", "col2"], col_metadata=0)
+documents = parser.load_data(Path("your/file/path.json"))
+```
+
+2. for dictory of documents:
+
+```python
+from pathlib import Path
+from llama_index.core import SimpleDirectoryReader
+from llama_index.readers.structured_data.base import StructuredDataReader
+
+parser = StructuredDataReader(col_index=[1, -1], col_metadata="col3")
+file_extractor = {
+    ".xlsx": parser,
+    ".csv": parser,
+    ".json": parser,
+    ".jsonl": parser,
+}
+documents = SimpleDirectoryReader(
+    "your/dic/path", file_extractor=file_extractor
+).load_data()
+```

llama-index-integrations/readers/llama-index-readers-structured-data/llama_index/readers/structured_data/BUILD

@@ -0,0 +1 @@
+python_sources()

llama-index-integrations/readers/llama-index-readers-structured-data/llama_index/readers/structured_data/__init__.py

@@ -0,0 +1,4 @@
+from llama_index.readers.structured_data.base import StructuredDataReader
+
+
+__all__ = ["StructuredDataReader"]

llama-index-integrations/readers/llama-index-readers-structured-data/llama_index/readers/structured_data/base.py

@@ -0,0 +1,138 @@
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Union
+
+import pandas as pd
+from fsspec import AbstractFileSystem
+from llama_index.core.readers.base import BaseReader
+from llama_index.core.schema import Document
+
+
+class StructuredDataReader(BaseReader):
+    """
+    Updated BaseReader parser to support JSON, JSONL, CSV and Excel (.xlsx) files.
+
+    ...
+
+    Args:
+        col_joiner (str): The string to join the columns with. Defaults to ', '.
+        col_index (str, int, or list): The list of columns to be used as index.
+        col_metadata (None, str, int, or list): The list of columns to be used as metadata.
+        ...
+    """
+
+    def __init__(
+        self,
+        *args: Any,
+        col_joiner: str = ", ",
+        pandas_config: dict = {},
+        col_index: Union[str, int, List],
+        col_metadata: Optional[Union[str, int, List]] = None,
+        **kwargs: Any,
+    ) -> None:
+        """Init params."""
+        super().__init__(*args, **kwargs)
+        self._col_joiner = col_joiner
+        self._pandas_config = pandas_config
+        self._col_index = col_index
+        self._col_metadata = col_metadata
+
+    def _load_dataframe(
+        self, file: Path, fs: Optional[AbstractFileSystem] = None
+    ) -> pd.DataFrame:
+        file_extension = file.suffix.lower()
+
+        read_funcs = {
+            ".csv": lambda f: pd.read_csv(f),
+            ".xlsx": lambda f: pd.read_excel(f),
+            ".json": lambda f: pd.read_json(f, encoding="utf-8"),
+            ".jsonl": lambda f: pd.read_json(f, encoding="utf-8", lines=True),
+        }
+
+        if file_extension not in read_funcs:
+            raise ValueError(
+                f"Unsupported file extension '{file_extension}'. Supported extensions are 'json', 'csv', 'xlsx', and 'jsonl'."
+            )
+
+        if fs:
+            with fs.open(file) as f:
+                df = read_funcs[file_extension](f, **self._pandas_config)
+        else:
+            df = read_funcs[file_extension](file, **self._pandas_config)
+        return df
+
+    def _validate_column(self, index_name, column_index, df):
+        if isinstance(column_index, int):
+            assert (
+                -len(df.columns) < column_index < len(df.columns)
+            ), f"The {index_name} {column_index} exceeds the range of columns in the dataframe: ({len(df.columns)})"
+        elif isinstance(column_index, str):
+            assert (
+                column_index in df.columns
+            ), f"The {index_name} must be in the dataframe"
+        else:
+            if all(isinstance(item, int) for item in column_index):
+                assert all(
+                    -len(df.columns) < item < len(df.columns) for item in column_index
+                ), f"Some items in {index_name} exceed the range of columns in the dataframe: ({len(df.columns)})"
+            elif all(isinstance(item, str) for item in column_index):
+                assert set(column_index).issubset(
+                    df.columns
+                ), f"All columns in {index_name} must be in the dataframe"
+            else:
+                raise ValueError(
+                    "Not support int and str columns both in column configs."
+                )
+
+    def load_data(
+        self,
+        file: Path,
+        extra_info: Optional[Dict] = None,
+        fs: Optional[AbstractFileSystem] = None,
+    ) -> List[Document]:
+        """Parse file."""
+        df = self._load_dataframe(file, fs)
+
+        assert self._col_index, f"The col_index must be specified"
+        self._validate_column("col_index", self._col_index, df)
+
+        if isinstance(self._col_index, int) or (
+            isinstance(self._col_index, list)
+            and all(isinstance(item, int) for item in self._col_index)
+        ):
+            df_text = df.iloc[:, self._col_index]
+        else:
+            df_text = df[self._col_index]
+
+        if isinstance(df_text, pd.DataFrame):
+            text_list = df_text.apply(
+                lambda row: self._col_joiner.join(row.astype(str).tolist()), axis=1
+            ).tolist()
+        elif isinstance(df_text, pd.Series):
+            text_list = df_text.tolist()
+
+        if not self._col_metadata:
+            return [
+                Document(text=text_tuple, metadata=(extra_info or {}))
+                for text_tuple in text_list
+            ]
+        else:
+            self._validate_column("col_metadata", self._col_metadata, df)
+            if isinstance(self._col_metadata, int) or (
+                isinstance(self._col_metadata, list)
+                and all(isinstance(item, int) for item in self._col_metadata)
+            ):
+                df_metadata = df.iloc[:, self._col_metadata]
+            else:
+                df_metadata = df[self._col_metadata]
+
+            if isinstance(df_metadata, pd.Series):
+                df_metadata = pd.DataFrame(df_metadata)
+
+            metadata_list = df_metadata.to_dict(orient="records")
+
+            return [
+                Document(
+                    text=text_tuple, metadata={**(metadata_tuple), **(extra_info or {})}
+                )
+                for text_tuple, metadata_tuple in zip(text_list, metadata_list)
+            ]