Skip to main content

​
Function Signature

def page_index(
    doc,
    model=None,
    toc_check_page_num=None,
    max_page_num_each_node=None,
    max_token_num_each_node=None,
    if_add_node_id=None,
    if_add_node_summary=None,
    if_add_doc_description=None,
    if_add_node_text=None
)
Source: pageindex/page_index.py:1103

​
Description

The page_index() function is the main entry point for generating a PageIndex tree structure from PDF documents. It automatically loads default configuration from config.yaml and merges user-provided parameters, then delegates to page_index_main() for processing.

​
Parameters

​
doc
str or BytesIO
required
Path to PDF file or BytesIO object containing the PDF data. Must be a valid PDF file path (ending in .pdf) or a BytesIO object.
​
model
str
default:"gpt-4o-2024-11-20"
OpenAI model to use for processing. Examples: "gpt-4o-2024-11-20", "gpt-4o", "gpt-4.1"
​
toc_check_page_num
int
default:"20"
Number of pages to check for table of contents detection. The function will scan this many pages from the beginning to find TOC pages.
​
max_page_num_each_node
int
default:"10"
Maximum number of pages allowed in each node. Nodes exceeding this limit will be recursively subdivided.
​
max_token_num_each_node
int
default:"20000"
Maximum token count per node. Used in conjunction with max_page_num_each_node to determine when to subdivide large nodes.
​
if_add_node_id
str
default:"yes"
Whether to add sequential node IDs to the tree structure. Valid values: "yes" or "no"
​
if_add_node_summary
str
default:"yes"
Whether to generate AI summaries for each node. Valid values: "yes" or "no"
​
if_add_doc_description
str
default:"no"
Whether to generate a one-sentence description for the entire document. Valid values: "yes" or "no"
​
if_add_node_text
str
default:"no"
Whether to include full text content in each node. Valid values: "yes" or "no"

​
Return Value

​
result
dict
A dictionary containing the document structure:

​
Example Usage

​
Basic Usage with Defaults

from pageindex import page_index

# Process PDF with default settings
result = page_index("document.pdf")

print(result['doc_name'])
print(f"Found {len(result['structure'])} top-level sections")

​
Custom Configuration

from pageindex import page_index

# Process with custom parameters
result = page_index(
    doc="financial_report.pdf",
    model="gpt-4o-2024-11-20",
    toc_check_page_num=30,
    max_page_num_each_node=15,
    max_token_num_each_node=25000,
    if_add_node_id="yes",
    if_add_node_summary="yes",
    if_add_doc_description="yes",
    if_add_node_text="no"
)

print(f"Document: {result['doc_name']}")
print(f"Description: {result['doc_description']}")

​
Processing BytesIO Objects

from io import BytesIO
from pageindex import page_index

# Read PDF from bytes
with open("document.pdf", "rb") as f:
    pdf_bytes = BytesIO(f.read())

result = page_index(pdf_bytes)

​
With Full Text Content

from pageindex import page_index

# Include full text in each node
result = page_index(
    doc="manual.pdf",
    if_add_node_text="yes"
)

# Access text from nodes
for node in result['structure']:
    print(f"Section: {node['title']}")
    print(f"Pages: {node['start_index']}-{node['end_index']}")
    print(f"Text preview: {node['text'][:200]}...")

​
Notes

  • All parameters are optional; defaults are loaded from config.yaml
  • Only parameters that differ from defaults need to be specified
  • The function uses ConfigLoader internally to merge user options with defaults
  • Processing time varies based on document length and enabled features
  • API costs depend on document size and whether summaries are generated

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us