Indexly Data Analysis & File Pipeline Overview

Who This Page Is For

• Users deciding between analyze-file, analyze-json, analyze-db, and analyze-autodoctor
• Developers tracing how Indexly routes structured files through the loader and orchestrator
• Operators analyzing AutoDoctor report JSON, telemetry JSON, or SQLite output with Indexly

Supported Formats

Indexly provides analysis and summarization for these structured formats:

CSV

• Delimiter detection
• Summary statistics, optional cleaning, visualization, and persistence via analyze-csv
• CSV routing through analyze-file when you want one command for mixed structured files
• Statistical inference through the Inference Docs

JSON and NDJSON

• Standard list and dictionary JSON
• NDJSON / record-list JSON
• .json files that contain NDJSON records
• Compressed *.json.gz files
• Socrata-style columns and data JSON
• Indexly search cache JSON
• AutoDoctor report JSON
• AutoDoctor telemetry JSON

SQLite

• Generic SQLite profiling
• Specialized AutoDoctor DB summaries when the schema matches AutoDoctor tables

Excel, Parquet, XML, YAML

• Sheet-aware Excel loading
• Efficient Parquet previews
• XML structure analysis and tree rendering
• Safe YAML loading into JSON-like structures
• YAML persistence through analyze-file writes to cleaned_data and stores YAML-specific metadata/artifact references when persistence is enabled

Choose The Right Command

Command Behaviors

indexly analyze-csv <file>

This is the dedicated CSV route.

It is best for:

• delimiter detection and numeric summary statistics
• optional --auto-clean, --normalize, and --remove-outliers
• terminal, static, or interactive CSV visualizations
• CSV analysis exports in txt, md, or json

For parser-accurate CSV options, see Analyze CSV Data and Clean CSV Data.

indexly analyze-file <file>

This is the universal dispatcher.

It:

• detects file type through universal_loader
• adds metadata hints for special formats such as AutoDoctor
• routes into the correct pipeline through the orchestrator

Use this when you want one command for mixed datasets.

For SQLite files, this route is intentionally a quick preview path. It loads bounded table previews for generic database inspection. Use analyze-db when you need relationship discovery, table profiling controls, diagrams, or exportable database summaries.

For YAML and YML files, persistence is handled by the same orchestrator path used by other structured formats. By default, analyze-file writes the cleaned preview and summary to ~/.indexly/indexly.db (cleaned_data), includes a JSON-safe yaml_table_output block in metadata, and records an auxiliary analysis artifact path with schema indexly.yaml.analysis.v1. Use --no-persist to skip both the analysis database write and YAML artifact creation for that run.

indexly analyze-json <file>

This is the JSON-focused route.

It is best for:

• plain JSON
• NDJSON
• compressed JSON
• Socrata-style table JSON
• JSON files that may need structural fallback logic

It now shares more routing behavior with the orchestrator, which helps prevent the old failure mode where NDJSON-style .json files summarized correctly but could not persist cleanly.

Use --chunk-size on this command when a newline-delimited source is too large to materialize fully.

See Analyze JSON And NDJSON Files.

indexly analyze-db <db>

This is the database-focused route.

It is best for:

• unknown SQLite databases
• table-by-table profiling
• relationship discovery
• schema exports and diagrams

By default, large tables are profiled with bounded sampling. Use --sample-size to choose a profile size, --fast or --fast-mode for lighter metrics, and --all-data only when full-table profiling is required.

When the database matches AutoDoctor’s schema, Indexly switches to an operational summary instead of staying in the generic inspection path.

indexly analyze-autodoctor <path>

This is the dedicated operational route for AutoDoctor artifacts.

It supports:

• AutoDoctor_Report.json
• Telemetry_*.json
• autodoctor.db

Use it when you want human-readable summaries first, not raw structure exploration.

See Analyze AutoDoctor Artifacts.

How Routing Works

Indexly’s structured-data analysis has three layers:

analyze-file / analyze-json / analyze-db / analyze-autodoctor
        |
        v
Universal Loader
        |
        v
Analysis Orchestrator
        |
        v
Specialized Pipelines

Universal Loader Responsibilities

• detect file type from extension and content
• distinguish JSON, NDJSON, SQLite, Excel, XML, YAML, and Parquet
• attach metadata hints such as AutoDoctor schema fingerprints

Analysis Orchestrator Responsibilities

• decide which analysis pipeline should run
• preserve JSON-aware persistence behavior
• reroute special formats such as AutoDoctor into dedicated summaries

Pipeline Responsibilities

Each specialized pipeline handles:

• validation
• normalization
• preview generation
• summary generation
• persistence/export handoff

AutoDoctor-Aware Analysis

Indexly now recognizes two AutoDoctor JSON families plus the AutoDoctor SQLite schema:

This avoids flattening operational documents into one synthetic table when a domain-specific summary is more useful.

For operational examples and artifact selection guidance, see:

• Analyze AutoDoctor Artifacts
• Telemetry and Persistence
• Generate and Share Support Bundle

Practical Examples

Preparing exported files before analysis

indexly rename-file ./exports --pattern "{date}-{title}" --recursive --dry-run
indexly rename-file ./exports --pattern "{date}-{title}" --recursive

Use Rename File when exported CSVs, reports, or logs need stable names before analysis or organization.

CSV analysis and cleaning

indexly analyze-csv sales.csv --show-summary
indexly analyze-csv sales.csv --auto-clean --show-summary --no-persist
indexly analyze-csv sales.csv --show-chart ascii --chart-type hist --transform auto

Generic structured-file analysis

indexly analyze-file sales.csv --auto-clean --show-summary
indexly analyze-file data.json --show-summary
indexly analyze-file metrics.parquet --show-summary
indexly analyze-file workbook.xlsx --sheet-name Sheet1 --show-summary

JSON and NDJSON analysis

indexly analyze-json iris.json --show-summary
indexly analyze-json events.ndjson --chunk-size 10000 --show-summary
indexly analyze-json records.json.gz --show-summary

SQLite analysis

indexly analyze-db chinook.db --show-summary --all-tables
indexly analyze-file chinook.db --show-summary

AutoDoctor analysis

indexly analyze-autodoctor .\AutoDoctor_Report.json --show-summary
indexly analyze-autodoctor .\Telemetry_20260416-081258-BTNB05.json --summary-only
indexly analyze-autodoctor .\autodoctor.db --show-summary

Related Pages

• Usage Guide
• Rename File
• Analyze JSON And NDJSON Files
• Clean CSV Data
• Analyze SQLite Databases
• Analyze AutoDoctor Artifacts
• Developer Guide