# disk_treemap
Just another disk usage analyzer with treemap GUI.
## Pros
* Written in Python. Easy to run, modify and extend. Cross platform support. (Tested on Linux, Windows, and Android with
Termux.)
* B/S structure. You can run the scanner on a remote machine, and view treemap via a browser on a local machine
* Support save/load a treemap. You can run the scanner on a machine, then copy result to another one and view it.
* Support S3 or S3 compatible object storage service
## Cons
* The web base frontend may suffer from performance issue if the treemap is too large.
* Scan speed is not ideal, especially on Windows.
# Installation
## Install via pip package installer
```shell
pip install disk_treemap
```
## Build and install from source
```shell
# clone the repository
git clone
git submodules update --recursively
# build static
python3 setup.py build_static
#build wheel
python3 setup.py bdist_wheel
#install built wheel
pip install dist/disk_treemap-1.0.0-py3-none-any.whl # change the filename
```
# Usage
```
usage: disk-treemap [-h] [--size-tree-path SIZE_TREE_PATH] [--overwrite]
[--scan-only] [--host HOST] [--port PORT] [--compression]
[--endpoint-url ENDPOINT_URL] [--follow-links]
[--follow-mounts] [--everything]
[paths ...]
positional arguments:
paths path(s) to scan. If multiple paths is provided, they
will be show in root side by side. S3 or compatible
object storage service is supported by a "s3://"
prefixed URI
optional arguments:
-h, --help show this help message and exit
--size-tree-path SIZE_TREE_PATH, --size_tree_path SIZE_TREE_PATH, -f SIZE_TREE_PATH
path to save scan result as a JSON file
--overwrite, -o overwrite existed JSON file. default to False
--scan-only, --scan_only, -s
scan and save JSON file but do not start web server.
default to False
--host HOST, -H HOST listening host of the web server
--port PORT, -p PORT listening port of the web server. default to 8000
--compression, -c enable compression of web server. require
flask_compress to operate. default to False
--endpoint-url ENDPOINT_URL
custom endpoint url, only affects S3
--follow-links, --follow_links
follow symlinks
--follow-mounts, --follow_mounts
follow mounts
--everything use Everything by voidtools to speedup scanning. The
result will be absolute path. Everything must be
running and only x64 version is supported.
```
You may also use the module directly: `python -m disk_treemap.main`. Same arguments apply.
A `size_tree.json` will be generated in the current directory. It contains file tree and file size information. Keep it
safe!
# Typical Usage
**Analyze an ordinary computer**
1. Run `disk-treemap <Paths to analyze>`
1. After `listening 127.0.0.1:8000` appearing, open browser and navigate to http://127.0.0.1:8000 .
**Analyze a Windows computer with Everything x64 installed**
1. Make sure Everything x64 is running.
1. Run `disk-treemap --everything <Paths to analyze>`
1. After `listening 127.0.0.1:8000` appearing, open browser and navigate to http://127.0.0.1:8000 .
**Analyze a remote Linux server, view on the local machine**
1. Run `disk-treemap <Paths to analyze> --host 0.0.0.0`.
If bandwidth between the server and the local machine is limited, try to install optional dependencies and append `--compression` to command above to enable compression.
1. After `listening 0.0.0.0:8000` appearing, open browser on the local machine and navigate to http:
//<IP address of the server>:8000 .
**Analyze a remote Linux server without external accessible IP, view on the local machine**
1. Run `disk-treemap <Paths to analyze> --scan_only`
1. After process exit without error. There should be a file named `size_tree.json` in the current directory. Copy the
file to local machine using `rsync` or other tools.
1. Run `disk-treemap` in the directory where the copied file located.
1. After `listening 127.0.0.1:8000` appearing, open browser and navigate to http://127.0.0.1:8000 .
**Install and analyze an Android phone with Termux**
1. Install Termux https://termux.com/
1. Install Python and pip in Termux https://wiki.termux.com/wiki/Python
1. Install disk_treemap via pip: `pip install termux`
1. Just do analyze as your phone is an ordinary computer.
# TODO
* Provide more visualization. Icicle/flame, sunburst maybe.
* Use NTFS USN Journal to speed up scanning on Windows.

挣扎的蓝藻
- 粉丝: 14w+
- 资源: 15万+
最新资源
- 机器学习-bert模型
- 【毕业设计】基于Python的Django-html基于语音识别的智能垃圾分类系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- ERA5再分析数据根区土壤水分下载_era5土壤湿度-CSDN博客.html
- 【毕业设计】基于Python的Django-html基于小波变换的数字水印研究系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html基于知识图谱的百科知识问答平台源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html开放领域事件抽取系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html旅游城市关键词分析系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html基于知识图谱的医疗问答系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html某大学学生影响力分析系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- Unity RPG地图场景资源
- 【毕业设计】基于Python的Django-html棉花数据平台建设与可视化系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- hymenoptera-data 数据集
- Java I/O流及其应用详解:字节流、字符流及相关工具类全面解析
- 机器学习-apriori算法-超市关联分析
- 【毕业设计】基于Python的Django-html企业物流管理系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
- 【毕业设计】基于Python的Django-html某医院体检挂号系统源码(完整前后端+mysql+说明文档+LW+PPT).zip
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈


