go-nate

CLI tool to dump, index and search bookmarks

Why 'go-nate' name?

Considering the bookmarks as a treasure the idea was to name the tool after the not-so-famous treasure hunter. Therefore, the nate part is coming after the Nathan Drake the treasure hunter of the Uncharted video game series. go part, obviously, stands for the main language the CLI tool is built on.

Motivation

The idea behind this tool is to make browser's bookmarks searchable by their category, title, and content. It could be done in 3 steps: dump, index and then run a webserver with available UI to search across the bookmarks.

Installation

Currently, the CLI tool could be installed using go install or docker capabilities.

Local

Clone the repository:

git clone https://github.com/Neurostep/go-nate.git && cd go-nate/

Run the following command:

go install github.com/Neurostep/go-nate

go-nate binary will be installed in $GOPATH/bin

Docker

To pull docker image run the following:

docker pull neurostep/go-nate:latest

go-nate while running stores logs and bookmark data (content, index information etc) locally. In order to make it persistent and accessible from the host we need to mount local directories to docker container, for example:

docker run -it -v "${HOME}/.gonate:/root/.gonate" -v "${HOME}/Library/Application Support/Google/Chrome:/root/bookmarks" neurostep/go-nate:latest go-nate dump -f bookmarks
docker run -it -v "${HOME}/.gonate:/root/.gonate" neurostep/go-nate:latest go-nate index
docker run -it -v "${HOME}/.gonate:/root/.gonate" neurostep/go-nate:latest go-nate repl
docker run -it -v "${HOME}/.gonate:/root/.gonate" neurostep/go-nate:latest go-nate server

Usage

By default, go-nate stores all auxiliary data in "${HOME}/.gonate/" directory. This can be changed by specifying environment variable GONATE_HOME.

go-nate --help

Usage: go-nate [flags] <command> [<args>]

Commands:
    dump      Saves bookmarks for the specified browser to the local DB. If bookmark URL is provided, it will dump that one only
    index     Indexes bookmarks from DB. If 'bookmark url' is provided, it will index only that bookmark
    watch     Runs a background check for the bookmark file change
    server    Runs HTTP server on provided port
    repl      Starts the go-nate REPL

Flags:
  --d  Turn on debug mode
  --i  Path to directory containing search index
  --l  Path to directory containing logs
  --s  Path to directory containing database files

Repl

go-nate repl --help

USAGE
  go-nate repl

go-nate repl will start the very simple REPL with prompt go-nate> . REPL supports following commands:

1. search <here goes search string> - will run the Query Search against the index. Index should be present.
2. set search searchResultSize <number> - specifies the number of result for search command. Default number is 10

Examples:

go-nate repl
go-nate> search golang
go-nate> set search searchResultSize 2

Dump

go-nate dump --help

USAGE
  go-nate dump [-f path] [-b browser] [-p profile] [-c concurrency] [-F force to dump] [bookmark url] [bookmark folder] [bookmark title]

FLAGS
  -F false                                                       If provided, then bookmark will be dumped even if it already exists
  -b chrome                                                      Browser for which bookmarks are being dumped
  -c 100                                                         Number of concurrent workers to dump the bookmarks
  -f ${HOME}/Library/Application Support/Google/Chrome           The path to local browser profile
  -p default                                                     The profile name of the browser

go-nate dump command does the following:

1. reads provided browser's bookmark file
2. tries to retrieve data for the bookmark using Go http library first
3. if it failed to retrieve content of the bookmark using Go http library, it will try to do that using Chrome web browser
4. then it stores content locally using Badger DB

Index

go-nate index --help

USAGE
  go-nate index [bookmark url]

For the index command it's required that dump step previously done. go-nate index will go over dumped data and will index that data using Bleve search engine

Server

go-nate server --help

USAGE
  go-nate server [-p port]

FLAGS
  -p 8080  Number represents the port server will listen to

Will spin-up the server. Navigate to http://localhost:8080/search/syntax/ and try search your bookmarks!

Watch

go-nate watch --help

USAGE
  go-nate watch [-i interval] [-f path] [-b browser] [-p profile]

FLAGS
  -b chrome                                                      Browser for which bookmarks are being watched and dumped
  -f ${HOME}/Library/Application Support/Google/Chrome           The path to local browser profile
  -i 30s                                                         The interval in which watch will perform the bookmark file check
  -p default                                                     The profile name of the browser

This command runs a background job which will be checking the provided bookmarks file for the update and run dump and index automatically.

Requirements

To run go-nate locally there are following requirements:

• Go Lang version 1.16+
• NodeJS version 16.3+
• Chrome Browser (version 91 tested)

There is also possibility to run go-nate using docker.