This plugin provides the following features:
- Fast and rich full text search to Redmine
- Display similar issues on an issue page
If you're using MySQL or MariaDB, you need Mroonga 9.03 or later.
If you're using PostgreSQL, you need PGroonga 2.2.0 or later.
Mroonga and PGroonga uses Groonga as full text search engine. You need Groonga 9.0.3 or later.
- ChupaText server: Text extractor server.
See Mroonga document
You can choose one of them to install ChupaText:
chupa-text-docker is recommended. See chupa-text-docker document to install chupa-text-docker.
See chupa-text-vagrant document to install chupa-text-vagrant.
chupa-text-http-server is a normal Ruby on Rails application like Redmine itself. You can deploy chupa-text-http-server as a normal Ruby on Rails application.
$ cd redmine
$ git clone https://github.com/clear-code/redmine_full_text_search.git plugins/full_text_search
$ bundle install
$ RAILS_ENV=production bin/rails redmine:plugins:migrateRestart Redmine.
NOTE for PGroonga:
If you use normal user for Redmine. You must run the following query
as a super user before run RAILS_ENV=production bin/rails redmine:plugins:migrate:
CREATE EXTENSION IF NOT EXISTS pgroonga;Open https://YOUR_REDMINE_SERVER/settings/plugin/full_text_search and
configure items in the page. If you install ChupaText server, you must
configure "ChupaText server URL". If you install your ChupaText server
by chupa-text-docker or chupa-text-vagrant on the same host, it's
http://127.0.0.1:20080/extraction.json.
You need to create index for existing data. You need to run
full_text_search:synchronize task until no more synchronize target
data.
$ cd redmine
$ RAILS_ENV=production bin/rails full_text_search:synchronize
$ RAILS_ENV=production bin/rails full_text_search:synchronize
$ RAILS_ENV=production bin/rails full_text_search:synchronize
...This plugin supports query expansion. You can use this feature to implement synonym search.
You can administrate query expansion list by Web UI in administration page or data file.
You can use the following format for data file:
- CSV
- JSON
If you use CSV, use the following format:
SOURCE1,DESTINATION1
SOURCE2,DESTINATION2
...
Example:
MySQL,MySQL
MySQL,MariaDB
MariaDB,MySQL
MariaDB,MariaDB
If you use JSON, use one of the following formats:
[
["SOURCE1", "DESTINATION1"],
["SOURCE2", "DESTINATION2"],
...
][
{"source": "SOURCE1", "destination": "DESTINATION1"},
{"source": "SOURCE2", "destination": "DESTINATION2"},
...
]Examples:
[
["MySQL", "MySQL"],
["MySQL", "MariaDB"],
["MariaDB", "MySQL"],
["MariaDB", "MariaDB"]
][
{"source": "MySQL", "destination": "MySQL"},
{"source": "MySQL", "destination": "MariaDB"},
{"source": "MariaDB", "destination": "MySQL"},
{"source": "MariaDB", "destination": "MariaDB"}
]You can synchronize query expansion list with the data file by the following command:
$ cd redmine
$ RAILS_ENV=production bin/rails full_text_search:query_expansion:synchronize INPUT=query-expansion.csvYou can confirm the current query expansion list in administration page.
Mroonga isn't crash safe. If MySQL is crashed while updating data in Mroonga, Mroonga data may be broken.
Here is the instruction to recover from broken Mroonga data.
If you're using Redmine plugin Delayed Job, you need to stop workers and delete jobs for this plugin:
$ sudo -H systemctl stop [email protected]
$ cd redmine
$ RAILS_ENV=production bin/rails runner 'Delayed::Job.where(queue: "full_text_search").delete_all'Stop MySQL:
$ sudo -H systemctl stop mysqldRemove Mroonga related files:
$ cd redmine
$ database_name=$(RAILS_ENV=production bin/rails runner 'puts ActiveRecord::Base.configurations[Rails.env]["database"]')
$ sudo -H sh -c "rm -rf /var/lib/mysql/${database_name}.mrn*"Start MySQL:
$ sudo -H systemctl start mysqldCheck that Mroonga has been properly installed based on the Mroonga manual. If Mroonga isn't installed, install Mroonga like the following:
$ mysql -u root -p < /usr/share/mroonga/install.sqlDestruct tables explicitly for this plugin:
$ mysql -u root -p ${database_name}
> DROP TABLE IF EXISTS fts_query_expansions;
> DROP TABLE IF EXISTS fts_targets;
> DROP TABLE IF EXISTS fts_tags;
> DROP TABLE IF EXISTS fts_tag_types;
> DROP TABLE IF EXISTS fts_types;Recreate schema for this plugin:
$ cd redmine
$ RAILS_ENV=production bin/rails redmine:plugins:migrate NAME=full_text_search VERSION=0
$ RAILS_ENV=production bin/rails redmine:plugins:migrate NAME=full_text_searchIf you're using Redmine plugin Delayed Job, you need to start workers:
$ sudo -H systemctl start [email protected]Synchronize:
$ cd redmine
$ RAILS_ENV=production bin/rails full_text_search:synchronize UPSERT=laterHere are some useful tools to prepare:
dev/run-mysql.shanddev/run-postgresql.sh: Run new RDBMS instance by Docker.dev/initialize-redmine.sh: Initialize Redmine.dev/run-test.sh: Run tests for the full text search plugin.
Clone source codes. This is required only once.
$ git clone https://github.com/redmine/redmine.git
$ cd redmine
$ git checkout 5.0-stable # or something
$ git clone [email protected]:${YOUR_FORK}/redmine_full_text_search.git plugins/full_text_searchYou can add more plugins to plugins/.
Choose suitable database configuration:
$ ln -fs ../plugins/full_text_search/config/database.yml.example.${REDMINE_VERSION}.${RDBMS} config/database.ymlHere is an example to use Redmine 5.0 and MySQL:
$ ln -fs ../plugins/full_text_search/config/database.yml.example.5.0.mysql config/database.ymlRun RDBMS.
For MySQL:
$ plugins/full_text_search/dev/run-mysql.sh /tmp/mysqlFor PostgreSQL:
$ plugins/full_text_search/dev/run-postgresql.sh /tmp/postgresqlInitialize Redmine:
$ plugins/full_text_search/dev/initialize-redmine.shRun tests:
$ plugins/full_text_search/dev/run-test.shYou need to create mapper classes for each search target. See
lib/full_text_search/*_mapper.rb for details.
You need to add require_dependency "full_text_search/XXX_mapper to
init.rb to load these new mapper classes.
You can confirm your changes by usual Redmine development ways.
For example, here is a command line to run Redmine:
$ bin/rails serverYou need to add tests to the following files:
test/unit/full_text_search/XXX_test.rbtest/functional/full_text_search/search_controller_test.rb
Here is a command line to run tests:
$ plugins/full_text_search/dev/run-test.shYou can specify test options by TESTOPTS:
$ plugins/full_text_search/dev/run-test.sh TESTOPTS="-n/test_XXX/"You can see all test options by TESTOPTS=--help:
$ plugins/full_text_search/dev/run-test.sh TESTOPTS=--help-
Kenji Okimoto
-
Sutou Kouhei
<[email protected]> -
Shimadzu Corporation
The MIT License. See LICENSE for details.
asserts/stylesheets/fontawesome*/**/*- Author: @fontawesome
- Fonts: SIL OFL 1.1 License
- Codes: MIT License
- See
asserts/stylesheets/fontawesome*/LICENSE.txtfor details
- Fork it ( http://github.com/clear-code/redmine_full_text_search/fork )
- Create your feature branch (
git checkout -b my-new-feature) - Commit your changes (
git commit -am 'Add some feature') - Push to the branch (
git push origin my-new-feature) - Create new Pull Request
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot]")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent