Skip to main content

elasticsearch catalog support for guillotina

Project description

.. contents::

GUILLOTINA_ELASTICSEARCH
========================

.. image:: https://travis-ci.org/guillotinaweb/guillotina_elasticsearch.svg?branch=master
:target: https://travis-ci.org/guillotinaweb/guillotina_elasticsearch

Elasticsearch integration for guillotina.


Configuration
-------------

config.json can include elasticsearch section::

"elasticsearch": {
"index_name_prefix": "guillotina-",
"connection_settings": {
"endpoints": ["localhost:9200"],
"sniffer_timeout": 0.5
}
}


Installation on a site
----------------------

POST SITE_URL/@catalog

{}

Uninstall on a site
-------------------

DELETE SITE_URL/@catalog

{}


Migrate Command
---------------

`guillotina_elasticsearch` comes with a `es-migrate` guillotina command::

./bin/g es-migrate


This command will perform a live migration on the index. It does this by
performing the reindex on a new index while the other one is still active.

New index and delete requests are performed on both indexes during live migration.

It is also smart about how to migrate, doing a diff on the mapping and only
reindexing the fields that changed.


Testing
--------

If container es (elasticsearch) fails to start when running tests,
you should increase max_map_count. command::

# Linux
sudo sysctl -w vm.max_map_count=262144


1.3.20 (2018-05-18)
-------------------

- Add scroll support to query
[gitcarbs]


1.3.19 (2018-05-14)
-------------------

- Raise an exception instead of returning an ErrorResponse object
[vangheem]


1.3.18 (2018-05-10)
-------------------

- Added context to the IIndexProgress event
[jordic]

1.3.17 (2018-05-09)
-------------------

- Do not set db write enabled on migrate
[vangheem]


1.3.16 (2018-05-09)
-------------------

- More logging


1.3.15
------

- Added new event on reindex start.
[jordic]


1.3.15 (2018-05-08)
-------------------

- Added IIndexProgress to hook on reindex process
[jordic]


1.3.13 (2018-05-02)
-------------------

- Format stored field results like source results
[vangheem]


1.3.12 (2018-05-01)
-------------------

- More vacuum improvements


1.3.11 (2018-04-30)
-------------------

- More vacuum improvements
[vangheem]

1.3.10 (2018-04-30)
-------------------

- migrate script should not use transactions
[vangheem]

1.3.9 (2018-04-30)
------------------

- Optimized vacuum script to use tid sorting which should prevent
needing to go through so many docs to do the vacuum check
[vangheem]


1.3.8 (2018-04-27)
------------------

- Provide `@name` in results
[vangheem]


1.3.7 (2018-04-26)
------------------

- change `@id` in search results to `@uid`
[vangheem]

- Add support for analysis character filter
[gitcarbs]


1.3.6 (2018-04-18)
------------------

- Work with store=true mappings
[vangheem]


1.3.5 (2018-04-15)
------------------

- Smaller bulk sizes for es vacuum
[vangheem]


1.3.4 (2018-04-15)
------------------

- Some performance fixes for vacuuming
[vangheem]


1.3.3 (2018-04-14)
------------------

- Provide profile data in results
[gitcarbs]


1.3.2 (2018-04-03)
------------------

- Upgrade tests to use pytest-docker-fixtures
[vangheem]


1.3.1 (2018-03-14)
------------------

- Pay attention to `index_data` configuration
[vangheem]


1.3.0 (2018-03-14)
------------------

- Upgrade to work with guillotina 2.4.x
[vangheem]


1.2.11 (2018-03-09)
-------------------

- Implement statement cache clearing
[vangheem]


1.2.10 (2018-03-02)
-------------------

- Do not use cached statement for vacuum
[vangheem]


1.2.9 (2018-02-07)
------------------

- Handle errors unpickling object for es vacuum
[vangheem]


1.2.8 (2018-01-11)
------------------

- Make sure to change transaction strategy for commands
[vangheem]


1.2.7 (2018-01-08)
------------------

- Make sure to close out connection in commands
[vangheem]


1.2.6 (2017-12-18)
------------------

- Use `traverse` instead of `do_traverse`.
[vangheem]


1.2.5 (2017-12-08)
------------------

- Retry on conflict for updates
[vangheem]


1.2.4 (2017-12-06)
------------------

- Use futures instead of threads for migrations
[vangheem]


1.2.3 (2017-11-21)
------------------

- Upgrade testing infrastructure with latest guillotina
[vangheem]


1.2.2 (2017-11-08)
------------------

- Fix test setup with jenkins
[vangheem]


1.2.1 (2017-11-08)
------------------

- Make logging less noisy
[vangheem]


1.2.0 (2017-11-06)
------------------

- Upgrade to guillotina 2.0.0
[vangheem]


1.1.24 (2017-10-12)
-------------------

- Close connection after done in vacuum command
[vangheem]


1.1.23 (2017-10-09)
-------------------

- Make sure to use `async with` syntax for manual api calls to elasticsearch using
session object.
[vangheem]


1.1.22 (2017-09-28)
-------------------

- Another tweak for missing indexes on running migration
[vangheem]


1.1.21 (2017-09-21)
-------------------

- All indexing/removing operations are already in a future so no need to add futures
to operations.
[vangheem]


1.1.20 (2017-09-21)
-------------------

- Use latest guillotina futures request api
[vangheem]


1.1.19 (2017-09-14)
-------------------

- Change page size and scroll of esvacuum to prevent timeouts
[vangheem]


1.1.18 (2017-08-22)
-------------------

- Fix TIDConflictError when registry is edited during a migration.
[vangheem]


1.1.17 (2017-08-11)
-------------------

- Handle missing types from migrations when mappings change better
[vangheem]


1.1.16 (2017-08-09)
-------------------

- Results from search should return sort value
[gitcarbs]


1.1.15 (2017-07-28)
-------------------

- Fix vacuum to handle empty scroll errors when iterating through all keys
[vangheem]


1.1.14 (2017-07-21)
-------------------

- Fix deadlock issue on unindex when migration is active
[vangheem]


1.1.13 (2017-07-12)
-------------------

- Wait a bit before running migration so addons that use async to calculate
dynamic types can load


1.1.12 (2017-07-12)
-------------------

- Fix scenario where doc type is missing in the upgraded mapping
[vangheem]


1.1.11 (2017-07-10)
-------------------

- Add update_by_query method
[vangheem]


1.1.10 (2017-07-06)
-------------------

- Add reindex command
[vangheem]


1.1.9 (2017-07-06)
------------------

- Fix invalid import in migrate script
[vangheem]


1.1.8 (2017-07-05)
------------------

- Add more logging for vacuum
[vangheem]


1.1.7 (2017-06-29)
------------------

- Add vacuum command
[vangheem]


1.1.6 (2017-06-23)
------------------

- Fix group query to not prepend 'group:' to security query for groups
[vangheem]


1.1.5 (2017-06-21)
------------------

- Fix migration when objects were deleted while migrating that were thought to
be orphaned
[vangheem]


1.1.4 (2017-06-20)
------------------

- Prevent auto mapping of metadata
[bloodbare]


1.1.3 (2017-06-15)
------------------

- Fix reindexing
[vangheem]


1.1.2 (2017-06-15)
------------------

- Fix import for client_exceptions aiohttp errors
[vangheem]


1.1.1 (2017-06-09)
------------------

- Move fixtures out of conftest.py into it's own module. This could break
tests that depend on it!
[vangheem]


1.1.0 (2017-06-09)
------------------

- Add Reindexer utility that specializes just in reindexing instead of
using the migrator
[vangheem]


1.0.29 (2017-06-08)
-------------------

- Clean mapping before it's compared to prevent false positives for detecting
differences in mappings
[vangheem]


1.0.28 (2017-06-08)
-------------------

- Be able to migrate the mapping only and skip working through content on the site
[vangheem]


1.0.27 (2017-06-07)
-------------------

- Make sure to flush reindexed items when calling reindex_all_content
[vangheem]


1.0.26 (2017-05-26)
-------------------

- When adding groups to query that is built, make sure to prepend each group with
"group:" so we can distinguish groups from users and somehow can not potentially
hack it where they add a "root" group they are a part of
[vangheem]


1.0.25 (2017-05-26)
-------------------

- Handle potential issue where data is changing while you're doing the reindex
[vangheem]


1.0.24 (2017-05-26)
-------------------

- Fix issue where a bad original index would screw up index diff calculation
[vangheem]

- Fix migration failure issue before flipping new index on
[vangheem]


1.0.23 (2017-05-25)
-------------------

- Fix reindexing on security change
[vangheem]

1.0.22 (2017-05-19)
-------------------

- Be able to cancel migration and have it do some cleanup
[vangheem]


1.0.21 (2017-05-19)
-------------------

- A bunch of migration fixes to make it more resilient
[vangheem]


1.0.20 (2017-05-18)
-------------------

- Implement live migration command
[vangheem]


1.0.19 (2017-05-16)
-------------------

- Reindex with clean option should delete, re-create index
[vangheem]


1.0.18 (2017-05-16)
-------------------

- Use dummy cache on reindex for all request types
[vangheem]

1.0.17 (2017-05-16)
-------------------

- Disable caching when reindexing
[vangheem]


1.0.16 (2017-05-15)
-------------------

- Use threads when bulk indexing on elasticsearch to make sure to always
keep elasticsearch busy.
[vangheem]

- Forcing the update of mapping
[bloodbare]


1.0.15 (2017-05-12)
-------------------

- close and open the indices to define the settings
[bloodbare]

1.0.14 (2017-05-12)
-------------------

- Enabling size on query by type
[bloodbare]


1.0.13 (2017-05-11)
-------------------

- Also set settings on force_mappings
[bloodare]


1.0.12 (2017-05-11)
-------------------

- Changing permissions name to guillotina
[bloodbare]


1.0.11 (2017-05-11)
-------------------

- Option to set the mapping without reindexing in case of mapping mutation
[bloodbare]


1.0.10 (2017-05-09)
-------------------

- Fix --update-missing argument with `es-reindex` command
[vangheem]


1.0.9 (2017-05-09)
------------------

- Fix bug when deleting nodes
[bloodbare]


1.0.8 (2017-05-08)
------------------

- Improve performance of reindexing of bushy content by using asyncio.gather
to traverse sub-trees and index at the same time.
[vangheem]

- Improve options to reindex command
[vangheem]


1.0.7 (2017-05-04)
------------------

- reindex_all_content takes update and update_missing params now
[vangheem]


1.0.6 (2017-05-03)
------------------

- Be able to update from an existing reindex
[vangheem]


1.0.5 (2017-05-02)
------------------

- Need to avoid using long running queries on reindex because they can timeout
for very large folders
[vangheem]


1.0.4 (2017-05-02)
------------------

- optimize reindex more
[vangheem]


1.0.3 (2017-05-02)
------------------

- More memory efficient reindex
[vangheem]


1.0.2 (2017-05-02)
------------------

- Fix reindexing content
[vangheem]


1.0.1 (2017-04-25)
------------------

- Provide as async utility as it allows us to close connections when the object
is destroyed
[vangheem]


1.0.0 (2017-04-24)
------------------

- initial release


Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

guillotina_elasticsearch-1.3.20.tar.gz (27.8 kB view details)

Uploaded Source

Built Distribution

guillotina_elasticsearch-1.3.20-py3-none-any.whl (40.1 kB view details)

Uploaded Python 3

File details

Details for the file guillotina_elasticsearch-1.3.20.tar.gz.

File metadata

File hashes

Hashes for guillotina_elasticsearch-1.3.20.tar.gz
Algorithm Hash digest
SHA256 6198483843716c39e7a67cee59574a0b9263995ef4ed2f9097e778cf425c01d5
MD5 51daeda3818428c26d8717cc6a830a7f
BLAKE2b-256 80ce8c3fde70f9e01090ee3938b8fe2c2f89799f4cc37e859f7f83b1bcd2b266

See more details on using hashes here.

File details

Details for the file guillotina_elasticsearch-1.3.20-py3-none-any.whl.

File metadata

File hashes

Hashes for guillotina_elasticsearch-1.3.20-py3-none-any.whl
Algorithm Hash digest
SHA256 2ed7d5cbc69ebb73a18f7c64f5dfd75ac00ff33c45d8b1df4c3118eb369afbb2
MD5 e89f63379997d614d24660eb809f3b64
BLAKE2b-256 a715b029a715729810fffb082be2cb065054af83d73c0e93b56bfc8d70a1434d

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page