Skip to content

RELEASE_NOTES.rst

Latest commit

 

History

History
284 lines (200 loc) · 11.6 KB

RELEASE_NOTES.rst

File metadata and controls

284 lines (200 loc) · 11.6 KB

Release notes

Update from 2.6.0 to 2.7.0

IMPORTANT: Allocate Sufficient Time for This Upgrade

Version 2.7.0 introduces a number of significant changes that require careful planning. If you maintain custom add-ons, you should take time to review the update and assess whether your code will need adjustments.

Especially the following upgrade steps are known to take long, as they involve migrating a lot of objects in the ZODB:

  • Upgrade step 2719: Migrate ARReport to Dexterity senaite#2831

IMPORTANT: Manual Reindexing of "modified" Index May Be Required

Version 2.7.0 introduces an update to the internal catalogs by adding a new modified index and metadata column. This change enables improved sorting and filtering by the last modification date for cataloged objects throughout SENAITE.

What does this mean for you?

  • The upgrade process will attempt to add the modified index and column to all SENAITE catalogs automatically.
  • However, for existing catalogs, you may need to manually reindex the new modified index to ensure correct catalog behavior and search results.

How to manually reindex the "modified" index:

  1. Click Site Setup in your SENAITE instance.
  2. Go to the Management Interface.
  3. Navigate to each catalog (e.g., senaite_catalog, senaite_catalog_analysis, etc.) that implements SENAITE catalog interfaces.
  4. In each catalog, select the "Indexes" tab.
  5. Find the modified index in the list.
  6. Click the "Reindex" button at the bottom of the page.

Update from 2.5.0 to 2.6.0

IMPORTANT: Allocate Sufficient Time for This Upgrade

Version 2.6.0 introduces a number of significant changes that require careful planning. If you maintain custom add-ons, you should take time to review the update and assess whether your code will need adjustments.

A key highlight of this release is the migration of many content types from Archetypes (AT) to Dexterity (DX). This is part of the broader effort to modernize the codebase and move closer to full Python 3 compatibility. As a result, the following content types have been migrated:

AnalysisCategories, AnalysisCategory, AnalysisProfiles, AnalysisProfile, AttachmentTypes, AttachmentType, BatchLabels, BatchLabel, ContainerTypes, ContainerType, Departments, Department, InstrumentTypes, InstrumentType, LabProducts, LabProduct, Manufacturers, Manufacturer, Preservations, Preservation, SampleConditions, SampleCondition, SampleMatrices, SampleMatrix, SamplePoints, SamplePoint, SampleTemplates, SampleTemplate, SampleTypes, SampleType, SamplingDeviations, SamplingDeviation, StorageLocations, StorageLocation, SubGroups, SubGroup, Suppliers, Supplier, WorksheetTemplates, WorksheetTemplate.

If you have any extenders or modifiers attached to these types in your own add-ons, you will need to port that functionality to use Dexterity behaviors.

In addition to the migration, version 2.6.0 introduces various improvements across the platform. Some of the most notable enhancements include support for Limit of Quantification (LOQ), improved handling of uncertainty values, better performance of analysis verification, a revamped sticker engine, new paste capabilities in sample registration, enhanced worksheet management, and the ability to remove users and customize user contact information. The release also includes refinements to result entry, multi-result analysis specifications, more robust catalog indexing, and many bug fixes and visual improvements across the user interface.

Several long-standing bugs have been addressed as well, including issues with result entry, catalog indexing, permissions, rejections, and Unicode handling. Performance improvements were made for analysis verification and recalculation, and additional developer utilities were added to the API.

Please note that the upgrade process may take a considerable amount of time. This is primarily due to upgrade step 2654, which involves processing and migrating a large amount of catalog data. Be prepared for a longer-than-usual upgrade window, especially for large databases.

This release has been thoroughly tested with Plone 5.2.15 and 5.2.14. For best compatibility and stability, it is strongly recommended that you pin your buildout to one of these versions when upgrading.

To ensure a smooth upgrade process, we also recommend taking a few precautionary steps beforehand:

  • Run a zeopack to compact the database
  • Create a full backup of your code and database
  • Ensure as many analyses as possible are in Verified or Published status
  • Stop any non-essential services or applications to free up memory
  • Clear or rotate log files to begin with a clean slate

Update from 2.0.0 to 2.1.0

!!! This update requires at least Plone 5.2.5 and Python 2.x. !!!

Please update your buildout.cfg configuration file to this version and re-run the buildout script.

You can use the Plone unified installer to install Plone 5.x first: https://github.com/plone/Installers-UnifiedInstaller

Please make sure to pin senaite.lims to version 2.1.0 to get the correct versions of all dependent SENAITE packages.

Update from 1.3.x to 2.0.0

!!! This update requires at least Plone 5.2.4 and Python 2.x. !!!

Please update your buildout.cfg configuration file to this version and re-run the buildout script.

You can use the Plone unified installer to install Plone 5.x first: https://github.com/plone/Installers-UnifiedInstaller

Please make sure to pin senaite.lims to version 2.0.0 to get the correct versions of all dependent SENAITE packages.

Run first the Plone upgrade before you continue to update SENAITE to version 2.x.

Update from 1.3.0 to 1.3.1

IMPORTANT: Plan the upgrade with enough time

This update might take long depending on the number of objects registered in the system:

  • Stale Sample and Partition objects have been removed from ZODB senaite#1351
  • Full Audit log has been added to Senaite senaite#1324

If you have your own add-on, please review the changes to check beforehand if some parts of your add-on require modifications. Worth to mention that the following tips are strongly recommended before proceeding with the upgrade:

  • Do a zeopack
  • Do a backup of both your code and database
  • Try to have as much analyses in verified/published statuses as possible
  • Stop unnecessary applications and services that may consume RAM
  • Start with a clean log file

Update from 1.2.9 to 1.3.0

IMPORTANT: Plan the upgrade with enough time

Version 1.3.0 is not a hotfix release, it rather comes with a lot of changes that require additional care when planning the update. If you have your own add-on, please review the changes to check beforehand if some parts of your add-on require modifications.

This update will take long (up to 5h for instances with large amounts of data). Therefore, is recommended to plan and allocate enough resources for the process to complete beforehand. For big databases, RAM is a critical factor to be considered before upgrading. Worth to mention that the following tips are strongly recommended before proceeding with the upgrade:

  • Do a zeopack
  • Do a backup of both your code and database
  • Try to have as much analyses in verified/published statuses as possible
  • Stop unnecessary applications and services that may consume RAM
  • Start with a clean log file

Most of the base code has been refactored keeping in mind the following objectives:

  • Less complexity: less code, better code
  • High test coverage: lower chance of undetected bugs
  • Boost performance: better experience, with no delays
  • Improve security: rely on Zope's security policies
  • Code responsibility: focus on core functionalities and let other add-ons to deal with the rest (senaite.lims, senaite.core.listing, etc.)

Besides of this refactoring, this version also comes with a myriad of new functionalities and enhancements: full-fledged sample partitions, reinvented listings and results entry, new adapters for extensibility, etc.

Version 1.3 is the result of hard, but exciting work at same time. Four months of walking through valleys of tears and fighting hydras. Four exciting months to be proud of.

Update from 1.2.8 to 1.2.9

IMPORTANT: Plan the upgrade with enough time This update might take long depending on the number of Analyses, Analysis Requests and Samples registered in the system:

  • Role mappings updated for Analaysis Requests and Samples (rejection) senaite#1041
  • Recatalog of invalidated/retest Analysis Requests (invalidation) senaite#1027
  • Reindex and recatalog of getDueDate for Analysis Requests senaite#1051
  • Reindex of getDueDate for Analyses: senaite#1032
  • Workflow: retract_ar transition has been renamed to invalidate senaite#1027

Update from 1.2.7 to 1.2.8

  • Operators for min and max values have been added. For specifications already present in the system, the result ranges are considered as bounded and closed: [min,max] = {result | min <= result <= max}. senaite#965

Update from 1.2.4 to 1.2.5

  • This update requires the execution of bin/buildout, because Products.TextIndexNG3 has been added. It will help to search by wildcards in TextIndexNG3 indexes instead of looking for the keyword inside wildcards. For now, it is used only in AR listing catalog. https://pypi.python.org/pypi/Products.TextIndexNG3/
  • This update might take long depending on the number of Analyses registered in the system, because the upgrade step will walk through all analyses in order to update those that do not have a valid (non-floatable) duplicate variation value (see #768).

Update from 1.2.3 to 1.2.4

Update from 1.2.2 to 1.2.3

  • IMPORTANT: Plan the upgrade with enough time This update might take long depending on the number of Analysis Requests registered in the system because a new index and column 'assigned_state' has been added in Analysis Requests catalog, that require the catalog to be reindexed (see #637).

Update from 1.2.1 to 1.2.2

  • IMPORTANT: Plan the upgrade with enough time This update might take long depending on the number of Batches registered in the system, because an index from their catalog needs to be reindexed (#574). Also, a new index that affects the Worksheets that have a Worksheet Template assigned has been added and needs to be indexed.

Update from 1.2.0 to 1.2.1

  • This update requires the execution of bin/buildout, because a new dependency has been added: Plone Subrequest
  • With this update, Analyses Services that are inactive, but have active dependent services, will be automatically transitioned to active state. This procedure fixes eventual inconsistencies amongst the statuses of Analyses Services. See #555