Commit 6b90f4ee authored by LE GAC Renaud's avatar LE GAC Renaud
Browse files

Move the CHANGELOG to the main repository.

parent fe647dff
--------------------------------- CHANGELOG ----------------------------------
HEAD
- Move to web2py 2.13.4, matplotlib 1.5.1 and pandas 0.18.0
- Release non backward compatible due to the evolution of web2py
via the introdution of pyDAL.
- The script build_version and the script run can use command jsduck, sencha,
sphinx-build or web2py behind a docker container.
0.9.5.3 (Feb 2016)
- Move to plugin_dbui 0.8.2.
- Series of modification to run via a docker image.
- build_version.py can install plugins.
0.9.5.2 (Jan 2016)
- Fix a bug in the controller edit_and_insert.
- Fix a bug in the function duplicate_conference.
- Fix bugs in graph_tools and improve the dashboard layout.
0.9.5 (Dec 2015)
- Require plugin_dbui 0.8.1 and libreoffice.
- Few modifications to be compliant with the navigator Chrome.
- List can be extract in OpenDocument format odt.
- Update the documentation.
0.9.3 (Dec 2015)
- Require plugin_dbui 0.7.3
- Add the wizard to create an harvester.
0.9.2 (Nov 2015)
- Require plugin_dbui 0.7.2 or later release.
- Major review of the user guide.
- Protect the table controllers against duplicate pair (harvester, category)
as well as duplicate usage of the publication category.
- Several bugs fixed.
- Fix bug in the list extraction to CSV file.
- Fix a bug in the graph generation when the database is empty.
0.9.0 (Nov 2015)
- Fix pylint warnings and errors in python modules.
- Migrate the documentation of the python modules to Sphinx.
- Add the field synonym to the tables collaborations, countries and publishers.
- Add the possibility to harvest several stores for a given automaton,
team, project and publication code.
- Add the subcommand loop, mysql and dump to the script run.
- Add a list of publishers. Abbreviations are almost compliant with ISO.
- Activate the authors roles.
- Add the preference add_rules_reg_institute.
- Graphs are rendered by the pandas and matplotlib libraries.
- Add protections and fix bugs.
0.8.14 (Sep 2015)
- Transform the modules invenio_tools and harvest_tools into packages.
- Institute parameters are searched in the inspirehep database.
- Major refactoring of the Marc12 decoding. New classes have been introduced:
RecordConf, RecodInst, RecordPubli, RecordThesis.
- Major redesign of harvester logic. Stop the processing as soon as
possible. The order of the checks is oai, authors, affiliation,
collaboration and specific items depending on the publication type.
- The class Automaton replaces PublicationsTool and many of its methods
have been renamed.
0.8.13 (Sep 2015)
- Maintenance release.
- The pytest framework run within the application.
- Series of python test to validate the Record class
on different type of publications (ACL, ACTI, COM, PHD)
- Bugs fixed and protections added in Record, CheckAndFix classes.
- Much more sofisticated algorithm to discover report number(s).
0.8.12 (Aug 2015)
- Bugs fixed release.
0.8.11 (Jul 2015)
- Release is non backward compatible. It mainly consolidate the user interface
- Require plugin_dbui 0.7.1 or later release.
- Rename the database application to preferences.
- Change the type of the database field preferences.value to JSON.
- Interaction with the preferences table is performed via the preferences widget
- Add the node application. It contains the leaves preferences.
- Move the CAS leaves to "application" and destroy the CAS node.
- Migrate the action "edit and insert" and "inser MARCXML" from the harvester
to the wizard node.
- Update the documentation with the new node structure..
- Protect harvester controller against REG_INSTITUTE not defined
- Protect harvester configuration against the same project, the same automaton
but different publications category.
0.8.10 (Mai 2015)
- Minor release, mainly bugs fixed.
- List can be extracted as a CSV or BibTex files.
0.8.9 (Apr 2015)
- Migrate to plugin_dbui 0.6.4 (sphinx documentation).
- Migrate the source code to gitLab.in2p3.fr.
- Use an improved git branching model (master, develop, feature).
- First version of the user documentation under sphinx.
- COM and ACTI publications are now independant object.
- Polish the harvester user interface.
0.8.8.2 (Apr 2015)
- Database(s) connection parameters are moved in the local file _db.py
- refactor the documentation additing few UML diagram (remove in 0.8.9)
- Bug fix in the wizard "edit and insert".
0.8.8.1 (Dec 2014)
- Require plugin_dbui 0.6.1.14 or later version.
Backward compatibility is broken (report interfaces, country default, ...).
The table axes, lists, metrics and publications have to be recreated.
- Remove all reference to CPPM.
- Consolidate harvesters software.
Modify the logic of the harvester by introducing the class CheckAndFix.
Validation and corrections of each record is performed at only one place.
Should improve code stability and maintenance.
- Review class naming of the invenio and harvester tools modules.
- Improve the UI related to the categories, harvester/controller and publication.
- New wizard EditAndInsert.
It can be run standalone or launched from the harvester report.
- Refactor the ux files to ease the buiding of a new database from scratch.
- Use the new callbacks INHIBIT_DELETE_UNDEF and INHIBIT_UPDATE_UNDEF.
- Redesing the lists and metrics interfaces as well as metric view.
- Remove obsolete controller toolbox. It is now replaced by standalone
scripts: export-to-csv, import-from-csv, fix-conference-dates,
fix-collaboration, fix-conference-url,fix-country, fix-defense,
fix-publication-url, fix-report-number, fix-submitted.
- The list of country is almost frozen by using the default list coming
from a geographical database (www.geonames.org).
Harvester can not add country anymore.
- Re-enforce rule for fields: collaborations, defense, conference_dates,
submitted and publications_url
- Add the graph reports.
0.8.7.2 (Sep 2014)
- Migrate to plugin_dbui 0.6.1.7.
- More robust harvester algorithms.
- Better coding of the check & Validate view.
0.8.7 (Jul 2014)
- Migrate to plugin_dbui 0.6.1.5
- Deploy the new functionality resetFields when duplicating publication.
The field status and origin are reset to their default value.
- The field origin is read only in the publication form.
- Add rules to Check & Validate: the teams and projects have to be defined.
- New application script static/app.js allowing to run the application
in debug mode using the dynamic loading of Ext JS. It is activated
using the URL /track_publications/admin?debug
0.8.6 (Jun 2014)
- Bugs fixed.
- Migrate to plugin_dbui 0.6.1.3
- Add and deploy the methods Record.host, oai and oai_url.
- Add the application property reg_institute.
- Check entries with the same value in the origin field.
- Redesign the harvester/check_validate views using inline javascript.
- Redesing the harvester/check_validate views returning LaTex document.
- Add the wizard compare_publications.
- Add the possibility to modify a record from the check_validate view.
0.8.5 (May 2014)
- Migrate to plugin_dbui 0.6.1.1 and web2py 2.9.5
- Define the length of each string field to garanty web2py compatibility.
- Use the controller plugin_dbui/about, documentation and versions.
- Add the field 'origin' in publications.
- Add the table 'application' for configuration parameters.
- Add the application properties authorize_user_login,
authorize_harvester_scan and havester_start_year.
- Add the logic to limit the year range to run harvesters.
- Add the logic to handle the field origin in harvesters.
- Add the login to inhibit user login usign application properties.
0.8.4 (Feb 2014)
- Migrate to plugin_dbui 0.6.0.11, Ext JS 4.2.1 and MathJax 2.0.3.
- Improve the ergonomie of the interface.
- Redesing the documentation page.
- This version is not backward compatible.
c6b881a
- Migrate to plugin_dbui 0.4.14.1
- Proceeding harvester logs publishers, volume and pages.
- Improved algorithm to determine conference location.
- The field section.sort_field is now actif.
- More robust algorithm to determine the publication year.
0.8.3 (Jul 2013)
- Migrate to plugin_dbui 0.4.13.1
- Add the module callbacks, check_tools and view_tools.
- More stringent tests in the "check and validate" procedure.
- Duplicate entries are reject on insert and look for during check.
Difference between duplicate publications are shown in
the "check and validate" report.
- Can't delete or update a publication marked OK.
- Can't delete a reference entry used by at least one publication.
- The wizard "check" can be ran by user and "check validate" by admin.
- Add the generic pdf converter to be used in view (view/tex2pdf.html).
- Pdf reports can be generated for "check and validate" and
"run harvester(s)" operations.
- Define default values for all string fields.
- Add the action compare and fix_nome in the toolbox.
0.8.2 (Apr 2013)
- Consolidation version.
- Migrate to plugin_dbui 0.4.12.1: replace local_import by the conventional
import statement, new syntax for the linked comboxes, and use the filter
CLEAN_SPACES.
- Bugs fixed and polish the user interface.
- Add constant to deal with migration flag and roles.
- Consolidate the authentification procedure.
- Improve the publication column template.
- Add the field category in the list selector.
- Check collaboration in Proceeding, Reports and Talks harvester.
- Rename the field / table "publishers" as "reviews" using label and
the french translation table (poor man approach).
- Deploy filters in the model and the invenio_tools.Record
- Remove the dot in the review abbreviation (CLEAN_PAPER_EDITOR)
- Review the documentaion of the python modules using epytext.
- All documantations have been moved to "static/docs".
- Add a leaf Help/documentations and remove the leaf database schema.
- Add the script buildVersion which is quite useful to build
epydoc documentation.
- First draft of a tool to find duplicate entries.
0.8.1 (Dec 2012)
- Migrate to plugin_dbui 0.4.10.1.
- New node CAS to manage user and their group.
- User and interface handle 3 groups None, user and admin.
- Merge the table diplomas and publications.
- Modify the metrics table by adding conditions and by removing tablename
- Add About and licence stuff.
- Simplify the interface using the teams/project relation ship.
- Polish the user role.
0.7.0 (Nov 2012)
- Migrate to plugin_dbui 0.4.9.7.
- Deploy the new class Selector.
- Deploy linked comboBox in selectors.
- New tree structure for the viewport.
- Add fields publisher, volume, id in the publication filter.
- Add tables axes, controllers and organisation.
- Add a wizard to check and validate records
- Add a wizard to run all harvesters in one go.
- Add a wizard to run a free harvester.
- Add the mode dry run for all harvester wizards.
0.6.0 (Oct 2012)
- Redesign the metric to use the axes approach.
- Remove obsolete code for lists and metrics reporting.
- Polish the user interface.
0.5.1 (Sept 2012)
- Consolidation and bugs fixed
- Myseed map the CPPM organisation
- Add a toolbox via the controller scripts
- Add a model of the MySQL database version 0.4.0.
It was use to migrate to the current version.
0.5.0 (Sept 2012)
- Bugs fixed
- Harvesters can run on INSPIREHEP store.
- New database schema with improved teams/projects axes.
- Improved version of the tools to generates list.
- Improved version of the class Record.
0.4.0 (July 2012)
- Review the logic of the harvester and the error messaging.
Harvester can be parameterised for a team, a project or both.
- Review the interface to enter an MARC XML record. It can be associated
for a team a project or both. It does not required the parameterisation
of an harvester.
- Add an helper tools to extract the list of authors from a string containing
author names and their affiliation.
- Increase the number of character allowed for the authors field.
- Redesign the tool to generate the list. It can be fully configure using
the table lists and sections.
0.3.2 (July 2012)
- Rename the table degrees as levels in order to run with MySQL.
- Add authentication procedure using tools provided by web2py.
- Connection to the database and authentication move in the access.py file.
- Sort the content of all tables (alphabetic, order of arrival, ...).
- Activate the Double click to start the RowEditor in update mode.
- Dedicated page to show software versions and database schema.
- Login via the viewport.
- Review default values in filter and selector.
0.2.1 (Jun 2012)
- Minor release
- ToolTip for all type of form
- Improve the French translation
0.2.0 (Jun 2012)
- First production release.
......@@ -37,12 +37,11 @@ It is in used since 2012 in CNRS/IN2P3.
## Changelog
Consult the Changelog page for fixes and enhancements of each version.
Consult the [Changelog](CHANGELOG) page for fixes and enhancements of
each version.
## License
Copyright Renaud Le Gac and others, 2010-2016.
Distributed under the terms of the
[CECIL](http://www.cecill.info/licences/Licence_CeCILL_V2.1-en.html) license,
LIMBRA is free and open source software.
## Contact
LIMBRA is free and open source software.
\ No newline at end of file
--------------------------------- CHANGELOG ----------------------------------
HEAD
- Move to web2py 2.13.4, matplotlib 1.5.1 and pandas 0.18.0
- Release non backward compatible due to the evolution of web2py
via the introdution of pyDAL.
- The script build_version and the script run can use command jsduck, sencha,
sphinx-build or web2py behind a docker container.
0.9.5.3 (Feb 2016)
- Move to plugin_dbui 0.8.2.
- Series of modification to run via a docker image.
- build_version.py can install plugins.
0.9.5.2 (Jan 2016)
- Fix a bug in the controller edit_and_insert.
- Fix a bug in the function duplicate_conference.
- Fix bugs in graph_tools and improve the dashboard layout.
0.9.5 (Dec 2015)
- Require plugin_dbui 0.8.1 and libreoffice.
- Few modifications to be compliant with the navigator Chrome.
- List can be extract in OpenDocument format odt.
- Update the documentation.
0.9.3 (Dec 2015)
- Require plugin_dbui 0.7.3
- Add the wizard to create an harvester.
0.9.2 (Nov 2015)
- Require plugin_dbui 0.7.2 or later release.
- Major review of the user guide.
- Protect the table controllers against duplicate pair (harvester, category)
as well as duplicate usage of the publication category.
- Several bugs fixed.
- Fix bug in the list extraction to CSV file.
- Fix a bug in the graph generation when the database is empty.
0.9.0 (Nov 2015)
- Fix pylint warnings and errors in python modules.
- Migrate the documentation of the python modules to Sphinx.
- Add the field synonym to the tables collaborations, countries and publishers.
- Add the possibility to harvest several stores for a given automaton,
team, project and publication code.
- Add the subcommand loop, mysql and dump to the script run.
- Add a list of publishers. Abbreviations are almost compliant with ISO.
- Activate the authors roles.
- Add the preference add_rules_reg_institute.
- Graphs are rendered by the pandas and matplotlib libraries.
- Add protections and fix bugs.
0.8.14 (Sep 2015)
- Transform the modules invenio_tools and harvest_tools into packages.
- Institute parameters are searched in the inspirehep database.
- Major refactoring of the Marc12 decoding. New classes have been introduced:
RecordConf, RecodInst, RecordPubli, RecordThesis.
- Major redesign of harvester logic. Stop the processing as soon as
possible. The order of the checks is oai, authors, affiliation,
collaboration and specific items depending on the publication type.
- The class Automaton replaces PublicationsTool and many of its methods
have been renamed.
0.8.13 (Sep 2015)
- Maintenance release.
- The pytest framework run within the application.
- Series of python test to validate the Record class
on different type of publications (ACL, ACTI, COM, PHD)
- Bugs fixed and protections added in Record, CheckAndFix classes.
- Much more sofisticated algorithm to discover report number(s).
0.8.12 (Aug 2015)
- Bugs fixed release.
0.8.11 (Jul 2015)
- Release is non backward compatible. It mainly consolidate the user interface
- Require plugin_dbui 0.7.1 or later release.
- Rename the database application to preferences.
- Change the type of the database field preferences.value to JSON.
- Interaction with the preferences table is performed via the preferences widget
- Add the node application. It contains the leaves preferences.
- Move the CAS leaves to "application" and destroy the CAS node.
- Migrate the action "edit and insert" and "inser MARCXML" from the harvester
to the wizard node.
- Update the documentation with the new node structure..
- Protect harvester controller against REG_INSTITUTE not defined
- Protect harvester configuration against the same project, the same automaton
but different publications category.
0.8.10 (Mai 2015)
- Minor release, mainly bugs fixed.
- List can be extracted as a CSV or BibTex files.
0.8.9 (Apr 2015)
- Migrate to plugin_dbui 0.6.4 (sphinx documentation).
- Migrate the source code to gitLab.in2p3.fr.
- Use an improved git branching model (master, develop, feature).
- First version of the user documentation under sphinx.
- COM and ACTI publications are now independant object.
- Polish the harvester user interface.
0.8.8.2 (Apr 2015)
- Database(s) connection parameters are moved in the local file _db.py
- refactor the documentation additing few UML diagram (remove in 0.8.9)
- Bug fix in the wizard "edit and insert".
0.8.8.1 (Dec 2014)
- Require plugin_dbui 0.6.1.14 or later version.
Backward compatibility is broken (report interfaces, country default, ...).
The table axes, lists, metrics and publications have to be recreated.
- Remove all reference to CPPM.
- Consolidate harvesters software.
Modify the logic of the harvester by introducing the class CheckAndFix.
Validation and corrections of each record is performed at only one place.
Should improve code stability and maintenance.
- Review class naming of the invenio and harvester tools modules.
- Improve the UI related to the categories, harvester/controller and publication.
- New wizard EditAndInsert.
It can be run standalone or launched from the harvester report.
- Refactor the ux files to ease the buiding of a new database from scratch.
- Use the new callbacks INHIBIT_DELETE_UNDEF and INHIBIT_UPDATE_UNDEF.
- Redesing the lists and metrics interfaces as well as metric view.
- Remove obsolete controller toolbox. It is now replaced by standalone
scripts: export-to-csv, import-from-csv, fix-conference-dates,
fix-collaboration, fix-conference-url,fix-country, fix-defense,
fix-publication-url, fix-report-number, fix-submitted.
- The list of country is almost frozen by using the default list coming
from a geographical database (www.geonames.org).
Harvester can not add country anymore.
- Re-enforce rule for fields: collaborations, defense, conference_dates,
submitted and publications_url
- Add the graph reports.
0.8.7.2 (Sep 2014)
- Migrate to plugin_dbui 0.6.1.7.
- More robust harvester algorithms.
- Better coding of the check & Validate view.
0.8.7 (Jul 2014)
- Migrate to plugin_dbui 0.6.1.5
- Deploy the new functionality resetFields when duplicating publication.
The field status and origin are reset to their default value.
- The field origin is read only in the publication form.
- Add rules to Check & Validate: the teams and projects have to be defined.
- New application script static/app.js allowing to run the application
in debug mode using the dynamic loading of Ext JS. It is activated
using the URL /track_publications/admin?debug
0.8.6 (Jun 2014)
- Bugs fixed.
- Migrate to plugin_dbui 0.6.1.3
- Add and deploy the methods Record.host, oai and oai_url.
- Add the application property reg_institute.
- Check entries with the same value in the origin field.
- Redesign the harvester/check_validate views using inline javascript.
- Redesing the harvester/check_validate views returning LaTex document.
- Add the wizard compare_publications.
- Add the possibility to modify a record from the check_validate view.
0.8.5 (May 2014)
- Migrate to plugin_dbui 0.6.1.1 and web2py 2.9.5
- Define the length of each string field to garanty web2py compatibility.
- Use the controller plugin_dbui/about, documentation and versions.
- Add the field 'origin' in publications.
- Add the table 'application' for configuration parameters.
- Add the application properties authorize_user_login,
authorize_harvester_scan and havester_start_year.
- Add the logic to limit the year range to run harvesters.
- Add the logic to handle the field origin in harvesters.
- Add the login to inhibit user login usign application properties.
0.8.4 (Feb 2014)
- Migrate to plugin_dbui 0.6.0.11, Ext JS 4.2.1 and MathJax 2.0.3.
- Improve the ergonomie of the interface.
- Redesing the documentation page.
- This version is not backward compatible.
c6b881a
- Migrate to plugin_dbui 0.4.14.1
- Proceeding harvester logs publishers, volume and pages.
- Improved algorithm to determine conference location.
- The field section.sort_field is now actif.
- More robust algorithm to determine the publication year.
0.8.3 (Jul 2013)
- Migrate to plugin_dbui 0.4.13.1
- Add the module callbacks, check_tools and view_tools.
- More stringent tests in the "check and validate" procedure.
- Duplicate entries are reject on insert and look for during check.
Difference between duplicate publications are shown in
the "check and validate" report.
- Can't delete or update a publication marked OK.
- Can't delete a reference entry used by at least one publication.
- The wizard "check" can be ran by user and "check validate" by admin.
- Add the generic pdf converter to be used in view (view/tex2pdf.html).
- Pdf reports can be generated for "check and validate" and
"run harvester(s)" operations.
- Define default values for all string fields.
- Add the action compare and fix_nome in the toolbox.
0.8.2 (Apr 2013)
- Consolidation version.
- Migrate to plugin_dbui 0.4.12.1: replace local_import by the conventional
import statement, new syntax for the linked comboxes, and use the filter
CLEAN_SPACES.
- Bugs fixed and polish the user interface.
- Add constant to deal with migration flag and roles.
- Consolidate the authentification procedure.
- Improve the publication column template.
- Add the field category in the list selector.
- Check collaboration in Proceeding, Reports and Talks harvester.
- Rename the field / table "publishers" as "reviews" using label and
the french translation table (poor man approach).
- Deploy filters in the model and the invenio_tools.Record
- Remove the dot in the review abbreviation (CLEAN_PAPER_EDITOR)
- Review the documentaion of the python modules using epytext.
- All documantations have been moved to "static/docs".
- Add a leaf Help/documentations and remove the leaf database schema.
- Add the script buildVersion which is quite useful to build
epydoc documentation.
- First draft of a tool to find duplicate entries.
0.8.1 (Dec 2012)
- Migrate to plugin_dbui 0.4.10.1.
- New node CAS to manage user and their group.
- User and interface handle 3 groups None, user and admin.
- Merge the table diplomas and publications.
- Modify the metrics table by adding conditions and by removing tablename
- Add About and licence stuff.
- Simplify the interface using the teams/project relation ship.
- Polish the user role.
0.7.0 (Nov 2012)
- Migrate to plugin_dbui 0.4.9.7.
- Deploy the new class Selector.
- Deploy linked comboBox in selectors.
- New tree structure for the viewport.
- Add fields publisher, volume, id in the publication filter.
- Add tables axes, controllers and organisation.
- Add a wizard to check and validate records
- Add a wizard to run all harvesters in one go.
- Add a wizard to run a free harvester.
- Add the mode dry run for all harvester wizards.
0.6.0 (Oct 2012)
- Redesign the metric to use the axes approach.
- Remove obsolete code for lists and metrics reporting.
- Polish the user interface.
0.5.1 (Sept 2012)
- Consolidation and bugs fixed
- Myseed map the CPPM organisation
- Add a toolbox via the controller scripts
- Add a model of the MySQL database version 0.4.0.
It was use to migrate to the current version.
0.5.0 (Sept 2012)
- Bugs fixed
- Harvesters can run on INSPIREHEP store.
- New database schema with improved teams/projects axes.
- Improved version of the tools to generates list.
- Improved version of the class Record.
0.4.0 (July 2012)
- Review the logic of the harvester and the error messaging.
Harvester can be parameterised for a team, a project or both.
- Review the interface to enter an MARC XML record. It can be associated
for a team a project or both. It does not required the parameterisation
of an harvester.
- Add an helper tools to extract the list of authors from a string containing
author names and their affiliation.
- Increase the number of character allowed for the authors field.
- Redesign the tool to generate the list. It can be fully configure using
the table lists and sections.
0.3.2 (July 2012)
- Rename the table degrees as levels in order to run with MySQL.
- Add authentication procedure using tools provided by web2py.
- Connection to the database and authentication move in the access.py file.
- Sort the content of all tables (alphabetic, order of arrival, ...).
- Activate the Double click to start the RowEditor in update mode.
- Dedicated page to show software versions and database schema.
- Login via the viewport.
- Review default values in filter and selector.
0.2.1 (Jun 2012)
- Minor release
- ToolTip for all type of form
- Improve the French translation
0.2.0 (Jun 2012)
- First production release.
../CHANGELOG
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment