Changes in version 2.2.1 (2025-02-21)                  

repana 2.2.1

Production of PDF using the master could fail in some systems. The test
of master with PDFs is removed. Producing HTML is more consistent and it
is the recommended way to use master

repana 2.2.0

Add a targets_structure function to use the targets system instead of
the default used by repana.

Add a logo!. Repana package has now a logo

repana 2.1.0

  - Complete vignette to render reports

repana 2.0.0

  - Modified format of NEWS.md file to track changes to the package.

  - Modify the documentation at package level as requested by CRAN

  - Allow to change the template file for a project

  - Simplify the template file

  - pkgdown is used to create a website for the package in github

  - Change the way timestamp is manage. It does not need anymore to
    include the start and end within the program.

  - Custom YAML tags now control if session info and signature are
    injected to the code

  - Master now works by default with the spin option. User format txt
    option for only txt output

  - Make structure now produce 00_clean.R instead of 01_clean.R

  - Simplified pattern manager for programs. Now all programs with
    format nn_ are included by default in the list of programs executed
    by master

  - Default db changed from SQLite to duckdb as duckdb have similar
    functionality, but is more R friendly and preserver factors and
    dates (see https://duckdb.org/)

repana 1.23.2

  - Fix a bug if an output_file is included as parameter in the
    render_function

repana 1.23.1

  - Remove unused dependencies. Check that report is render only if
    pandoc is available. Test for render reports only if pandoc is
    available

  - Fix the problem of creating repeated *.Rproj in the .gitignore file

#repana 1.23.0

  - The confirm_libraries function has been removed. It was not really
    useful and potentially would not work well if many libraries are
    installed in the system.

  - Messages on the make_structure are now messages and not calls to
    cat() function.

repana 1.22.0

  - Modification of the version to have three numbers, change author
    address to submit to CRAN.

  - Modify license.

  - Modify render_report to not delete the output file if the output
    directory is the same as the directory of the input file.

repana 1.21

  - Small correction in the template

repana 1.20

  - Modify templates. Now only 1 exists which is more friendly for pdf
    and html render and improve the search of Rscript to be windows
    friendly.

  - DBI is declared as dependent and therefore it is loaded
    automatically with the package.

  - Master spin is now by default html but a new parameters allow to
    select the type of render.

  - Name of reports from rcode is now uniformly called spin instead of
    SPIN

repana 1.19

  - Modify opening template to have plyr before dplyr and load tidyverse

repana 1.18

  - Modify the templates to have one for the heading and one for the
    closing part of a SPIN

repana 1.17

  - a master_spin function to run the code as SPINs

repana 1.16

  - New RStudio addin to insert a Template documentation for a SPIN
    program

repana 1.15

  - Include a function to wrap the render of rmarkdown documents within
    a program

repana 1.14

  - Include a wrap to writeDataTable. Change as.character to deparse to
    find the name of the data.frame in update_table.

repana 1.12

  - Fix tests for the make_structure now creates 01_clean.R file

repana 1.11

  - Include the get_dirs() function

repana 1.10

  - update table now accept an argument for the name of the table

repana 1.9

*Fix get_con to use a different config.yml if needed

Version 1.8

  - Include pool connections and update vignette and documentation

Version 1.7

  - change the gitignore entry in config.yml to
    clean_before_new_analysis which is more explanatory for non git
    users.

  - This entry is not back compatible any more with the previous
    config.yml

  - The way to specify the database also changes. Now config.yml
    configure a RSQLite in-memory database. This is also not back
    compatible with previous config.yml

Version 1.6

  - Change the way to define the gitignore, as a list instead of an
    expression

Version 1.5

  - Fix the duplication of entries in gitignore

Version 1.5

  - gitignore is now processed based on the names of gitignore entry of
    the config.yml. Previously was a fixed set of directories

  - make_structure create directories and subdirectories

Version 1.4*

  - New function confirm_libraries()

Version 1.3

  - Include config.yml in .gitignore

  - Change the NEWS to NEWS.md

Version 1.2

*Improve seeking of Rscript package. Works for Linux and OS

Version 1.1

*Include parameter stop to master function

Version 1.0

  - Package ready for deployment