* behavior outline
* FIrst pass at allow external plugins
* remove template exporter
* Add listing of active plugins to '--version' output
* Documentation for plugins
* [Docs] add custom imports and exporters to site TOC
* [Docs] better linewrapping
* enforce positive initial linewrap
Check column widths
update gitignore
throw error when linewrap too small
simply check for large enough linewrap value
* delete unused error message
* PR feedback
make exception more informative
update check_linewrap signature in src and test
make check_linewrap a free function
* delete unused function
* delete else..pass block
* newline for make format
* Include dates_exporter
* Use Base classes for importer and exporters.
* [Docs] improve documentation of custom Importers and Exporters
* [Testing] separate run with external plugin!
* basic behavior test
* prototype unittest for JSON Exporter
test for unimplemented method
* make format
delete unused imports
* Remove 'importer' or 'exporter' from filenames where not needed
* [Test] run different tests with or without the external plugins installed
* [Test] move test rot13 plugin into git tree
from 0dc912af82
* consolidate demo plugins to common package
* [Docs] name page for plugins
* [Docs] include the sample plug in code files directly
* style fixes
* [test] determine whether to run external plug in tests based on installed packages
* improved code documentation
* style fixes for GitHub actions
* Convert "short" and "pretty" (and "default") formaters to plugins
further to https://github.com/jrnl-org/jrnl/pull/1177
* more code clean up
tests pass locally...now for GitHub...
* [tests] dynamically determine jrnl version for plugin tests
* [GitHub Actions] direct install of testing plugins
* Remove template code
* [plugins] meta --> collector
* [Docs] create scripted entries using an custom importer
* (closer to) being able to run behave tests outside project root directory
* We already know when exporter to use
Don't re-calculate it!
* [Tests] don't name test plugin 'testing"
If so named, pip won't install it.
* [Test] run behave tests with test plugins outside project root
* [Test] behave tests pass locally
* [Docs] fix typo
* [GitHub Actions] run test commands from poetry's shell
* black-ify code
* [GitHub Actions] move downstream (rather than up) to run tests
* [GitHub Actions] set shell to poetry
* [GitHub Workflows] Manually activate virtual environment
* [GitHub Actions] Skip Windows & Python 3.8
Can't seem to find Python exe?
* [GiotHub Actions] explicitly use virtual env
* [GitHub Actions] create virutal env directly
* [GitHub Actions] better activate of Windows virtual env
* [GitHub Actions] create virtual env on Mac
* [Github Actions] install wheel and upgrade pip
* [GitHub Actions] skip virtual environments altogether
* [GitHub Actions] change directory for behave test
* Remove Windows exclusions from CI as per note -- they should be working now
Co-authored-by: Suhas <sugas182@gmail.com>
Co-authored-by: Micah Jerome Ellison <micah.jerome.ellison@gmail.com>
* Fix inverted POSIX check, refactor os_compat, and add tests for it
* Fix missing parentheses and remove skip_win on test that is passing in Windows now
* Fix expected quotes in quoted args
* Make output clearer on failing test
* Bringing skip_win back to test whose failure is a bit more complicated than expected
* make behave slightly less verbose for use with behave --format progress2
* standardize behave tests
* move tests around to be more behavior driven
* clean up txt file after tests
* add more tests, add more functionality to behave for calling mock editor
* move around behave tests, get rid of regression files
* clean up some code around keyrings
* add more placeholder test scenarios (marked with @todo)
You can run just these tests with `behave --no-skipped --tags=todo`
* fix "missing_directory" test
This test was missing the config file it was trying to use. So, it was
really a very useless, broken test that we absolutely should not have
approved the PR (#963) for.
* add write tests for each journal type
* update version tests, add new regex match behave step
* add config test outlines
* add journal types to some search tests
* change "basic" config reference to "simple"
* update configs
* add more journal types in search
* fix basic folder journal reference
* add flush output steps to behave, update delete flag tests
* fix failing test with a flush
* update more delete flag tests to include other journal types
* fix file cleanup after failed test with no debug on
* fix password test
* fix DayOne tag sample data, move search/format tag tests, and run them on multiple jrnl types
* added ability to auto-prompt for password for encrypted journals
Only uses password when prompted, and doesn't get in the way of other
input prompts. This allows us to run the same scenarios on both
encrypted journals and other journal types.
* fold encrypted scenarios into the rest of the scenarios where possible
* remove apostrophe that is breaking tests on CI
* add more journal type tests to import feature
* standardize whitespace in behave tests, take out duplicate test
* update handling of cache directories in test suite (easier syntax)
* skip failing YAML exporter emoji test on Windows
* added @todo tags for things that need follow-up
Co-authored-by: Micah Jerome Ellison <micah.jerome.ellison@gmail.com>
* More refactoring of cli.py
break up code from cli.py (now in jrnl.py) up into smaller functions
get rid of export mode
move --encrypt and --decrypt to commands.py
clean up the help screen even more
update flag name for import
* reorganize code, move around lots of functions
* clean up import statements
* move run function out of cli and into jrnl
* rename confusingly named function
* move editor function into editor file
* rename parse_args.py to args.py to make room for more args functions
* Fix error in test suite for windows
I accidentally flipped the conditional, so this fixes it.
Co-authored-by: Micah Jerome Ellison <micah.jerome.ellison@gmail.com>
* Update app description on help screen
Co-authored-by: Micah Jerome Ellison <micah.jerome.ellison@gmail.com>
This adds the ability to run commands in a cache directory without the
test writer knowing where the cache directory is located. This will let
us expand later if we want to start using system temp folders, without
having to rewrite any of our tests.
* clean up extra directories after running behave
* clean up white space issues
* move repeated code into function
* clean up behave code for creating cache directories
* Fix for windows shell parsing in our test suite
Co-authored-by: Micah Jerome Ellison <micah.jerome.ellison@gmail.com>
c.f. commit 7cbca9f60f
c.f. commit 398283a8db3177b1ae122e5b9d6a9d623cf9106f
[Dayone] Skip failing tests on Travis
Travis sets the timezone to UTC, which causes many tests to fail
fstring wip
Run pyupgrade
fix broken pyupgrade fstring
run pyupgrade on plugin dir
fixup! remove py2 remnants and use mocks in tests
small print bugfix
The file=sys.stderr was part of the format(), so an error got printed to
stdout
Drop use of codecs package
Use builtins.open() instead
fixup! remove py2 remnants and use mocks in tests
