Skip to main content

Scriv changelog management tool

Project description

Scriv changelog management tool

PyPI Build status Documentation Supported Python versions License Sponsor me on GitHub nedbat on Twitter

Overview

Scriv is a command-line tool for helping developers maintain useful changelogs. It manages a directory of changelog fragments. It aggregates them into entries in a CHANGELOG file.

Getting Started

Scriv writes changelog fragments into a directory called “changelog.d”. Start by creating this directory. (By the way, like many aspects of scriv’s operation, you can choose a different name for this directory.)

To make a new changelog fragment, use the scriv create command. It will make a new file with a filename using the current date and time, your GitHub or Git user name, and your branch name. Changelog fragments should be committed along with all the other changes on your branch.

When it is time to release your project, the scriv collect command aggregates all the fragments into a new entry in your changelog file.

Documentation

Full documentation is at https://scriv.readthedocs.org.

License

The code in this repository is licensed under the Apache Software License 2.0 unless otherwise noted.

Please see LICENSE.txt for details.

How To Contribute

Contributions are very welcome.

Changelog

Unreleased

See the fragment files in the changelog.d directory.

0.15.0 — 2022-04-24

Removed

  • Dropped support for Python 3.6.

Added

  • The github-release command parses the changelog and creates GitHub releases from the entries. Changed entries will update the corresponding release.
  • Added a --version option.

Changed

  • Parsing of fragments now only attends to the top-level section headers, and includes nested headers instead of splitting on all headers.

0.14.0 — 2022-03-23

Added

  • Add an anchor before each version section in the output of scriv collect so URLs for the sections are predictable and stable for each new version (Fixes issue 46). Thanks Abhilash Raj and Rodrigo Girão Serrão.

Fixed

  • Markdown fragments weren’t combined properly. Now they are. Thanks Rodrigo Girão Serrão.

0.13.0 — 2022-01-23

Added

  • Support finding version information in TOML files (like pyproject.toml) using the literal configuration directive. Thanks, Kurt McKee

0.12.0 — 2021-07-28

Added

  • Fragment files in the fragment directory will be skipped if they match the new configuration value skip_fragments, a glob pattern. The default value is “README.*”. This lets you put a README.md file in that directory to explain its purpose, as requested in issue 40.

Changed

  • Switched from “toml” to “tomli” for reading TOML files.

Fixed

  • Setting format=md didn’t properly cascade into other default settings, leaving you with RST settings that needed to be explicitly overridden (issue 39). This is now fixed.

0.11.0 — 2021-06-22

Added

  • A new poorly documented API is available. See the Scriv, Changelog, and Fragment classes in the scriv.scriv module.

Changed

  • Python 3.6 is now the minimum supported Python version.

Fixed

  • The changelog is now always written as UTF-8, regardless of the default encoding of the system. Thanks, Hei (yhlam).

0.10.0 — 2020-12-27

Added

  • Settings can now be read from a pyproject.toml file. Install with the “[toml]” extra to be sure TOML support is available. Closes issue 9.
  • Added the Philosophy section of the docs.

Changed

  • The default entry header no longer puts the version number in square brackets: this was a misunderstanding of the keepachangelog formatting.
  • Respect the existing newline style of changelog files. (#14) This means that a changelog file with Linux newlines on a Windows platform will be updated with Linux newlines, not rewritten with Windows newlines. Thanks, Kurt McKee.

Fixed

  • Support Windows’ directory separator (\) in unit test output. (#15) This allows the unit tests to run in Windows environments. Thanks, Kurt McKee.
  • Explicitly specify the directories and files that Black should scan. (#15) This prevents Black from scanning every file in a virtual environment. Thanks, Kurt McKee.
  • Using “literal:” values in the configuration file didn’t work on Python 3.6 or 3.7, as reported in issue 18. This is now fixed.

0.9.2 — 2020-08-29

  • Packaging fix.

0.9.0 — 2020-08-29

Added

  • Markdown format is supported, both for fragments and changelog entries.
  • Fragments can be mixed (some .rst and some .md). They will be collected and output in the format configured in the settings.
  • Documentation.
  • “python -m scriv” now works.

Changed

  • The version number is displayed in the help message.

0.8.1 — 2020-08-09

Added

  • When editing a new fragment during “scriv create”, if the edited fragment has no content (only comments or blank lines), then the create operation will be aborted, and the file will be removed. (Closes issue 2.)

Changed

  • If the fragment directory doesn’t exist, a simple direct message is shown, rather than a misleading FileNotFound error (closes issue 1).

Fixed

  • When not using categories, comments in fragment files would be copied to the changelog file (issue 3). This is now fixed.
  • RST syntax is better understood, so that hyperlink references and directives will be preserved. Previously, they were mistakenly interpreted as comments and discarded.

0.8.0 — 2020-08-04

Added

  • Added the collect command.
  • Configuration is now read from setup.cfg or tox.ini.
  • A new configuration setting, rst_section_char, determines the character used in the underlines for the section headings in .rst files.
  • The new_entry_template configuration setting is the name of the template file to use when creating new entries. The file will be found in the fragment_directory directory. The file name defaults to new_entry.FMT.j2. If the file doesn’t exist, an internal default will be used.
  • Now the collect command also includes a header for the entire entry. The underline is determined by the “rst_header_char” settings. The heading text is determined by the “header” setting, which defaults to the current date.
  • The categories list in the config can be empty, meaning entries are not categorized.
  • The create command now accepts –edit (to open the new entry in your text editor), and –add (to “git add” the new entry).
  • The collect command now accepts –edit (to open the changelog file in an editor after the new entries have been collected) and –add (to git-add the changelog file and git rm the entries).
  • The names of the main git branches are configurable as “main_branches” in the configuration file. The default is “master”, “main”, and “develop”.
  • Configuration values can now be read from files by prefixing them with “file:”. File names will be interpreted relative to the changelog.d directory, or will be found in a few files installed with scriv.
  • Configuration values can interpolate the currently configured format (rst or md) with “${config:format}”.
  • The default value for new templates is now “file: new_entry.${config:format}.j2”.
  • Configuration values can be read from string literals in Python code with a “literal:” prefix.
  • “version” is now a configuration setting. This will be most useful when used with the “literal:” prefix.
  • By default, the title of collected changelog entries includes the version if it’s defined.
  • The collect command now accepts a --version option to set the version name used in the changelog entry title.

Changed

  • RST now uses minuses instead of equals.
  • The create command now includes the time as well as the date in the entry file name.
  • The –delete option to collect is now called –keep, and defaults to False. By default, the collected entry files are removed.
  • Created file names now include the seconds from the current time.
  • “scriv create” will refuse to overwrite an existing entry file.
  • Made terminology more uniform: files in changelog.d are “fragments.” When collected together, they make one changelog “entry.”
  • The title text for the collected changelog entry is now created from the “entry_title_template” configuration setting. It’s a Jinja2 template.
  • Combined the rst_header_char and rst_section_char settings into one: rst_header_chars, which much be exactly two characters.
  • Parsing RST fragments is more flexible: the sections can use any valid RST header characters for the underline. Previously, it had to match the configured RST header character.

Fixed

  • Fragments with no category header were being dropped if categories were in use. This is now fixed. Uncategorized fragments get sorted before any categorized fragments.

0.1.0 — 2019-12-30

  • Doesn’t really do anything yet.

Project details


Download files

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

Source Distribution

scriv-0.15.0.tar.gz (60.2 kB view hashes)

Uploaded source

Built Distribution

scriv-0.15.0-py2.py3-none-any.whl (27.5 kB view hashes)

Uploaded py2 py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page