Skip to content

CHANGELOG.md

Latest commit

 

History

History
97 lines (58 loc) · 3.68 KB

CHANGELOG.md

File metadata and controls

97 lines (58 loc) · 3.68 KB

1.9.1 (2025-03-19)

  • Added a warning when the language model is not explicitly specified. The default is GPT 3.5, but this is no longer a good option today. In the future, this setting will be required.

1.9.0 (2025-03-19)

  • Added REASONING_EFFORT config option for reasoning models.

1.8.0 (2025-03-18)

  • Supported -t default or TEMPERATURE=default for models that do not support setting the temperature parameter. Currently, some reasoning models such as o3-mini throw an error when temperature is specified.
  • The shorthand model name 4 now points to gpt-4o instead of gpt-4-turbo. However, shorthand model names are now deprecated. Specify full chat model names instead.

1.7.1 (2024-04-12)

  • Fixed broken module import that made the CLI not functioning at all

1.7.0 (2024-04-10)

  • GPT-4 Turbo has entered general availability, so the model shorthand 4 now points to gpt-4-turbo instead of gpt-4. gpt-4-turbo is faster, smarter, and more cost-effective; there is almost no practical reason to choose gpt-4 now.

1.6.0 (2024-04-04)

  • You can now specify more than one file to translate multiple source files sequentially (#20).
  • Added OVERWRITE_POLICY (-w/--overwrite-policy) option, which allows you to specify what happens if the output file already exists (#20).
  • When one of the API calls causes a critical error, the other API calls are now aborted as well.
  • Fixed an error where the translation status was still reported in a non-TTY environment.
  • Fixed an issue that occurred when the status text was longer than the terminal width.

1.5.1 (2024-04-03)

  • Fixed error handling when checking if the output file is writable.
  • Hid a deprecated and undocumented CLI option from --help.

1.5.0 (2024-04-03)

  • Added OUTPUT_FILE_PATH option, which can flexibly transform the input file path into the output file path.
  • Deprecated OUT_SUFFIX option in favor of OUTPUT_FILE_PATH.
  • Enhanced error reporting to be friendlier when a file system error occurs.

1.4.0 (2024-03-30)

  • Checks the permission of the output file/directory before actually calling the API (#17).
  • Fixed: out CLI option is now relative to BASE_DIR rather than the target file, following standard convention of other CLI tools (#16).

1.3.2 (2024-03-14)

  • Fixed a typo in config name API_ENDOPOINT => API_ENDPOINT (#15)

1.3.1 (2024-02-21)

  • Fixed help message.

1.3.0 (2024-02-21)

  • Fixed a bug where some config values could not be set to zero.
  • Added -q/--quiet CLI option.
  • The tool now suppresses progress output by default when not in TTY.

1.2.0 (2024-02-02)

  • Added API_ENDPOINT config option.

1.1.0 (2023-09-01)

  • Added CODE_BLOCK_PRESERVATION_LINES config option.
  • Added OUT_SUFFIX config option.
  • Added out CLI option.

1.0.2 (2023-08-31)

  • Removed unnecessary dependencies.

1.0.1 (2023-08-31)

  • Critical fix for 'bin' executables.

1.0.0 (2023-08-31)

This is the first version released on NPM. It's now published as a standalone CLI tool that can be installed with npm install -g.

BREAKING

  • Renamed the package/repository name from 'markdown-gpt-translator' to 'chatgpt-md-translator'.
  • This tool no longer reads config data from environment variables (except for HTTPS_PROXY). Settings must be written in the config file.
  • Changed a config key name from GPT_TRANSLATOR_BASE_DIR to BASE_DIR.

NEW

  • Changed command-line parser to let you write options like -m3 or --model=3 instead of -m 3 (old style will work, too).
  • Added support for indented code blocks (#7).

0.3.0 (2023-05-21)

  • Added the temperature option.
  • Added the API call interval option.
  • Added HTTPS Proxy support (via HTTPS_PROXY in .env)