PoC/beta version for combining and publishing PDF files for different localisations, as a basis for the Software Project course topic.
under settings there should be the entries for:
lecture_slides_dir: This tells the program from where to fetch the topic slides for the lectures.
header, footer and divider files: Which file should be used. This file should be in the folder with the course specific files.
Fill these with what font, fontsize colour and maxlines you want to use
Settings for calling generative AI
- AI: name of the AI used, eg. Google
- API_KEY: ...
- Model: Model to be called
- Minimum_translations: under this amount of texts the translation is done manually to save API calls
- Request_timeout_ms: minimum timeout between calls
- Max_requests_per_minute: maximum amount of requests per minute
- These should be named as the name of the course and there can be as many as you want. They need to have the following:
- Coursecode: The code for the course
- ai_prompt: Prompt to describe this particular course for the AI
- translate_to: language codes, separated by "," for translations, for example: en, sv for english and swedish translations
- publish_dir: The teams folder where the course slides should be published.
- coursesize: description of the course size
- lectures: How many lectures the course contains
- coursename: The name of the course
- filename_prefix: prefix for the files.
- lectureterm: what the lecture is called (i.e. oppitunti, luento)
- course_slides_dir: where the course specific slides are.
- After these you should have an equal amount of numbered entries as your lecture count followed by your coursename and the topics for the course, separated by semicolons (which cannot be used in course name) ex.
1 = lecturename; topic1; topic2
2 = lecturename2; topic3; topic5
The program allows you to health check the links present in published slides. This is used by using the --linkcheck or -l flag. This prints out any links which no longer hold the linked resource or is not a working website at all.
You can check links on a specific file by specifying it via --checkfile or -f. -f <path to file>
The flag --silent or -s is also available to minimize prints.
The dead links the program finds are saved to a database, which can be externally viewed to see which links have returned which codes and when they have failed the test.