Repository navigation
Proposing a new challenge
- It must have between 20 and 30 questions.
- Images must comply with the source image license obligations.
- The content must not have anything that can be discriminating on the basis of race, color, national origin, sex, sexual identity, gender identity, religion or age.
- The content must be appropriate for children and must not contain drugs, pornography or explicit violence.
Images must comply with the source image license obligations.
At the time of writing, the feature that allows credit attribution in the images has not been implemented yet. For that reason, we recommend using images licensed under Creative Commons 0 or made by you yourself.
The images must be PNG files with a size of 256x256 pixels, and should be as much compressed as possible. The name of the file indicates the order of the questions, being 1.png the image used for the first question and 2.png the image used for the second question.
TIP: If you use GIMP for generating the images, every time you are about to export the image as PNG make sure to change its mode to Indexed (Image > Mode > Indexed...). Also, when the export prompt pops up, uncheck every option.
The data of the challenge is composed of:
- description (string): this is the text that will be rendered just above the
Will you guess all?paragraph in the header of the challenge. Please provide a helpful description introducing and explaining the challenge to the users. Note: You can expand to different paragraphs using\n. - startingDate (string): This is when the challenge will be available to the users. Please provide a string with an ISO format.
- contributorName (string): This is the name/nickname that will appear in the footer. Leave empty if you want the challenge to be anonymous.
- questions ({ solutions: string[] }[]): These are the questions that will be rendered, the only field they need at the moment is an array with the hashed solutions.
- solutions(string[]): This is the list of the hashed solutions against which the answers will be checked. More information on how to hash them below.
In the end, we will end up with a json like the following
{
"description": "One line.\nOther line.",
"startingDate": "2022-04-22",
"contributorName": "ILoveFoss",
"questions": [
{
"solutions": ["la tia de maria (hashed)", "la tía de maría (hashed)"]
},
// ...
]
}
The solutions are hashed using the Argon2 password hashing algorithm. You can find a web implementation in this website.
The hashed solutions must be generated with the following configuration in order to be understood:
- Salt: Tobertet
- Memory: 1024
- Iterations: 1
- Hash length: 32
- Parallelism: 1
- Type: Argon2di
The answer provided by the user is both trimmed and converted to lower case. For this reason, the password must be provided in lower case. Also, if the password contains any accent, it is recommended to generate a version for that.
If the answer were
The dog, then I would only need to hash the stringthe dog. If the answer wereLa tía de María, then I would need to hash at leastla tia de mariaandla tía de maría. In the later example it is recommended as well to hashla tía de mariaandla tia de maría.
Once the password (unhashed solution) has been typed in, click on the Run WebAssembly button. The hashed solution will appear in the console after the string Hash:.