@jutge.org/toolkit 4.4.20 → 4.4.22

package/docs/getting-started-guide.md

@@ -121,16 +121,35 @@ Before using the toolkit, you should configure it with your preferences:
     jtk generate problem
     ```
 
-    The toolkit will ask you for:
+    The flag `--kind` can be used to specify the type of problem to generate:
+    - `io`: The problem consists of reading input from standard input and writing output to standard output (use it for most problems and to be used by many programming languages).
+
+    - `funcs`: The problem consists of implementing one or more functions (use it for problems that require the user to implement specific functions that are tested directly with special compilers such as RunPython, RunHaskell or RunClojure).
+
+    ```bash
+    jtk generate problem --kind io
+    jtk generate problem --kind funcs
+    ```
+
+    In the `io` case, the toolkit will ask you for:
     - Problem title
     - Problem description
     - Original language
     - Programming language for the solution
+    - Other languages for the statement
 
     Be specific about the problem description. Here is a possible example:
 
     > The task is to read all numbers in the input and print how much of them are prime numbers. Add a cute, short story around it. The input will only contain natural numbers less than 2^32-1.
 
+    In the `funcs` case, the toolkit will ask you for:
+    - Problem title
+    - Problem description (do not describe the functions yet)
+    - Specification of the functions to implement (this is the place to describe the functions, their headers uses to be enough)
+    - Original language
+    - Programming language for the solution
+    - Other languages for the statement
+
 2. **Review the generated content** in the `my-problem.pbm` directory.
 
 ### Building Your Problem

package/docs/problem-anatomy.md

@@ -188,12 +188,19 @@ updated_at: 2026-01-13T12:47:12.884Z
 
 Statement metadata is stored using YAML syntax in files named `problem.lang.yml`, where `lang` denotes the language code for which the metadata is given (`en`, `es`, `ca`, ...).
 
+A problem should have one unique original language and may have several translations. All translations should refer to the same original language. The system will find the original author through it.
+
 In the case that `lang` denotes the original language of the problem, `problem.lang.yml` should contain the following fields:
 
 - `title`: Title of the problem in the original language.
 - `author`: Full name of the problem setter.
 - `email`: Email of the problem setter.
 
+It can also contain the following fields:
+
+- `event`: Name of the event for which the problem was created.
+- `date`: Date of the event for which the problem was created.
+
 If `lang` denotes a translation, `problem.lang.yml` should contain the following fields:
 
 - `title`: Title of the problem in the translated language.
@@ -201,7 +208,12 @@ If `lang` denotes a translation, `problem.lang.yml` should contain the following
 - `translator_email`: Email of the problem translator.
 - `original_language`: Code for the original language of the problem.
 
-A problem should have one unique original language and may have several translations. All translations should refer to the same original language [CHECK]. The system will find the original author through it.
+It can also contain the following fields:
+
+- `event`: Name of the event for which the problem was created.
+- `date`: Date of the event for which the problem was created.
+
+If the `event` and `date` fields are not given for a translation, the system will use the values of the original language.
 
 All the values for the metadata fields should be given as Unicode-based plain text. Remember to enclose them in quotes if they contain special characters such as `:` or `'`.
 
@@ -246,6 +258,8 @@ There are also optional arguments that may be used:
 
 - `presentation_error`: If set to `1` (by default), the PE (Presentation Error) verdict will be enabled. Otherwise, the verdict will be WA (Wrong Answer) if the files are not identical.
 
+- `tolerance`: Tolerance for the graphic checker. It is a number between 0 and 1.
+
 - `solution`: Programming language of the golden solution. By default, it is C++. Currently supported programming languages are C, C++, Java, Python3, Haskell, Clojure, Rust, RunPython, RunHaskell, and RunClojure.
 
 - `pylibs`: Problems in Python can allow the usage of certain non-standard Python libraries. `pylibs` is a list of libraries that may be used to solve the current problem. For instance, `pylibs: [numpy, networkx]`. By default, this list is empty. Available libraries are at https://jutge.org/documentation/pylibs.

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "@jutge.org/toolkit",
     "description": "Toolkit to prepare problems for Jutge.org",
-    "version": "4.4.20",
+    "version": "4.4.22",
     "homepage": "https://jutge.org",
     "author": {
         "name": "Jutge.org",