fix GitHub PAT regex (#341)

Author: filipchristiansen

README.md

@@ -54,7 +54,7 @@ You can also replace `hub` with `ingest` in any GitHub URL to access the corresp
 ## 📚 Requirements
 
 - Python 3.8+
-- For private repositories: A GitHub Personal Access Token (PAT). You can generate one at [https://github.com/settings/personal-access-tokens](https://github.com/settings/personal-access-tokens) (Profile → Settings → Developer Settings → Personal Access Tokens → Fine-grained Tokens)
+- For private repositories: A GitHub Personal Access Token (PAT). [Generate your token **here**!](https://github.com/settings/tokens/new?description=gitingest&scopes=repo)
 
 ### 📦 Installation
 

src/gitingest/utils/git_utils.py

@@ -9,7 +9,7 @@
 
 from gitingest.utils.exceptions import InvalidGitHubTokenError
 
-GITHUB_PAT_PATTERN = r"^(?:github_pat_|ghp_)[A-Za-z0-9_]{36,}$"
+GITHUB_PAT_PATTERN = r"^(?:gh[pousr]_[A-Za-z0-9]{36}|github_pat_[A-Za-z0-9]{22}_[A-Za-z0-9]{59})$"
 
 
 def is_github_host(url: str) -> bool:

tests/test_git_utils.py

@@ -29,9 +29,12 @@
     "token",
     [
         # Valid tokens: correct prefixes and at least 36 allowed characters afterwards
-        "github_pat_" + "a" * 36,
+        "github_pat_" + "a" * 22 + "_" + "b" * 59,
         "ghp_" + "A" * 36,
-        "github_pat_1234567890abcdef1234567890abcdef1234",
+        "ghu_" + "B" * 36,
+        "ghs_" + "C" * 36,
+        "ghr_" + "D" * 36,
+        "gho_" + "E" * 36,
     ],
 )
 def test_validate_github_token_valid(token: str) -> None:
@@ -47,6 +50,7 @@ def test_validate_github_token_valid(token: str) -> None:
         "ghp_" + "b" * 35,  # one character short
         "invalidprefix_" + "c" * 36,  # Wrong prefix
         "github_pat_" + "!" * 36,  # Disallowed characters
+        "github_pat_" + "a" * 36,  # Too short after 'github_pat_' prefix
         "",  # Empty string
     ],
 )