Markdown-Checker 1.2.2__tar.gz → 1.3.0__tar.gz

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: markdown-checker
-Version: 1.2.2
+Version: 1.3.0
 Summary: A markdown link validation reporting tool.
 Author-email: John Aziz <johnaziz269@gmail.com>
 Maintainer-email: John Aziz <johnaziz269@gmail.com>
@@ -26,6 +26,7 @@ Project-URL: Documentation, https://markdown-checker.readthedocs.io/en/latest/
 Project-URL: Homepage, https://github.com/john0isaac/markdown-checker
 Project-URL: Issues, https://github.com/john0isaac/markdown-checker/issues
 Project-URL: Releases, https://github.com/john0isaac/markdown-checker/releases
+Import-Name: markdown_checker
 
 [![PyPi](https://img.shields.io/pypi/v/markdown-checker)](https://pypi.org/project/markdown-checker/)
 [![Documentation Status](https://readthedocs.org/projects/markdown-checker/badge/?version=latest)](https://markdown-checker.readthedocs.io/en/latest/?badge=latest)
@@ -77,7 +78,7 @@ exits with a non-zero status:
 🔍 Checked 42 links in 10 files.
 😭 Found 1 issues in the following files:
     File 'docs/index.md', line 5
-./missing.md is broken.
+./missing.md could not be found at /docs/missing.md.
 ```
 
 ## Using `markdown-checker` in GitHub Actions

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/README.md

@@ -48,7 +48,7 @@ exits with a non-zero status:
 🔍 Checked 42 links in 10 files.
 😭 Found 1 issues in the following files:
     File 'docs/index.md', line 5
-./missing.md is broken.
+./missing.md could not be found at /docs/missing.md.
 ```
 
 ## Using `markdown-checker` in GitHub Actions

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/explanation/link-detection.rst

@@ -38,16 +38,19 @@ Fragments on paths
 ----------------------
 
 A path may carry a trailing ``?query`` and/or ``#anchor`` - for example
-``docs/usage.md#configuration``. Both checks that operate on paths strip
-this fragment before resolving the file, so only the ``docs/usage.md``
-part needs to exist on disk; the anchor itself is never validated against
-the target file's headings.
+``docs/usage.md#configuration``. As in any URL, everything from the first
+``?`` or ``#`` onward is stripped before resolving the file, so only the
+``docs/usage.md`` part needs to exist on disk; the anchor itself is never
+validated against the target file's headings. The remaining path is then
+percent-decoded, so ``my%20file.md`` refers to ``my file.md``. A file name
+that really contains ``#`` or ``?`` must be written as ``%23``/``%3F``.
 
 Resolving a path to a file
 --------------------------------
 
 ``check_broken_paths`` tries to resolve a path relative to the markdown
 file it was found in first (e.g. ``../img.png`` in ``docs/usage.md``
-resolves against ``docs/``), then falls back to resolving it as an
-absolute or current-working-directory-relative path. A path is reported as
-broken only if neither resolution finds an existing file.
+resolves against ``docs/``), then relative to the current working
+directory. Root-relative paths (``/docs/usage.md``) are tried the same way
+with the leading ``/`` dropped. A path is reported as broken only if
+neither resolution finds an existing file.

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/howto/pyproject-config.rst

@@ -18,6 +18,7 @@ Add a ``[tool.markdown-checker]`` table to your project's
     extensions = [".md", ".ipynb"]
     skip-files = ["CODE_OF_CONDUCT.md", "SECURITY.md"]
     tracking-domains = ["github.com", "microsoft.com", "visualstudio.com", "aka.ms", "azure.com"]
+    tracking-param = ["wt.mc_id"]
     guide-url = "https://github.com/john0isaac/markdown-checker/blob/main/CONTRIBUTING.md"
     max-workers = 10
     report-format = "markdown"

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/howto/skip-links.rst

@@ -17,8 +17,9 @@ check automatically (e.g. it blocks bots) but you still want to keep the link:
 
     markdown-checker . -f check_broken_urls --skip-domains=example.com,test.com
 
-Matching is a substring match against the URL's hostname, so
-``--skip-domains=example.com`` also matches ``www.example.com``.
+A domain matches the URL's hostname exactly or as a parent domain, so
+``--skip-domains=example.com`` also matches ``www.example.com`` (but not
+``notexample.com``).
 
 Skip URLs containing a substring
 ------------------------------------
@@ -35,7 +36,7 @@ List syntax
 -----------
 
 Every list option (``--skip-domains``, ``--skip-urls-containing``,
-``--extensions``, ``--skip-files``, ``--tracking-domains``) takes a single
+``--extensions``, ``--skip-files``, ``--tracking-domains``, ``--tracking-param``) takes a single
 comma-separated value:
 
 - Correct: ``--skip-domains=example.com,test.com``

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/reference/checks.rst

@@ -26,7 +26,7 @@ Sample finding
     .. code-block:: text
 
         File 'docs/index.md', line 5
-        ./missing.md is broken.
+        ./missing.md could not be found at /docs/missing.md.
 
 ``check_broken_urls``
 ------------------------
@@ -55,8 +55,10 @@ Sample finding
 
 Flags
     URLs whose path contains a locale segment, e.g. ``/en-us/`` in
-    ``https://example.com/en-us/docs``. A "locale segment" is a
-    language-country pair matching ``xx-xx`` (case-insensitive).
+    ``https://example.com/en-us/docs`` or a trailing ``/en-us``. A "locale
+    segment" is a known language-country pair of the form ``xx-xx``
+    (case-insensitive), so look-alikes such as ``/ci-cd/`` aren't flagged.
+    The host, query string and anchor are not searched.
 Ignores
     ``www.nvidia.com`` (built in - its docs always include a locale
     segment), plus any host in ``--skip-domains`` or URL in
@@ -73,8 +75,8 @@ Sample finding
 --------------------------
 
 Flags
-    URLs on a configured tracking domain that are missing a ``wt.mc_id``
-    query parameter.
+    URLs on a configured tracking domain that are missing every
+    ``--tracking-param`` query parameter (``wt.mc_id`` by default).
 Ignores
     URLs on hosts *not* listed in ``--tracking-domains`` (the check does
     nothing for them), plus any host in ``--skip-domains`` or URL in
@@ -83,7 +85,8 @@ Default tracking domains
     ``github.com``, ``microsoft.com``, ``visualstudio.com``, ``aka.ms``,
     ``azure.com``.
 Options that affect it
-    ``--tracking-domains``, ``--skip-domains``, ``--skip-urls-containing``.
+    ``--tracking-domains``, ``--tracking-param``, ``--skip-domains``,
+    ``--skip-urls-containing``.
 Sample finding
     .. code-block:: text
 
@@ -94,10 +97,12 @@ Sample finding
 ---------------------------
 
 Flags
-    Every relative path link missing a ``wt.mc_id`` query parameter. Unlike
-    ``check_urls_tracking``, there is no domain filter - all paths are checked.
+    Every relative path link missing every ``--tracking-param`` query
+    parameter (``wt.mc_id`` by default). Unlike ``check_urls_tracking``,
+    there is no domain filter - all paths are checked.
 Options that affect it
-    ``SRC``/``--dir``, ``--extensions``, ``--skip-files`` only.
+    ``SRC``/``--dir``, ``--extensions``, ``--skip-files``,
+    ``--tracking-param``.
 Sample finding
     .. code-block:: text
 

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/reference/cli.rst

@@ -78,14 +78,33 @@ Type
 Default
   ``github.com,microsoft.com,visualstudio.com,aka.ms,azure.com``.
 Constraints
-  Only affects ``check_urls_tracking``.
+  Only affects ``check_urls_tracking``. Subdomains are included, e.g.
+  ``microsoft.com`` also matches ``learn.microsoft.com``.
 Description
-  Hostnames that must carry a ``wt.mc_id`` tracking parameter.
+  Hostnames that must carry a tracking parameter (see ``--tracking-param``).
 
 .. code-block:: bash
 
     markdown-checker . -f check_urls_tracking --tracking-domains=github.com,microsoft.com
 
+``-tp``, ``--tracking-param``
+---------------------------------
+
+Type
+  ``list[str]``, comma-separated.
+Default
+  ``wt.mc_id``.
+Constraints
+  Only affects ``check_urls_tracking`` and ``check_paths_tracking``.
+  Matched case-insensitively as a query parameter (after ``?`` or ``&``).
+Description
+  Tracking query parameters a link must carry. A link passes if it has any
+  one of them.
+
+.. code-block:: bash
+
+    markdown-checker . -f check_urls_tracking --tracking-param=wt.mc_id,utm_source
+
 ``-sf``, ``--skip-files``
 -----------------------------
 
@@ -110,8 +129,9 @@ Type
 Default
   ``[]`` (none).
 Constraints
-  Only affects ``check_broken_urls`` and ``check_urls_locale``. Substring
-  match against the URL's hostname.
+  Only affects ``check_broken_urls`` and ``check_urls_locale``. Matches the
+  URL's hostname exactly or as a parent domain (``example.com`` matches
+  ``www.example.com`` but not ``notexample.com``).
 Description
   Domains to exclude from URL checks.
 

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/reference/configuration.rst

@@ -47,6 +47,9 @@ Keys
    * - ``tracking-domains``
      - ``--tracking-domains``
      - array of strings
+   * - ``tracking-param``
+     - ``--tracking-param``
+     - array of strings
    * - ``skip-files``
      - ``--skip-files``
      - array of strings
@@ -96,6 +99,7 @@ Example
     skip-files = ["CODE_OF_CONDUCT.md", "SECURITY.md"]
     skip-domains = []
     tracking-domains = ["github.com", "microsoft.com", "visualstudio.com", "aka.ms", "azure.com"]
+    tracking-param = ["wt.mc_id"]
     guide-url = "https://github.com/john0isaac/markdown-checker/blob/main/CONTRIBUTING.md"
     timeout = 20
     retries = 3

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/reference/report-formats.rst

@@ -46,7 +46,7 @@ A structured document including both errors and warnings:
 .. code-block:: json
 
     {
-      "tool": {"name": "markdown-checker", "version": "1.2.2"},
+      "tool": {"name": "markdown-checker", "version": "1.3.0"},
       "check": "check_broken_urls",
       "summary": {
         "files_checked": 10,

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/docs/tutorial.rst

@@ -63,7 +63,7 @@ You'll see output similar to:
     🔍 Checked 2 links in 2 files.
     😭 Found 1 issues in the following files:
         File 'docs-sample/index.md', line 5
-    ./missing.md is broken.
+    ./missing.md could not be found at /docs/missing.md.
 
 Because an error-level issue was found, markdown-checker also writes a
 ``comment.md`` file in the current directory with the same information as a

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/pyproject.toml

@@ -67,7 +67,7 @@ typing = [
 ]
 
 [build-system]
-requires = ["flit_core<4"]
+requires = ["flit_core<5"]
 build-backend = "flit_core.buildapi"
 
 [project.scripts]

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/__init__.py

@@ -2,7 +2,7 @@
 markdown-checker - a markdown link validation reporting tool.
 """
 
-__version__ = "1.2.2"
+__version__ = "1.3.0"
 
 from markdown_checker.cli import main
 from markdown_checker.models import Config

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/checks/broken_paths.py

@@ -28,6 +28,7 @@ class BrokenPathsCheck(BaseCheck[MarkdownPath]):
         detected_issues: list[MarkdownPath] = []
         for path in links.paths:
             if not path.exists():
-                path.issue = "is broken"
+                path.resolved_path = path.get_full_path()
+                path.issue = f"could not be found at {path.resolved_path.as_posix()}"
                 detected_issues.append(path)
         return detected_issues

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/checks/broken_urls.py

@@ -58,7 +58,7 @@ class BrokenURLsCheck(BaseCheck[MarkdownURL]):
         :meth:`collect` to resolve later.
         """
         config = config or Config()
-        effective_skip = [domain.lower() for domain in [*config.skip_domains, *_BUILTIN_SKIP_DOMAINS]]
+        effective_skip = [*config.skip_domains, *_BUILTIN_SKIP_DOMAINS]
         skip_urls_containing = config.skip_urls_containing
 
         # Standalone fallback (direct calls, tests): own a private service for this run only.
@@ -67,10 +67,7 @@ class BrokenURLsCheck(BaseCheck[MarkdownURL]):
 
         pending: list[tuple[MarkdownURL, Future[URLCheckResult]]] = []
         for url in links.urls:
-            hostname = url.host_name().lower()
-            if any(domain in hostname for domain in effective_skip) or any(
-                substring in url.link for substring in skip_urls_containing
-            ):
+            if url.matches_domain(effective_skip) or any(substring in url.link for substring in skip_urls_containing):
                 continue
             pending.append((url, _service.submit(url)))
         return (pending, _service if owns_service else None)

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/checks/locale.py

@@ -40,10 +40,7 @@ class URLsLocaleCheck(BaseCheck[MarkdownURL]):
 
         detected_issues: list[MarkdownURL] = []
         for url in links.urls:
-            hostname = url.host_name().lower()
-            if any(domain.lower() in hostname for domain in effective_skip) or any(
-                substring in url.link for substring in skip_urls_containing
-            ):
+            if url.matches_domain(effective_skip) or any(substring in url.link for substring in skip_urls_containing):
                 continue
             if url.has_locale():
                 url.issue = "has locale"

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/checks/tracking.py

@@ -1,4 +1,4 @@
-"""The ``check_urls_tracking``/``check_paths_tracking`` checks: flag links missing a ``wt.mc_id`` tracking ID."""
+"""The ``check_urls_tracking``/``check_paths_tracking`` checks: flag links missing a tracking ID."""
 
 from markdown_checker.checks.base import BaseCheck
 from markdown_checker.models import Config
@@ -21,8 +21,9 @@ class URLsTrackingCheck(BaseCheck[MarkdownURL]):
         service: URLCheckService | None = None,
     ) -> list[MarkdownURL]:
         """Flag every URL in ``links.urls`` whose host matches
-        ``config.tracking_domains`` but is missing a ``wt.mc_id`` query
-        parameter (see :meth:`MarkdownLinkBase.has_tracking
+        ``config.tracking_domains`` but is missing all of the
+        ``config.tracking_param`` query parameters (see
+        :meth:`MarkdownLinkBase.has_tracking
         <markdown_checker.models.base.MarkdownLinkBase.has_tracking>`). URLs
         on other hosts, or matching ``config.skip_domains``/
         ``skip_urls_containing``, are ignored. Sets
@@ -36,12 +37,9 @@ class URLsTrackingCheck(BaseCheck[MarkdownURL]):
 
         detected_issues: list[MarkdownURL] = []
         for url in links.urls:
-            hostname = url.host_name().lower()
-            if any(domain.lower() in hostname for domain in skip_domains) or any(
-                substring in url.link for substring in skip_urls_containing
-            ):
+            if url.matches_domain(skip_domains) or any(substring in url.link for substring in skip_urls_containing):
                 continue
-            if any(domain.lower() in hostname for domain in tracking_domains) and not url.has_tracking():
+            if url.matches_domain(tracking_domains) and not url.has_tracking(config.tracking_param):
                 url.issue = "is missing tracking id"
                 detected_issues.append(url)
         return detected_issues
@@ -59,16 +57,18 @@ class PathsTrackingCheck(BaseCheck[MarkdownPath]):
         config: Config | None = None,
         service: URLCheckService | None = None,
     ) -> list[MarkdownPath]:
-        """Flag every path in ``links.paths`` missing a ``wt.mc_id`` query
-        parameter (see :meth:`MarkdownLinkBase.has_tracking
+        """Flag every path in ``links.paths`` missing all of the
+        ``config.tracking_param`` query parameters (see
+        :meth:`MarkdownLinkBase.has_tracking
         <markdown_checker.models.base.MarkdownLinkBase.has_tracking>`),
         setting ``issue="is missing tracking id"`` (error-level) on each.
         Unlike ``check_urls_tracking``, there is no domain filter: every
-        path is checked. ``config`` and ``service`` are unused.
+        path is checked. ``service`` is unused.
         """
+        config = config or Config()
         detected_issues: list[MarkdownPath] = []
         for path in links.paths:
-            if not path.has_tracking():
+            if not path.has_tracking(config.tracking_param):
                 path.issue = "is missing tracking id"
                 detected_issues.append(path)
         return detected_issues

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/cli.py

@@ -11,6 +11,7 @@ import click
 from markdown_checker import __version__
 from markdown_checker.checker import run_check_on_files
 from markdown_checker.models.config import Config
+from markdown_checker.patterns import DEFAULT_TRACKING_PARAMS
 from markdown_checker.reports import build_report
 from markdown_checker.reports import get_renderer
 from markdown_checker.reports import RENDERERS
@@ -123,6 +124,15 @@ class ListOfStrings(click.Option):
     help="List of tracking domains to check.",
     required=False,
 )
+@click.option(
+    "-tp",
+    "--tracking-param",
+    cls=ListOfStrings,
+    type=list[str],
+    default=list(DEFAULT_TRACKING_PARAMS),
+    help="List of tracking query parameters; a link passes if it has any of them.",
+    required=False,
+)
 @click.option(
     "-sf",
     "--skip-files",
@@ -258,6 +268,7 @@ def main(
     skip_domains: list[str],
     skip_urls_containing: list[str],
     tracking_domains: list[str],
+    tracking_param: list[str],
     timeout: int,
     retries: int,
     retry_on_429: bool,
@@ -309,6 +320,7 @@ def main(
         skip_domains=skip_domains,
         skip_urls_containing=skip_urls_containing,
         tracking_domains=tracking_domains,
+        tracking_param=tracking_param,
         timeout=timeout,
         retries=retries,
         retry_on_429=retry_on_429,

markdown_checker-1.3.0/src/markdown_checker/models/base.py

@@ -0,0 +1,84 @@
+"""Shared base dataclass for the two kinds of link markdown-checker extracts: URLs and paths."""
+
+import locale
+from abc import ABC
+from collections.abc import Sequence
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Literal
+from urllib.parse import urlparse
+
+from markdown_checker.patterns import DEFAULT_TRACKING_PARAMS
+from markdown_checker.patterns import LOCALE_PATTERN
+from markdown_checker.patterns import tracking_pattern
+
+
+@dataclass(slots=True)
+class MarkdownLinkBase(ABC):
+    """Base class for a single link found in a markdown file."""
+
+    link: str
+    """The raw link destination as written in the file (e.g.
+    ``"https://example.com"`` or ``"../img.png"``).
+    """
+
+    line_number: int
+    """1-based line number where the link was found."""
+
+    file_path: Path
+    """Path to the markdown file the link was found in."""
+
+    issue: str = ""
+    """Human-readable description set by a check when this link fails it
+    (e.g. ``"is broken"``); empty when no issue was found.
+    """
+
+    issue_level: Literal["error", "warning"] = "error"
+    """Severity of ``issue`` once set. ``"error"`` fails the CLI run (exit
+    code 1); ``"warning"`` is reported but never fails it.
+    """
+
+    resolved_path: Path | None = None
+    """The link's resolved on-disk target path, set by
+    ``check_broken_paths`` (see :meth:`MarkdownPath.get_full_path
+    <markdown_checker.models.path.MarkdownPath.get_full_path>`); left unset
+    (``None``) by every other check, including URL checks.
+    """
+
+    def has_locale(self) -> bool:
+        """
+        Check if the link's path contains a locale segment, e.g. ``/en-us/``.
+
+        Only the path is searched (not the host, query or anchor), and a
+        candidate ``xx-yy`` segment only counts if it's a known locale, so
+        segments like ``/ci-cd/`` aren't reported.
+
+        Returns:
+            True if the link has a locale, False otherwise
+        """
+        path = urlparse(self.link).path
+        return any(
+            f"{match['language']}_{match['region']}".lower() in locale.locale_alias
+            for match in LOCALE_PATTERN.finditer(path)
+        )
+
+    def has_tracking(self, params: Sequence[str] = DEFAULT_TRACKING_PARAMS) -> bool:
+        """
+        Check if the link carries any of the tracking query parameters in
+        ``params`` (case-insensitive), e.g. ``?wt.mc_id=`` or ``&wt.mc_id=``.
+
+        Args:
+            params: Tracking parameter names; any one of them counts.
+
+        Returns:
+            True if the link has a tracking ID, False otherwise
+        """
+        if not params:
+            return False
+        return bool(tracking_pattern(tuple(params)).search(self.link))
+
+    def __str__(self) -> str:
+        return self.link
+
+    def __repr__(self) -> str:
+        return self.link

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/models/config.py

@@ -4,6 +4,8 @@ from typing import Literal
 
 import httpx2
 
+from markdown_checker.patterns import DEFAULT_TRACKING_PARAMS
+
 DEFAULT_HEADERS: dict[str, str] = {
     "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
     "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7)",
@@ -33,8 +35,10 @@ class Config:
     """Holds all runtime configuration for a check run."""
 
     skip_domains: list[str] = field(default_factory=list)
-    """Hostnames to exclude from URL checks (substring match against the
-    URL's hostname). Used by ``check_broken_urls`` and ``check_urls_locale``.
+    """Domains to exclude from URL checks. A domain matches the URL's
+    hostname exactly or as a parent domain (``example.com`` matches
+    ``www.example.com`` but not ``notexample.com``). Used by
+    ``check_broken_urls``, ``check_urls_locale`` and ``check_urls_tracking``.
     """
 
     skip_urls_containing: list[str] = field(default_factory=list)
@@ -43,9 +47,16 @@ class Config:
     """
 
     tracking_domains: list[str] = field(default_factory=list)
-    """Hostnames that must carry a ``wt.mc_id`` tracking parameter. Only
-    used by ``check_urls_tracking``; URLs on other hosts are ignored by
-    that check.
+    """Domains whose URLs must carry a tracking parameter (see
+    ``tracking_param``), matched like ``skip_domains`` (subdomains
+    included). Only used by ``check_urls_tracking``; URLs on other hosts are
+    ignored by that check.
+    """
+
+    tracking_param: list[str] = field(default_factory=lambda: list(DEFAULT_TRACKING_PARAMS))
+    """Tracking query parameter names a link must carry; any one of them
+    counts, matched case-insensitively. Used by ``check_urls_tracking`` and
+    ``check_paths_tracking``.
     """
 
     timeout: int = 20

{markdown_checker-1.2.2 → markdown_checker-1.3.0}/src/markdown_checker/models/path.py

@@ -1,8 +1,10 @@
 import os
 from dataclasses import dataclass
 from pathlib import Path
+from urllib.parse import unquote
 
 from markdown_checker.models.base import MarkdownLinkBase
+from markdown_checker.patterns import QUERY_OR_ANCHOR_PATTERN
 from markdown_checker.patterns import TRACKING_QUERY_PATTERN
 
 
@@ -26,46 +28,36 @@ class MarkdownPath(MarkdownLinkBase):
     def remove_fragments(self) -> str:
         """
         Strip the tracking query string and any trailing ``?query``/``#anchor``
-        fragment from the link, leaving just the file path portion.
+        fragment from the link, then percent-decode what's left (e.g.
+        ``my%20file.md`` becomes ``my file.md``), leaving just the file path.
+
+        As in any URL, the first ``?`` or ``#`` starts the fragment, so a
+        literal ``#`` or ``?`` in a file name must be written as ``%23``/``%3F``.
 
         Returns:
             The link with its fragment removed, as a string.
         """
         cleaned = TRACKING_QUERY_PATTERN.sub("", self.link)
-
-        # Find the last occurrence of the dot
-        dot_index = cleaned.rfind(".")
-
-        # If a dot is found, slice the string up to the end of the extension
-        if dot_index != -1:
-            # Find the next slash or fragment after the dot
-            slash_index = cleaned.find("/", dot_index)
-            fragment_index = cleaned.find("#", dot_index)
-
-            # Determine the earliest occurrence of either slash or fragment or query
-            end_index = min(
-                slash_index if slash_index != -1 else len(cleaned),
-                fragment_index if fragment_index != -1 else len(cleaned),
-            )
-            return cleaned[:end_index]
-        else:
-            return cleaned
+        return unquote(QUERY_OR_ANCHOR_PATTERN.sub("", cleaned))
 
     def get_full_path(self) -> Path:
         """
-        Resolve the link (with its fragment stripped) to an absolute path,
-        relative to the current working directory.
+        Resolve the link (with its fragment stripped) to an absolute path.
 
-        This is used for links that are themselves absolute (e.g.
-        ``/docs/usage.md``) or otherwise resolvable without knowing which
-        file the link appeared in. If resolution fails (e.g. on an unusual
-        path), falls back to :meth:`get_full_path_relative`.
+        Root-relative links (e.g. ``/docs/usage.md``) resolve against the
+        current working directory, which is treated as the repository root.
+        Other links resolve relative to the current working directory too;
+        if that fails (e.g. on an unusual path), this falls back to
+        :meth:`get_full_path_relative`.
 
         Returns:
             The resolved absolute path.
         """
+        cleaned = self.remove_fragments()
+        if cleaned.startswith("/"):
+            return Path(os.path.normpath(Path.cwd() / cleaned.lstrip("/")))
         try:
-            return self.path_without_fragments.resolve()
+            return Path(cleaned).resolve()
         except (FileNotFoundError, RuntimeError):
             return Path(self.get_full_path_relative())
 
@@ -85,39 +77,21 @@ class MarkdownPath(MarkdownLinkBase):
         """
         Check whether the link's target file exists on disk.
 
-        Tries resolution relative to the containing markdown file first
-        (see :meth:`get_full_path_relative`), then as an absolute/CWD-relative
-        path (see :meth:`get_full_path`). Used by ``check_broken_paths`` to
-        flag links whose target is missing.
+        Tries resolution relative to the containing markdown file first, then
+        relative to the current working directory. Root-relative links (e.g.
+        ``/docs/usage.md``) are tried the same way with the leading ``/``
+        dropped. Used by ``check_broken_paths`` to flag links whose target is
+        missing.
 
         Returns:
             True if the path exists, False otherwise
         """
-        # For paths starting with /, strip the leading / to allow relative resolution
-        link = self.link.lstrip("/") if self.link.startswith("/") else self.link
-        cleaned = TRACKING_QUERY_PATTERN.sub("", link)
-
-        # Strip fragments/anchors the same way remove_fragments() does
-        dot_index = cleaned.rfind(".")
-        if dot_index != -1:
-            slash_index = cleaned.find("/", dot_index)
-            fragment_index = cleaned.find("#", dot_index)
-            end_index = min(
-                slash_index if slash_index != -1 else len(cleaned),
-                fragment_index if fragment_index != -1 else len(cleaned),
-            )
-            cleaned = cleaned[:end_index]
-
-        # Check relative path resolution
+        cleaned = self.remove_fragments().lstrip("/")
+
         relative_path = os.path.normpath(os.path.join(os.path.dirname(self.file_path), cleaned))
         if os.path.exists(relative_path):
             return True
-        # Check absolute path resolution
         try:
-            if Path(cleaned).resolve().exists():
-                return True
+            return Path(cleaned).resolve().exists()
         except (FileNotFoundError, RuntimeError, OSError):
-            resolved = Path(relative_path)
-            if resolved.exists():
-                return True
-        return False
+            return False