Add new recursive argument to include directive (#208)

* 'recursive' option added * exclude directive now avoids processing the desired files all together * add recursive description in README * Add translations * Add test * Refactor * Revert change in order of execution of directives * Add recursive to global config in documentation * Bump version --------- Co-authored-by: Álvaro Mondéjar Rubio <mondejar1994@gmail.com>
mondeja · Jun 1, 2024 · 47f57fb · 47f57fb
1 parent eb65645
commit 47f57fb
Show file tree

Hide file tree

Showing 10 changed files with 106 additions and 49 deletions.
diff --git a/README.md b/README.md
@@ -59,6 +59,7 @@ plugins:
       heading_offset: 0
       start: <!--start-->
       end: <!--end-->
+      recursive: true
 ```
 
 #### `opening_tag` and `closing_tag`
@@ -245,6 +246,9 @@ Includes the content of a file or a group of files.
 - <a name="include_encoding" href="#include_encoding">#</a>
   **encoding** (_utf-8_): Specify the encoding of the included file.
   If not defined `utf-8` will be used.
+- <a name="recursive" href="#include_recursive">#</a>
+  **recursive** (_true_): When this option is disabled, included files are not
+  processed for recursive includes. Possible values are `true` and `false`.
 
 ##### Examples
 

diff --git a/locale/es/README.md b/locale/es/README.md
@@ -48,6 +48,7 @@ plugins:
       heading_offset: 0
       start: <!--start-->
       end: <!--end-->
+      recursive: true
 ```
 
 #### `opening_tag` y `closing_tag`
@@ -236,6 +237,10 @@ Los valores posibles son `true` y `false`.
 - <a name="include_encoding" href="#include_encoding">#</a> **encoding**
 (*utf-8*): Especifica la codificación del archivo incluído. Si no se define,
 se usará `utf-8`.
+- <a name="recursive" href="#include_recursive">#</a> **recursive** (*true*):
+Cuando esta opción está deshabilitada, los archivos incluidos no son
+procesados para incluir de forma recursiva. Los valores posibles son `true` y
+`false`.
 
 ##### Ejemplos
 

diff --git a/locale/es/README.md.po b/locale/es/README.md.po
@@ -406,3 +406,13 @@ msgstr ""
 
 msgid "[pypi-link]: https://pypi.org/project/mkdocs-include-markdown-plugin"
 msgstr "[pypi-link]: https://pypi.org/project/mkdocs-include-markdown-plugin"
+
+msgid ""
+"<a name=\"recursive\" href=\"#include_recursive\">#</a> **recursive** "
+"(*true*): When this option is disabled, included files are not processed for"
+" recursive includes. Possible values are `true` and `false`."
+msgstr ""
+"<a name=\"recursive\" href=\"#include_recursive\">#</a> **recursive** "
+"(*true*): Cuando esta opción está deshabilitada, los archivos incluidos no "
+"son procesados para incluir de forma recursiva. Los valores posibles son "
+"`true` y `false`."
diff --git a/locale/fr/README.md b/locale/fr/README.md
@@ -47,6 +47,7 @@ plugins:
       heading_offset: 0
       start: <!--start-->
       end: <!--end-->
+      recursive: true
 ```
 
 #### `opening_tag` et `closing_tag`
@@ -235,6 +236,9 @@ valeurs possibles sont `true` et `false`.
 - <a name="include_encoding" href="#include_encoding">#</a> **encoding**
 (*utf-8*): Spécifiez l'encodage du fichier inclus. S'il n'est pas défini,
 `utf-8` sera utilisé.
+- <a name="recursive" href="#include_recursive">#</a> **recursive** (*true*):
+Lorsque cette option est désactivée, les fichiers inclus ne sont pas traités
+pour des inclusions récursives. Les valeurs possibles sont `true` et `false`.
 
 ##### Exemples
 

diff --git a/locale/fr/README.md.po b/locale/fr/README.md.po
@@ -402,6 +402,15 @@ msgid ""
 msgstr ""
 "[Globs génériques Bash]: https://facelessuser.github.io/wcmatch/glob/#syntax"
 
-#, fuzzy
 msgid "[pypi-link]: https://pypi.org/project/mkdocs-include-markdown-plugin"
 msgstr "[pypi-link]: https://pypi.org/project/mkdocs-include-markdown-plugin"
+
+msgid ""
+"<a name=\"recursive\" href=\"#include_recursive\">#</a> **recursive** "
+"(*true*): When this option is disabled, included files are not processed for"
+" recursive includes. Possible values are `true` and `false`."
+msgstr ""
+"<a name=\"recursive\" href=\"#include_recursive\">#</a> **recursive** "
+"(*true*): Lorsque cette option est désactivée, les fichiers inclus ne sont "
+"pas traités pour des inclusions récursives. Les valeurs possibles sont "
+"`true` et `false`."
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "mkdocs-include-markdown-plugin"
-version = "6.0.7"
+version = "6.1.0"
 description = "Mkdocs Markdown includer plugin."
 readme = "README.md"
 license = "Apache-2.0"

diff --git a/src/mkdocs_include_markdown_plugin/config.py b/src/mkdocs_include_markdown_plugin/config.py
@@ -24,3 +24,4 @@ class PluginConfig(Config):  # noqa: D101
     end = Optional(MkType(str))
     exclude = ListOfItems(MkType(str), default=[])
     cache = MkType(int, default=0)
+    recursive = MkType(bool, default=True)
diff --git a/src/mkdocs_include_markdown_plugin/directive.py b/src/mkdocs_include_markdown_plugin/directive.py
@@ -34,6 +34,7 @@ class DirectiveBoolArgument:  # noqa: D101
             'comments': bool,
             'rewrite-relative-urls': bool,
             'heading-offset': int,
+            'recursive': bool,
             'start': str | None,
             'end': str | None,
         },
@@ -96,6 +97,7 @@ def str_arg(arg: str) -> re.Pattern[str]:
     'dedent': arg('dedent'),
     'trailing-newlines': arg('trailing-newlines'),
     'rewrite-relative-urls': arg('rewrite-relative-urls'),
+    'recursive': arg('recursive'),
 
     # int
     'heading-offset': arg('heading-offset'),

diff --git a/src/mkdocs_include_markdown_plugin/event.py b/src/mkdocs_include_markdown_plugin/event.py
@@ -103,6 +103,22 @@ def get_file_content(  # noqa: PLR0913, PLR0915
         http_cache: Cache | None = None,
 ) -> str:
     """Return the content of the file to include."""
+    settings_ignore_paths = []
+    if settings.exclude is not None:
+        for path in glob.glob(
+                [
+                    os.path.join(docs_dir, fp)
+                    if not os.path.isabs(fp)
+                    else fp for fp in settings.exclude
+                ],
+                flags=GLOB_FLAGS,
+                root_dir=docs_dir,
+        ):
+            if path not in settings_ignore_paths:
+                settings_ignore_paths.append(path)
+        if page_src_path in settings_ignore_paths:
+            return markdown
+
     def found_include_tag(  # noqa: PLR0912, PLR0915
             match: re.Match[str],
     ) -> str:
@@ -125,19 +141,7 @@ def found_include_tag(  # noqa: PLR0912, PLR0915
         arguments_string = match['arguments']
 
         exclude_match = ARGUMENT_REGEXES['exclude'].search(arguments_string)
-        ignore_paths = []
-        if settings.exclude:
-            ignore_paths.extend(
-                glob.glob(
-                    [
-                        os.path.join(docs_dir, fp)
-                        if not os.path.isabs(fp)
-                        else fp for fp in settings.exclude
-                    ],
-                    flags=GLOB_FLAGS,
-                    root_dir=docs_dir,
-                ),
-            )
+        ignore_paths = [*settings_ignore_paths]
         if exclude_match is not None:
             exclude_string = parse_string_argument(exclude_match)
             if exclude_string is None:
@@ -151,14 +155,13 @@ def found_include_tag(  # noqa: PLR0912, PLR0915
                     f' {file_lineno_message(page_src_path, docs_dir, lineno)}',
                 )
 
-            ignore_paths.extend(
-                resolve_file_paths_to_exclude(
+            for path in resolve_file_paths_to_exclude(
                     exclude_string,
                     page_src_path,
                     docs_dir,
-                ),
-            )
-        ignore_paths = list(set(ignore_paths))
+            ):
+                if path not in ignore_paths:
+                    ignore_paths.append(path)
 
         file_paths_to_include, is_url = resolve_file_paths_to_include(
             filename,
@@ -181,7 +184,8 @@ def found_include_tag(  # noqa: PLR0912, PLR0915
             files_watcher.included_files.extend(file_paths_to_include)
 
         bool_options, invalid_bool_args = parse_bool_options(
-            ['preserve-includer-indent', 'dedent', 'trailing-newlines'],
+            ['preserve-includer-indent', 'dedent',
+                'trailing-newlines', 'recursive'],
             defaults,
             arguments_string,
         )
@@ -264,16 +268,17 @@ def found_include_tag(  # noqa: PLR0912, PLR0915
                         expected_but_any_found[i] = False
 
             # nested includes
-            new_text_to_include = get_file_content(
-                new_text_to_include,
-                file_path,
-                docs_dir,
-                tags,
-                defaults,
-                settings,
-                files_watcher=files_watcher,
-                http_cache=http_cache,
-            )
+            if bool_options['recursive'].value:
+                new_text_to_include = get_file_content(
+                    new_text_to_include,
+                    file_path,
+                    docs_dir,
+                    tags,
+                    defaults,
+                    settings,
+                    files_watcher=files_watcher,
+                    http_cache=http_cache,
+                )
 
             # trailing newlines right stripping
             if not bool_options['trailing-newlines'].value:
@@ -349,19 +354,7 @@ def found_include_markdown_tag(  # noqa: PLR0912, PLR0915
         arguments_string = match['arguments']
 
         exclude_match = ARGUMENT_REGEXES['exclude'].search(arguments_string)
-        ignore_paths = []
-        if settings.exclude is not None:
-            ignore_paths.extend(
-                glob.glob(
-                    [
-                        os.path.join(docs_dir, fp)
-                        if not os.path.isabs(fp)
-                        else fp for fp in settings.exclude
-                    ],
-                    flags=GLOB_FLAGS,
-                    root_dir=docs_dir,
-                ),
-            )
+        ignore_paths = [*settings_ignore_paths]
         if exclude_match is not None:
             exclude_string = parse_string_argument(exclude_match)
             if exclude_string is None:
@@ -374,14 +367,13 @@ def found_include_markdown_tag(  # noqa: PLR0912, PLR0915
                     f' directive at'
                     f' {file_lineno_message(page_src_path, docs_dir, lineno)}',
                 )
-            ignore_paths.extend(
-                resolve_file_paths_to_exclude(
+            for path in resolve_file_paths_to_exclude(
                     exclude_string,
                     page_src_path,
                     docs_dir,
-                ),
-            )
-        ignore_paths = list(set(ignore_paths))
+            ):
+                if path not in ignore_paths:
+                    ignore_paths.append(path)
 
         file_paths_to_include, is_url = resolve_file_paths_to_include(
             filename,
@@ -667,6 +659,7 @@ def on_page_markdown(
             'comments': config.comments,
             'rewrite-relative-urls': config.rewrite_relative_urls,
             'heading-offset': config.heading_offset,
+            'recursive': config.recursive,
             'start': config.start,
             'end': config.end,
         },

diff --git a/tests/test_unit/test_nested_includes.py b/tests/test_unit/test_nested_includes.py
@@ -226,6 +226,32 @@
             ],
             id='start-end-not-found (second-level)',
         ),
+        # recursive inclusion disabled with `include` directive
+        pytest.param(
+            '''# Header
+
+{%
+  include "{filepath}"
+  comments=false
+  recursive=false
+%}''',
+            '''# Header 2
+
+{% include "{filepath}" %}
+''',
+            '''# Header 3
+
+This content must not be included.
+''',
+            '''# Header
+
+# Header 2
+
+{% include "{filepath}" %}
+''',
+            [],
+            id='include-recursive=false',
+        ),
     ),
 )
 def test_nested_include(
@@ -249,6 +275,9 @@ def test_nested_include(
     second_includer_content = second_includer_content.replace(
         '{filepath}', included_file.as_posix(),
     )
+    expected_result = expected_result.replace(
+        '{filepath}', included_file.as_posix(),
+    )
 
     first_includer_file.write_text(first_includer_content)
     second_includer_file.write_text(second_includer_content)