Repository files navigation

Include extension for Python Markdown. It lets you include local or remote (downloadable) files into your markdown at arbitrary positions.

This project is motivated bymarkdown-include and provides the same functionalities with some extras.

Inclusion for local file is by default recursive and for remote file non-recursive. You can change this behavior through configuration.

You can include part of the file by slicing according to line/column number.

File/Downloaded contents are cached, i.e if you include same file multiple times in multiple places, they won't be read/downloaded more than once. This behavior can also be changed with configuration.

Circular inclusion by default raises an exception. You can change this behavior to include the affected files in non-recursive mode through configuration.

You should not use markdown-include along with this extension, choose either one, not both.

1. Simple:{! file_path_or_url !}
2. With explicit encoding:{! file_path_or_url | encoding !}
3. With recurs_state on:{!+ file_path_or_url !} or{!+ file_path_or_url | encoding !}. This makes the included file to be able to include other files. This is meaningful only when recursion is set toNone. If it is set toFalse, this explicit recurs_state defintion can not force recursion. This is a depth 1 recursion, so you can choose which one to recurs and which one to not.
4. With recurs_state off:{!- file_path_or_url !} or{!- file_path_or_url | encoding !}. This will force not to recurs even when recursion is set toTrue.
5. Applying indentation{!> file_path_or_url!}. This will apply the indentation found in the include line before the include for all the lines in the included file.
6. Escaped syntax: You can escape it to get the literal. For example,\{! file_path_or_url !} will give you literal{! file_path_or_url !} and\\\{! file_path_or_url !} will give you\{! file_path_or_url !}
7. File slice: You can slice a file by line and column number. The syntax is{! file_path [ln:l.c-l.c,l.c-l.c,...] !}. No spaces allowed inside file slice syntax[ln:l.c-l.c,l.c-l.c,]. See more detals inFile slicing section.

General syntax:{!recurs_state apply_indent file_path_or_url [ln:slice_syntax] | encoding !}

The spaces are not necessary. They are just to make it look nice :) . No spaces allowed between{! and recurs_state (+-). If apply indentation is specified then it must follow recurse_state immediately or the{! if recurse_state is not specified.

If you don't like the syntax you can change it through configuration.

There might be some complications with the syntax{!file!}, for example, conflict withmarkdown.extensions.attr_list which uses{:?something}. As the: is optional, a typical problem that occurs is this one:

A paragraph{!our syntax!}

would produce:

<psyntax_="syntax!"_our="!our">A paragraph</p>

If you really want to avoid this type of collision, find some character sequence that is not being used by any extension that you are using and use those character sequences to make up the syntax.

See the configuration section for details

Install from Pypi:

pip install mdx_include

text=r"""some text {! test1.md !} some more text {! test2.md | utf-8 !}Escaping will give you the exact literal \{! some_file !}If you escape, then the backslash will be removed.If you want the backslash too, then provide two more: \\\{! some_file !}"""md=markdown.Markdown(extensions=['mdx_include'])html=md.convert(text)print(html)

Example output:

(when test1.md contains a single line**This is test1.md** and test2.md contains**This is test2.md**)

<p>some text<strong>This is test1.md</strong> some more text<strong>This is test2.md</strong></p><p>Escaping will give you the exact literal {! some_file !}</p><p>If you escape, then the backslash will be removed.</p><p>If you want the backslash too, then provide two more: \{! some_file !}</p>

configs= {'mdx_include': {'base_path':'mdx_include/test/','encoding':'utf-8','allow_local':True,'allow_remote':True,'truncate_on_failure':False,'recurs_local':None,'recurs_remote':False,'syntax_left':r'\{!','syntax_right':r'!\}','syntax_delim':r'\|','syntax_recurs_on':'+','syntax_recurs_off':'-','syntax_apply_indent':r'\>',            },        }text=r"""some text {! some_file !} some more text {! some_more_file | utf-8 !}Escaping will give you the exact literal \{! some_file !}If you escape, then the backslash will be removed.If you want the backslash too, then provide two more: \\\{! some_file !}"""md=markdown.Markdown(extensions=['mdx_include'],extension_configs=configs)html=md.convert(text)print(html)

You can include part of the file from certain line/column number to certain line/column number.

The general file slice syntax is:[ln:l.c-l.c,l.c-l.c,...], where l is the line number and c is the column number. All indexes are inclusive.

Examples:

Multiple slicing can be done by adding more slice expressions with commas (s,). In this case, a separator (default is two newlines) is inserted between each slice. For example, with slice expression1-2,4-9, two newlines will be inserted between the lines 1-2 and 4-9.

More details on thercslice doc

The configuration gives you enough cache control, but that's not where it ends :). You can do manual cache cleaning instead of letting the extension handle it for itself. First turn the auto cache cleaning off by settingcontent_cache_clean_local and/orcontent_cache_clean_remote toFalse (this is default), then call the cache cleaning function manually on the markdown object whenever you want:

md.mdx_include_content_cache_clean_local()md.mdx_include_content_cache_clean_remote()

You can also get the internal cache dictionary and make inplace modification (e.g cleaning a specific cache for a specific file/URL, or even modify the cached content):

local_cache_dict=md.mdx_include_get_content_cache_local()remote_cache_dict=md.mdx_include_get_content_cache_remote()

Let's say, there are three files, A, B and C. A includes B, B includes C and C inclues A and we are doing recursive include.

If circular inclusion is not allowed in the config i.e ifallow_circular_inclusion isFalse (which is the default) then it will raise an exception.

Ifallow_circular_inclusion is set toTrue, then it will work like this:

1. A and B will be normally included
2. B includes C normally too
3. C includes A which is a circular inclusion (C>A>B>C>A>B>C...). Thus A will be included in non-recursive mode asallow_circular_inclusion is set toTrue i.e C will include A literally without parsing A anymore.

The following markdown:

Including a gist:    ```python{! https://gist.github.com/drgarcia1986/3cce1d134c3c3eeb01bd/raw/73951574d6b62a18b4c342235006ff89d299f879/django_hello.py !}```

will produce (with fenced code block enabled):

<p>Including a gist:</p><pre><codeclass="python"># -*- coding: utf-8 -*-# Settingsfrom django.conf import settingssettings.configure(    DEBUG=True,    SECRET_KEY='secretfoobar',    ROOT_URLCONF=__name__,    MIDDLEWARE_CLASSES=(        'django.middleware.common.CommonMiddleware',        'django.middleware.csrf.CsrfViewMiddleware',        'django.middleware.clickjacking.XFrameOptionsMiddleware',    ))# Viewsfrom django.http import HttpResponsefrom django.conf.urls import urldef index(request):    return HttpResponse('&lt;h1&gt;Hello Word&lt;/h1&gt;')# Routesurlpatterns = (    url(r'^$', index),)# RunServerif __name__ == '__main__':    from django.core.management import execute_from_command_line    import sys    execute_from_command_line(sys.argv)</code></pre>