I would like to generate PDF from Markdown that look "like" when reading the markdown on github, that is, with a similar / same style.

https://github.com/trishume/syntect

RGBA PNG images are not visible in the output PDF, though they do take up space. PDF renderers used include PDF.js, qpdfviewer, apvlv, and whatever the reMarkable tablet uses.

Using imagemagick to remove the alpha channel makes the images show up:

convert -alpha 1 flamer-1.png flamer-1-alpha-removed.png

Example:

test.md

# `flamer-1.png`:

![](flamer-1.png)

# `flamer-1-alpha-removed.png`:

![](flamer-1-alpha-removed.png)

flamer-1.png:

flamer-1-alpha-removed.png

test.pdf

Ref https://www.makeareadme.com/

Seems like a useful feature to have, but CommonMark doesn't define any way to create page breaks. This stackoverflow post suggests using some html to do it:

<div style="page-break-after: always;"></div>

We could also try making extra long thematic breaks page breaks. I.E. five or more on a line = page break.

*****

Or two thematic breaks in a row:

***
***

I'm open to discussion on if and how explicit page breaks could be implemented.

Instead of putting notes at the bottom of the page, put the notes at the end of the PDF. This is easier to implement, and conforms more closely to how most markdown to html generators work.

Later this could be a style option, letting you choose between bottom of the page, side of the page, or at the end.

Basically, you would give it a number and it would append that many blank pages to the end.

My use case is taking notes on the remarkable tablet. It does not allow you to append pages to PDFs, so this is an easy workaround for that.

Perhaps this would also be useful for someone who is generating a book? Maybe just in a use case similar to mine. I don't know if dead tree publishers take PDFs, it care about having blank pages at the end.

It would be cool to render mdBooks directly as PDF :)

I think this is the appropriate link to that topic: rust-lang/mdBook#88

https://spec.commonmark.org/0.28/

One caveat is that mdproof will not implement all html, even though markdown is technically a superset of it. Some html will be recognized, but not all.

Required:

• Block Quotes
• Thematic Breaks
• Unordered Lists
• Ordered Lists
• Images
• Footnotes
  • Render links as footnotes (?)
• Entity references
  • NVM, these are handled by pulldown-cmark.

Good to have:

• Tables
• Checklists

Because right now it is unclear. Would be nice to have it compatible with mdBook one by the way: https://github.com/rust-lang-nursery/mdBook/blob/master/LICENSE

Having migrated away from TeX for most of my uni-related work to Markdown, I could not agree more with this: (from the Readme)

However, all of these methods pull in massive dependencies, and can be difficult to set up correctly.

I have had setup problems for all of this time, and I think I'm still not 100% done.

So, thank you for making it possible to work with MD in a simpler way!

Have a good weekend :)

tree -L 1     
.
├── Cargo.lock
├── Cargo.toml
├── Chinese.md
├── Chinese.pdf
├── images
├── README.md
├── src
├── target
├── test.md
└── test.pdf

thread 'main' panicked at 'byte index 7 is not a char boundary; it is inside '─' (bytes 6..9) of ├── src ', libcore/str/mod.rs:2111:5

First I just wanted to say this is a really cool project, I'll definitely be using this.

There should be a default ouput name where running mdproof foo.md should create foo.pdf rather than giving an error.

I tried running

RUST_BACKTRACE=1 cargo run -- example_documents/pagination.md

and I got

    Finished dev [unoptimized + debuginfo] target(s) in 3.06s
     Running `target/debug/mdproof example_documents/pagination.md`
thread 'main' panicked at 'attempted to leave type `linked_hash_map::Node<std::string::String, object::Object>` uninitialized, which is invalid', /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/core/src/mem/mod.rs:660:9
stack backtrace:
   0: rust_begin_unwind
             at /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/std/src/panicking.rs:515:5
   1: core::panicking::panic_fmt
             at /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/core/src/panicking.rs:92:14
   2: core::panicking::panic
             at /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/core/src/panicking.rs:50:5
   3: core::mem::uninitialized
             at /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/core/src/mem/mod.rs:660:9
   4: linked_hash_map::LinkedHashMap<K,V,S>::insert
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/linked-hash-map-0.3.0/src/lib.rs:203:52
   5: lopdf::object::Dictionary::set
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/lopdf-0.17.0/src/object.rs:280:3
   6: <lopdf::object::Dictionary as core::iter::traits::collect::FromIterator<(K,lopdf::object::Object)>>::from_iter
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/lopdf-0.17.0/src/object.rs:351:4
   7: printpdf::types::plugins::xmp::xmp_metadata::XmpMetadata::into_obj
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/printpdf-0.2.10/src/types/plugins/xmp/xmp_metadata.rs:72:30
   8: printpdf::types::pdf_metadata::PdfMetadata::into_obj
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/printpdf-0.2.10/src/types/pdf_metadata.rs:66:10
   9: printpdf::types::pdf_document::PdfDocumentReference::save
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/printpdf-0.2.10/src/types/pdf_document.rs:267:58
  10: mdproof::main::run
             at ./src/bin/mdproof.rs:67:5
  11: mdproof::main
             at /home/scott/.cargo/registry/src/github.com-1ecc6299db9ec823/quicli-0.3.1/src/main_macro.rs:45:29
  12: core::ops::function::FnOnce::call_once
             at /rustc/a178d0322ce20e33eac124758e837cbd80a6f633/library/core/src/ops/function.rs:227:5
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.