https://github.com/doriantaylor/xslt-latex

Use XSLT (1.0, lol, so you can even do it in the browser) to turn webpages into opinionated but pretty LaTeX
https://github.com/doriantaylor/xslt-latex

Last synced: 24 days ago
JSON representation

Use XSLT (1.0, lol, so you can even do it in the browser) to turn webpages into opinionated but pretty LaTeX

• Host: GitHub
• URL: https://github.com/doriantaylor/xslt-latex
• Owner: doriantaylor
• License: apache-2.0
• Created: 2021-12-11T15:38:59.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2022-10-07T02:54:57.000Z (over 2 years ago)
• Last Synced: 2023-03-12T10:11:35.817Z (almost 2 years ago)
• Language: XSLT
• Size: 22.5 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# `latex.xsl`: You Love to See It

OK folks this is my highly-opinionated (X)HTML to LaTeX transform. It

is a work in progress, but the ultimate goal is to be able to turn

collections of Web pages into nice-looking PDFs.

## Getting Started

You will need to `make fetch` before you can do anything, because this

thing uses my [RDFa](https://github.com/doriantaylor/rdfa-xslt) and

[transclusion](https://github.com/doriantaylor/xslt-transclusion) XSLT

libraries, as well as [XSLTSL](http://xsltsl.sourceforge.net/).

## Formatting your (X)HTML

The RDFa and transclusion libraries both require the ``

tag to be set on all resources *to the URLs where they actually are*,

because XSLT otherwise has no way of knowing where anything is.

This transform also takes advantage of RDFa 1.1, and will resolve

certain embedded metadata:

* `rdf:type` for `\documentclass{}`

* `dct:creator` for `\author{}`

* `dct:issued`, `dct:created`, `dct:date` for `\date{}`

* `dct:abstract` for `\abstract{}`

The following RDF classes map to the following document types:

* `bibo:Book` and subclasses map to `\documentclass{book}`

* `bibo:Report`, `bibo:Manual`, `bibo:Specification`, `bibo:Standard`,

  all map to `\documentclass{report}`

* `bibo:PersonalCommunicationDocument` and subclasses map to

  `\documentclass{letter}`

* `bibo:Article` and everything else maps to `\documentclass{article}`

Section levels are handled automatically, with `book` and `report`

defaulting to `\chapter`, and everything else defaulting to

`\section`. You can set the default section to `\part` by making the

type a `bibo:Collection` or `bibo:MultiVolumeBook`.

> (I am not in love with that last mapping but needed some way to

> express it.)

## Running the Beast

You will have to run `xsltproc` like so, in order to resolve the

embedded URI references:

```bash

SGML_CATALOG_FILES=./catalog.xml xsltproc --catalogs latex.xsl whatever.xhtml

```

If you want to access the output in a browser, you can make the

beginning of your document look like this:

```xml

...

```

…assuming your server is set up to clip file extensions. If not,

you'll have to go in and add `.xsl` to everything.

## Copyright & License

©2018-2021 [Dorian Taylor](https://doriantaylor.com/)

This software is provided under

the [Apache License, 2.0](https://www.apache.org/licenses/LICENSE-2.0).