[Tex/LaTex] LaTeX to HTML: PolyTeXnic and PlasTeX

htmllatex2htmlpythontex4ht

I was stunned by the elegant appearance of the Tau manifesto page, for which the source is in LaTeX. The author used a home-made tool that he calls PolyTeXnic, but which he is not willing to release.

How difficult would it be to reproduce his result? I'm leaning towards using PlasTeX as the base tool.

Edit I was not referring to the contents of the Tau manifesto page 🙂 but rather to its general appearance: nice cross references, equations, figures, tables, cross-references to them, footnotes, bibliographical references, etc.
The maths is actually what worries me the least, thanks to MathJax. I was rather thinking on all the other aspects of the document, and whether PlasTeX might be up to the task.

Best Answer

Let me answer myself to myself: one of the most reliable ways to convert LaTeX into html is definitely TeX4ht, because it uses TeX as the underlying engine.

TeX4ht takes care of the cross references, the figures, footnotes, bibliography, and more. The math parts are transformed into MathML which is then beautifully rendered using MathJax.

Related Solutions

[Tex/LaTex] HTML package, useful to put htmlonly and latexonly environments, not found

As mentioned in comments, latex4html is not developed any more, so it's unlikely it will be possible to find this file (which is n any case very generically named and so hard to track down). You will be best looking at tex4ht as an alternative.

[Tex/LaTex] HTML to LaTex: How to use python and lxml to convert an html document to LaTeX with custom macros

I looked at many tools and settled on lxml with a recursive function to map html tags to LaTeX markup. It gives you a single place to easily define your mapping with Python. I believe I worked from an example in the book, Foundations of Python Network Programming

Here is a minimum working example in Python 2.7:

# convert html document to LaTeX
import lxml.html # http://lxml.de/lxmlhtml.html
from lxml import etree
from io import StringIO, BytesIO

def html2latex(el): # fill in this function to catch and convert html tags
    result = []
    if el.text:
        result.append(el.text)
    for sel in el:
        if False: # get info
            print('tag',sel.tag)
            print('text',sel.text)
            print('tail',sel.tail)
            print('attrib',sel.attrib)
        if sel.tag in ["h1"]:
            result.append('\hmchapter{%s}' % html2latex(sel))
        elif sel.tag in ["td", "table"]:
            result.append("<%s>" % sel.tag)
            result.append(html2latex(sel))
            result.append("</%s>" % sel.tag)
        elif sel.tag in ["span"]:  #
            for att in sel.attrib.keys():
                if att =='style':
                    if sel.attrib[att] == 'font-style:italic':
                        result.append(r'\textit{%s}' % (html2latex(sel)))
        else:
            result.append(html2latex(sel))
        if sel.tail:
            result.append(sel.tail)
    return "".join(result)

def main():
    # must be unicode or lxml parse crashes
    html = u'''
<!DOCTYPE html>
    <html>
  <head>
    <title></title>
  </head>
<body >
  <h1 class="hmchapter" data-hmvarbodychaptertitle = "My title">My title</h1>
  text <span style="font-style:italic">in a specific context</span> and more.
</body>
</html>
'''
    parser = etree.HTMLParser()
    tree   = etree.parse(StringIO(html), parser) # expects a file, use StringIO for string
    root = tree.getroot()
    latex = html2latex(root)
    print latex

if __name__ == '__main__':
    main()

which prints:

\hmchapter{My title}
text \textit{in a specific context} and more.

Best Answer

Related Solutions

[Tex/LaTex] HTML package, useful to put htmlonly and latexonly environments, not found

[Tex/LaTex] HTML to LaTex: How to use python and lxml to convert an html document to LaTeX with custom macros

Related Question