Is there a way I can rename or alias latexmath in an AsciiDoc document?
In an ideal world, I'd like to set up an AsciiDoc such that $...$ is interpreted as LaTeX math, and
$$...$$ is interpreted as a block equation. In general, I'm just trying to reduce the number of characters involved in defining a math block since
where $c$ is the speed of light and $m_0$ is the rest mass
is significantly more readable (to someone who's used LaTeX for years) than
where latexmath:[$c$] is the speed of light and latexmath:[$m_0$] is the rest mass
The use case I have is that I'm writing technical documentation for upload to a GitLab repository. I'd like to be able to exploit GitLab's ability to automatically render AsciiDoc format files. However, these documents are math heavy, so I find the large numbers of latexmath:[...] blocks hard to read while editing.
latexmath is the name of the macro that handles parsing the LaTeX markup. If you don't specify latexmath, asciidoctor doesn't know to pass control to an alternate parser.
You could achieve what you're after by writing a pre-processing step that identifies contiguous blocks of LaTeX markup and wraps that markup in latexmath:[...]. The updated markup can then be processed by Asciidoctor like normal (assuming that your LaTeX markup identification is accurate). How you go about implementing that is up to you.
Another way, assuming you have some Ruby skills, would be to modify the extension that implements the latexmath macro such that it was called, say, L. Then your markup would be the more concise:
where L:[$c$] is the speed of light and L:[$m_0$] is the rest mass
Related
For example, when the goal is to represent several lines of a program's source code, then the AsciiDoc block markup choices are using
a literal block (....), which is converted to the literallayout DocBook tag
a listing block (----), which is converted to the screen DocBook tag
These are solid defaults, but the proper semantic markup would be programlisting in this case.
Using passthrough blocks is one solution, but at the cost of polymorphism (or convenience, if I decide to pepper my documents with ifdefs / ifndefs):
WARNING
Using passthroughs to pass content (without substitutions) can couple your content to a specific output format, such as HTML. In these cases, you should use conditional preprocessor directives to route passthrough content for different output formats based on the current backend.
I don't mind using the conditionals, but simply wondering if I am missing something obvious or straightforward? Such as passing semantic tag names to a block attrlist or something. Read through the entire AsciiDoc manual, the asciidoctor man page, the Generate DocBook from AsciiDoc article, and tried keyword searches, but couldn't find a thing. (It is highly probably though that I missed it..:)
I'm currently trying to customize the standard writer built into Pandoc to produce output in the ConTeXt format from Markdown input. Unfortunately, the documentation to create a custom writer found at the Pandoc website is not giving me too much information apart from how to write a custom HTML writer. So, I would like to ask for help with some fundamental ideas:
What would be the preferrable way to add some (probably) very simple functionality to the ConTeXt writer, e.g.: I would like to rewrite the sequence of characters " - " (in a Markdown document) as another sequence "~-- " (in the resulting ConTeXt document).
If I understood correctly, I'm supposed to base my custom writer on the standard (built-in) writers... But where can I find these? There doesn't seem to be anything in /usr/share/pandoc/(I'm working on Linux).
The website mentions the "classic style" and the "new style". Apart from one obviously being newer, what style am I supposed to use?
I know that these questions may sound rather simple, but there doesn't seem to be a lot of information available beyond the usual basic stuff. Any help would be much appreciated.
Pandoc has another feature that is similar to custom writers, called "Lua filters". Filters are quite likely a simpler and better choice in this case: They allow to modify the internal document representation. E.g.:
function Inlines (inlines)
for i=#inlines, 3, -1 do -- iterate backwards through the list of inlines
if inlines[i-2].t == 'Space' and inline[i-1] == pandoc.Str '-' and
inlines[i].t == 'Space' then
-- Replace elements with raw ConTeXt
inlines[i-2] = pandoc.RawInline('context', '~--')
inlines:remove(i)
inlines:remove(i-1)
end
end
return inlines
end
The above would be used b writing it to a file, and then passing that file to pandoc via --lua-filters FILENAME.lua
The documentation for Lua filters is also less sparse and hopefully more approachable than the docs for custom writers.
I have a legacy VB application that still has some life in it, and I am wanting to translate it to another language.
I plan to write a Ruby script, possibly utilising a parser, to extract all strings from the three million lines of source, replace them with constants, and move them to a string resource file that can be used to provide translations.
Is anyone aware of a script/library that could be used to intelligently extract the strings?
I'm not aware of any existing off-the-shelf tool that you could use. We created a tool like this at my work and it worked well. The FRM file format is quite simple (although only briefly documented). We wrote a tool that (1) extracted all strings from control definitions and (2) generated the code to reload them at runtime during Form_Load.
I have an application that needs to parse a subset of Markdown. I basically only want to support inline elements (bold, italic, links, etc), not block level elements (p, h1, h2, etc).
There are a lot of different libraries, so I need some help narrowing it down (and a code sample would be helpful). I started using RedCarpet until I realized that I can't specify which elements I want to parse.
What Ruby Markdown library can I use to achieve this?
I haven't found a library that allows you to specify on a granular level what parts of Markdown syntax are allowed. RDiscount has some configurability, however it doesn't take into account block level elements.
You could also give Sanitize a try (I know, parsing twice isn't exactly an ideal solution) and strip out the elements you don't want afterward.
I am using git-wiki for my personal note storage. It works very well, except that WikiWords are converted to links before the markdown parsing stage, using a regular expression. This messes up scores of things, for instance links that point to outside wiki pages, or block quotes (if I am quoting something, I do not want a WikiWord to be changed into a link).
Are there ruby-based Markdown parsers that understand WikiLinks?
The best parser around is the C-based one (upskirt/sundown), whose ruby iteration is red carpet:
https://github.com/tanoku/redcarpet
It is better for performance and security reasons.
For the wiki links, pre-process them before sending your text to the markdown parser.