The new version of TeXMath can translate from its type system into
LaTeX. So instead of writing the LaTeX ourself, we write to the TeXMath
`Exp` type, and let TeXMath do the rest.
Moved `MediaBag` definition and functions from Shared:
`lookupMedia`, `mediaDirectory`, `insertMedia`, `extractMediaBag`.
Removed `emptyMediaBag`; use `mempty` instead, since `MediaBag`
is a Monoid.
This will make paragraphs styled with `Author`, `Title`, `Subtitle`,
`Date`, and `Abstract` into pandoc metavalues, rather than text. The
implementation only takes those elements from the beginning of the
document (ignoring empty paragraphs).
Multiple paragraphs in the `Author` style will be made into a metaList,
one paragraph per item. Hard linebreaks (shift-return) in the paragraph
will be maintained, and can be used for institution, email, etc.
If header anchors (bookmarks in a header paragraph) already have an
auto-id, which will happen if they're generated by pandoc, we don't want
to rename it twice, and thus end up with an unnecessary number at the
end. So we add a state value to check if we're in a header. If we are,
we don't rename the bookmark -- wait until we rename it in our header
handling.
This modifies the Docx type in the parser to avoid all the extra files
(Notes, numbering, etc). A reader monad keeps track of these, and applies
them at the end. The reader monad is stacked with ErrorT to enable better
error-handling than the old Maybes. (Note that the better error handling
isn't really there yet, but it is now possible.)
One long-term goal of these changes is to make it easier to write the Docx
type. This should make it easier to develop a standalone docx package in the
future.
This is a first stab at writing out equations in LaTeX based on
omml equations in Word. There are some glitches: unicode chars not known to
LaTeX are silently skipped, and functions (such as `\oiiint`) not in the
standard LaTeX packages are inserted, which can lead to pdf compilation
errors (depending, of course, on your preamble).
Adding, for example, `\usepackage[charter]{mathdesign}` to the preamble will
allow you to use most of the more esoteric functions.
This gets rid of `divAttrToContainers`: an internal convenience function
which had become pretty inconvenient. Rather than converting classes and
indentations to string lists and back, we deal with the `pPr` attribute
directly.
Here, when hanging indents are greater than or equal to left indents, we
don't set it to block quote. Such indents are frequently used in
academic bibliographies. (Thanks to Caleb McDaniel.)
Previously, a fresh state was created for the purpose of updating. In
the future, when there is more than one field in the state, this
obviously won't work.
This is a ReaderT State stack, which keeps track of some environment info, such
as the options and the docx doc. The state will come in handy in the future,
for a couple of planned features (rewriting the section anchors as auto_idents,
and hopefully smart-quoting).
The normalizing tests revealed a problem with unformatted spaces, brought about
by `spanTrim`. This fixes by not trimming the spaces out of spans until they
are in their final form.
There were some problems with the old str normalization. This fixes those
problems. Also, since it drills down on its own, it only needs to be
mapped over the blocks, not walked over the tree.
`normalize` from Text.Pandoc.Shared is more general. In tests, though,
it more than doubles the run time. `strNormalize` does less, but it does
what we need. This comment is added for future maintainability.
Use a function `stripSpaces`, instead of recursion. Makes it a bit
easier to read and mantain, and simplify normalizing DefinitionList,
which was left out the first time.
