This handles row and colspans. Partially addresses #6344.
It also ensures that cells won't wrap text in places where it
wouldn't normally wrap, even if this means making the cells wider
than requested by the colspec. (Closes#9001. Closes 7641.)
Parameters are different, so this is a breaking [API change].
Markdown, RST, and Muse writers have been adjusted to
use the new `gridTable`.
Previously inline TeX was handled in a way that was different
from org's own export, and that could lead to information loss.
This was particularly noticeable for inline math environments
such as `equation`. Previously, an `equation` environment
starting at the beginning of a line would create a raw block,
splitting up the paragraph containing it (see #10836).
On the other hand, an `equation` environment not at the beginning
of a line would be turned into regular inline elements
representing the math. (This would cause the equation number to
go missing and in some cases degrade the math formatting.)
Now, we parse all of these as raw "latex" inlines, which will be
omitted when converting to formats other than LaTeX (and other
formats like pandoc's Markdown that allow raw LaTex).
Closes#10836.
Allows using `-t plain+four_space_rule` to emulate the output of pandoc
before #7172. This is useful for users that use pandoc e.g. to
autoformat commit messages.
This fixes a regression introduced by af57648.
We used to renumber the Relationships so they didn't conflict with
the set of fixed Relationships we imposed.
We are now preserving the ids from the reference doc's
document.xml.refs, so we shouldn't renumber them or references
introduced by the user (e.g. in a template) will fail.
Closes#10769.
Also, when falling back to a Div with class `figure` for a figure
that can't be represented any other way, include a Div with class
`caption` containing the caption.
Closes#10755.
...to determine what to do about `.pdfhref` macros in `ms`
output.
When no PDF engine is specified, we don't use the `.pdfhref`
macros at all. This gives better results for links in formats
other than PDF, since the link text would simply disappear if
it exists only in a `.pdfhref` macro.
When a PDF engine is specified, escape the argument of
`.pdfhref O` in a way that is appropriate.
Remove `groff` extension.
Text.Pandoc.Extensions: remove `Ext_groff` constructor.
See #10738. This revises the earlier commit
3adcb4bd80.
If `--pdf-engine` is specified or if a PDF is being produced,
we set the `pdf-engine` variable. This allows writers and
templates to behave differently depending on the PDF engine.
When `groff` is used as a PDF engine, the `groff` extension
to `ms` is automatically enabled.
Limitations:
- `groff` currently produces larger PDFs than `pdfroff`.
- With `groff`, a table of contents produced with
`--table-of-contents/--toc` will always be placed at the end of
the document.
- Certain characters (e.g. Greek characters) may be dropped in
the PDF outline.
Closes#10738.
When this extension is enabled, the `ms` writer won't
escape `.pdfhref O` macros in the way that is required
when `pdfroff` is used (octal-encoded UTF-16).
The extension is not enabled by default.
Text.Pandoc.Extensions: add `Ext_groff` constructor [API change].
We now use the plain writer to render these, so that Greek characters
etc. will show up properly.
Note: the ms output this produces works well with pdfroff but not
with `groff -Tpdf`.
The form `\[e aa]` works for groff but not for all roff's (not for
macOS man for instance). So instead we'll emit `e` followed by
an escape for a unicode combining accent.
See #10716.
This affects T.P.RoffChar, T.P.Writers.Roff,
and the Man and Ms writers.
That is, `\(xy` instead of `\[xy]`. This was the original AT&T troff
form and is the most widely supported. The bracketed form causes
problem for some tools, e.g. `makewhatis` on macOS.
Closes#10716.
In LaTeX's tabular environment, the tabular newline takes an optional
argument that we skip. But it only takes a single optional argument, and
any further square-bracketed text that follows shouldn't be skipped.
Fixes#7512, and also adds a test for the original problem raised in
that issue which was already fixed at some point.
John MacFarlane