CVS log for pkgsrc/converters/pandoc/Makefile

Up to [cvs.NetBSD.org] / pkgsrc / converters / pandoc

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

converters/pandoc: Turn the package into a meta package

Background: The upstream package, pandoc, historically contained the CLI
tool but it was split into two separate packages: pandoc and
pandoc-cli. When that happened I (pho@) renamed the original
converters/pandoc to converters/hs-pandoc-base to install the Pandoc
library, and replaced the contents of converters/pandoc with the upstream
pandoc-cli.

The intention was to make the change less surprising to our users. That is,
I didn't want the CLI tool to suddenly disappear from
converters/pandoc. But it complicated matters because the upstream
pandoc-cli used a distinct versioning scheme from pandoc. As a result I was
forced to use a weird versioning scheme for converters/pandoc, whose
PKGVERSION actually tracked the version of converters/hs-pandoc-base,
despite its contents being that of pandoc-cli. The difference in package
name, upstream "pandoc" vs. pkgsrc "hs-pandoc-base", was also confusing.

So I:

* renamed converters/hs-pandoc-base to converters/hs-pandoc, which tracks
  upstream "pandoc",
* created converters/pandoc-cli, which tracks upsream "pandoc-cli",
* and turned converters/pandoc into a singleton meta-package which depends
  only on converters/pandoc-cli

so that we can eliminate the weird versioning scheme and can still provide
the CLI tool (indirectly) from converters/pandoc.

Revbump all Haskell after updating lang/ghc96

converters/pandoc: Fix build with GHC 9.6

Bump Haskell packages after updating lang/ghc94

revbump all the Haskell packages after modifying the runtime system of lang/ghc94

converters/pandoc: Update to 3.0.1

The release note is too long to paste here. See
https://hackage.haskell.org/package/pandoc-3.0.1/changelog

As of 3.0, the pandoc package has been split into 3 packages:

  * pandoc:            Base library         (converters/hs-pandoc-base)
  * pandoc-server:     HTTP REST API        (converters/hs-pandoc-server)
  * pandoc-cli:        Command-line tool    (converters/pandoc; this package)
  * pandoc-lua-engine: Lua scripting engine (converters/hs-pandoc-lua-engine)

The original package "pandoc" no longer contains the pandoc executable, and
now it resides in "pandoc-cli". This sucks, as simply creating a new pkgsrc
package for the CLI means that converters/pandoc suddenly loses the
executable and will confuse existing users a lot. So I had no choice but to
replace the content of converters/pandoc with pandoc-cli, and synchronize
its version with converters/hs-pandoc-base.

Recursive bump for recently updated Haskell packages

*: bump for hs-vector-algorithms buildlink3.mk change

hs*: recursive bump for new dependencies needed

for hs-aeson, hs-vector

converters/pandoc: Link pandoc with static Haskell libraries

Bump all Haskell packages after enabling "split sections" in mk/haskell.mk

revbump after changing the default Haskell compiler

Update to pandoc-2.17.1.1

The list of changes since 2.13 is too long to paste here:
https://hackage.haskell.org/package/pandoc-2.17.1.1/changelog

Bump packages that depends on GHC

*: Bump PKGREVISION for ghc-9.0.1

Update to pandoc-2.13

The change log is too long to paste here. See
http://hackage.haskell.org/package/pandoc-2.13/changelog

Import pandoc-2.9.1.1 from wip

Pandoc is a Haskell library for converting from one markup format to
another, and a command-line tool that uses this library. It can read
several dialects of Markdown and (subsets of) HTML, reStructuredText,
LaTeX, DocBook, JATS, MediaWiki markup, DokuWiki markup, TWiki markup,
TikiWiki markup, Jira markup, Creole 1.0, Haddock markup, OPML, Emacs
Org-Mode, Emacs Muse, txt2tags, ipynb (Jupyter notebooks), Vimwiki,
Word Docx, ODT, EPUB, FictionBook2, roff man, and Textile, and it can
write Markdown, reStructuredText, XHTML, HTML 5, LaTeX, ConTeXt,
DocBook, JATS, OPML, TEI, OpenDocument, ODT, Word docx, PowerPoint
pptx, RTF, MediaWiki, DokuWiki, XWiki, ZimWiki, Textile, Jira, roff
man, roff ms, plain text, Emacs Org-Mode, AsciiDoc, Haddock markup,
EPUB (v2 and v3), ipynb, FictionBook2, InDesign ICML, Muse, LaTeX
beamer slides, and several kinds of HTML/JavaScript slide shows (S5,
Slidy, Slideous, DZSlides, reveal.js).

In contrast to most existing tools for converting Markdown to HTML,
pandoc has a modular design: it consists of a set of readers, which
parse text in a given format and produce a native representation of
the document, and a set of writers, which convert this native
representation into a target format. Thus, adding an input or output
format requires only adding a reader or writer.