@@ -1 +1,87 @@
3+* Revisions for building with windows under cygwin:
4+ Cabal under windows produces 'pandoc.exe', and some of the scripts
7+* Consider making section headers block titles rather than blocks.
8+ Instead of: [Header 1 "My title", Block1, Block2, Block3],
9+ Section "My title" [Block1, Block2, Block3].
10+ This seems cleaner and would facilitate a docbook writer.
11+ It might also simplify the rst reader.
13+* pandoc's HTML output fails to validate completely (w3c).
14+ There are a few quirks:
15+ + HTML doesn't like the \> at the end of <meta tags.
16+ But if we remove them, we'll have trouble with S5 output,
17+ which seems to need the xhtml header?
18+ + There's also a problem with the email obfuscation scheme.
19+ <noscript> isn't allowed inside <p> blocks. <script> is
21+ - come up with another scheme, perhaps more like markdown.pl's
22+ - ignore the validation problems
25+* Consider adding support for acronyms.
26+ Perhaps like this: [AAAS]
27+ [AAAS]: "American association for the advancement of science"
28+ <acronym title="American association for the advancement
29+ of science">AAAS</acronym>
31+* Consider changing footnote syntax so that all footnotes in markdown
32+ are embedded (and automatic).^[Like this. Here's a footnote. It
33+ is parsed like a block, so you can have embedded code blocks:
37+ ] That was the end of the note. This means having block elements
38+ embedded in inline elements, which is possible.
39+ Advantage: Much easier to write. You don't have to pick a label,
40+ move down to type your note, move back up.
41+ Disadvantage: Perhaps slightly harder to read. (But HTML and LaTeX
42+ output will still be easy to read.)
44+* Consider scrapping most of the wrapper scripts in favor of having
45+ symlinks to pandoc. Modify pandoc so that it changes its defaults
46+ depending on the name of the calling program (getProgName).
47+ This would eliminate a lot of complexity and allow better handling
48+ of options (eliminating the need for a separation between wrapper
49+ and pandoc options, for example).
51+ If we do this, we should change option parsing in pandoc to allow
52+ options after arguments. This will preserve backward-compatibility
53+ with the present wrapper system. We'd also want to add an -o
54+ option to pandoc (output file). When -o foo is specified, pandoc
55+ should print "Created foo" to stderr on success (unless --quiet
58+ A disadvantage is that we'd lose iconv conversion. But maybe this
59+ isn't needed anymore; UTF-8 seems to be standard on most systems now.
61+ The tricky wrappers to replace are markdown2pdf and html2markdown.
65+ save working_directory
67+ if markdown2latex "$@" >tempdir/output 2>tempdir/logfile; then
68+ extract output-file from logfile (this will be foo.pdf)
69+ if output-file found:
70+ mv foo.pdf tempdir/foo.tex
72+ mv tempdir/output tempdir/foo.tex
74+ run pdflatex on foo.tex to produce foo.pdf
75+ mv foo.pdf working_directory/foo.pdf
77+ display logfile to inform user
81+ html2markdown: needs to run the HTML through tidy (mainly because
82+ pandoc's html parser requires closing tags, etc.) So we probably
83+ need something like the existing wrapper script here. roktas
84+ suggests perhaps keeping html2markdown simple and using a separate
85+ script, web2markdown. note: we also need iconv here, since web
86+ pages may not be in UTF8.