d829c48 · typeable/pandoc

git.iko.soy/typeable/pandoc

Merged changes from branches/wrappers since r177.

Summary of main changes: + Added -o/--output and -d/--debug options to pandoc. + Modified pandoc to behave differently depending on the name of the program. For example, if the program name is 'html2latex', the default reader will be html and the default writer latex. + Removed most of the old wrappers, replacing them with symlinks to pandoc. + Rewrote markdown2pdf and created a new wrapper web2markdown, with the functionality of the old html2markdown script. These new scripts exploit pandoc's -d option to avoid having to do complex command-line parsing. + Revised man pages and documentation appropriately. git-svn-id: https://pandoc.googlecode.com/svn/trunk@279 788f1e2b-df1e-0410-8736-df70ead52e1b

commitd829c4820adbe7a7634f1c1d825d0d206512e6e7 authorfiddlosopher <fiddlosopher@788f1e2b-df1e-0410-8736-df70ead52e1b> dateFri 22 Dec 2006, 20:16 (20 years ago) parentcfaf0c1

27 files changed+710−796

compare with master

unifiedsplit

Makefile+14−10

@@ -23,15 +23,16 @@ EXECSBASE := $(shell sed -ne 's/^[Ee]xecutable:[[:space:]]*//p' $(CABAL).in)

2323#-------------------------------------------------------------------------------

2424# Install targets

2525#-------------------------------------------------------------------------------

26+WRAPPERS := web2markdown markdown2pdf

27+SYMLINKS := markdown2html markdown2latex markdown2s5 markdown2rst \

28+ markdown2rtf html2markdown latex2markdown rst2markdown

29+PROGS := $(EXECS) $(WRAPPERS)

2630# Add .exe extensions if we're running Windows/Cygwin.

2731EXTENSION := $(shell uname | tr '[:upper:]' '[:lower:]' | \

2832 sed -ne 's/^cygwin.*$$/\.exe/p')

2933EXECS := $(addsuffix $(EXTENSION),$(EXECSBASE))

3034# First entry in Cabal's executable stanza is the main executable.

3135MAIN := $(firstword $(EXECS))

32−WRAPPERS := html2markdown latex2markdown markdown2html \

33− markdown2latex markdown2pdf

34−PROGS := $(EXECS) $(WRAPPERS)

3536DOCS := README.html README BUGS

3637

3738#-------------------------------------------------------------------------------

@@ -92,6 +93,12 @@ all: build-program

9293templates: $(SRCDIR)/templates

9394 $(MAKE) -C $(SRCDIR)/templates

9495

96+.PHONY: symlinks

97+cleanup_files+=$(SYMLINKS)

98+symlinks: $(SYMLINKS)

99+$(SYMLINKS): $(MAIN)

100+ ln -sf ./$(MAIN) $@

101+

95102define generate-shell-script

96103echo "Generating $@..."; \

97104awk ' \

@@ -131,7 +138,7 @@ build: configure

131138 $(BUILDCMD) build

132139

133140.PHONY: build-exec

134−build-exec: $(PROGS)

141+build-exec: $(PROGS) $(SYMLINKS)

135142cleanup_files+=$(EXECS)

136143$(EXECS): build

137144 for f in $@; do \

@@ -191,8 +198,9 @@ install-exec: build-exec

191198 fi; \

192199 $(INSTALL_PROGRAM) $$f $(BINPATH)/; \

193200 done

201+ cd $(BINPATH); for f in $(SYMLINKS); do ln -sf $(MAIN) $$f; done

194202uninstall-exec:

195− -for f in $(notdir $(PROGS)); do rm -f $(BINPATH)/$$f; done

203+ -for f in $(notdir $(PROGS) $(SYMLINKS)); do rm -f $(BINPATH)/$$f; done ;

196204

197205# Program + user documents installation.

198206.PHONY: install-program uninstall-program

@@ -277,15 +285,11 @@ osx-dmg: ../$(osx_dmg_name)

277285 -rm -f $(osx_dmg_name)

278286 mv $(osx_udzo_name) ../$(osx_dmg_name)

279287

280−.PHONY: test test-markdown test-wrapper

288+.PHONY: test test-markdown

281289test: $(MAIN)

282290 @cd $(TESTDIR) && perl runtests.pl -s $(PWD)/$(MAIN)

283291test-markdown: $(MAIN)

284292 @cd $(TESTDIR)/MarkdownTest_1.0.3 && perl MarkdownTest.pl -s $(PWD)/$(MAIN) -tidy

285−cleanup_files+=testwrapper

286−test-wrappers: testwrapper

287− @echo "Running $<..."

288− @sh testwrapper

289293

290294# Stolen and slightly improved from a GPLed Makefile. Credits to John Meacham.

291295src_all:=$(shell find $(SRCDIR) -type f -name '*hs' | egrep -v '^\./(_darcs|lib|test)/')

README+154−97

@@ -20,7 +20,7 @@ or output format requires only adding a reader or writer.

2020[reStructuredText]: http://docutils.sourceforge.net/docs/ref/rst/introduction.html

2121[S5]: http://meyerweb.com/eric/tools/s5/

2222[HTML]: http://www.w3.org/TR/html40/

23−[LaTeX]: http://www.latex-project.org/

23+[LaTeX]: http://www.latex-project.org/

2424[RTF]: http://en.wikipedia.org/wiki/Rich_Text_Format

2525[Haskell]: http://www.haskell.org/

2626

@@ -30,9 +30,53 @@ any kind. (See COPYRIGHT for full copyright and warranty notices.)

3030Recai Oktaş (roktas at debian dot org) deserves credit for the build

3131system, the debian package, and the robust wrapper scripts.

3232

33−[GPL]: http://www.gnu.org/copyleft/gpl.html

33+[GPL]: http://www.gnu.org/copyleft/gpl.html "GNU General Public License"

3434

35−# Using Pandoc

35+Requirements

36+============

37+

38+The `pandoc` program itself does not depend on any external libraries

39+or programs. The convenience programs `markdown2html`, `markdown2latex`,

40+`markdown2rst`, `markdown2rtf`, `markdown2s5`, `html2markdown`,

41+`latex2markdown`, and `rst2markdown` are implemented as symbolic links to

42+`pandoc`.

43+

44+The wrapper script `web2markdown` requires

45+

46+ - `html2markdown` (included with Pandoc)

47+ - a POSIX-compliant shell (installed by default on all linux and unix

48+ systems, including Mac OS X, and in [Cygwin] for Windows),

49+ - `HTML Tidy`

50+ - `iconv` (for character encoding conversion). (If `iconv` is absent,

51+ `web2markdown` will still work, but it will treat everything as UTF-8.)

52+

53+[Cygwin]: http://www.cygwin.com/

54+[HTML Tidy]: http://tidy.sourceforge.net/

55+[`iconv`]: http://www.gnu.org/software/libiconv/

56+

57+The wrapper script `markdown2pdf` requires

58+

59+ - `markdown2latex` (included with Pandoc)

60+ - a POSIX-compliant shell

61+ - `pdflatex`, which should be part of any [LaTeX] distribution

62+ - the [unicode] and [fancyvrb] LaTeX packages, which are included

63+ in many LaTeX distributions. The [unicode] package allows LaTeX to

64+ process UTF-8 characters. [fancyvrb] allows code blocks and verbatim

65+ text to be used within footnotes. If your installation of LaTeX

66+ does not include these packages, you will get an error (complaining

67+ about missing `ucs.sty` or `fancyvrb.sty`) when you try to compile

68+ a LaTeX file produced by Pandoc, or when you use the `markdown2pdf`

69+ script (described below). If this happens, install the [unicode] and

70+ [fancyvrb] packages package from [CTAN]. (Get the zip file from CTAN

71+ and unpack it into `~/texmf/tex/latex/`. You may also need to run

72+ `mktexlsr` or `texhash` before the files can be found by TeX.)

73+

74+[CTAN]: http://www.ctan.org "Comprehensive TeX Archive Network"

75+[unicode]: http://www.ctan.org/tex-archive/macros/latex/contrib/unicode/

76+[fancyvrb]: http://www.ctan.org/tex-archive/macros/latex/contrib/fancyvrb/

77+

78+Using Pandoc

79+============

3680

3781If you run `pandoc` without arguments, it will accept input from

3882STDIN. If you run it with file names as arguments, it will take input

@@ -66,10 +110,14 @@ a subset of reStructuredText syntax. For example, it doesn't handle

66110tables, definition lists, option lists, or footnotes. It handles only the

67111constructs expressible in unextended markdown. But for simple documents

68112it should be adequate. The `latex` and `html` readers are also limited

69−in what they can do.

113+in what they can do. Because the `html` reader is picky about the HTML

114+it parses, it is recommended that you pipe HTML through [HTML Tidy] before

115+sending it to `pandoc`, or use the `web2markdown` script described below.

116+

117+By default, `pandoc` writes its output to STDOUT. If you want to

118+write to a file, use the `-o` option or shell redirection:

70119

71−`pandoc` writes its output to STDOUT. If you want to write to a file,

72−use redirection:

120+ pandoc -o hello.html hello.txt

73121

74122 pandoc hello.txt > hello.html

75123

@@ -77,13 +125,14 @@ Note that you can specify multiple input files on the command line.

77125`pandoc` will concatenate them all (with blank lines between them)

78126before parsing:

79127

80− pandoc -s chapter1.txt chapter2.txt chapter3.txt references.txt > book.html

128+ pandoc -s chapter1.txt chapter2.txt references.txt > book.html

81129

82130(The `-s` option here tells `pandoc` to produce a standalone HTML file,

83131with a proper header, rather than a fragment. For more details on this

84132and many other command-line options, see below.)

85133

86−# Character encodings

134+Character encodings

135+-------------------

87136

88137Unfortunately, due to limitations in GHC, `pandoc` does not automatically

89138detect the system's local character encoding. Hence, all input and

@@ -97,92 +146,65 @@ will convert `source.txt` from the local encoding to UTF-8, then

97146convert it to HTML, then convert back to the local encoding,

98147putting the output in `output.html`.

99148

100−[`iconv`]: http://www.gnu.org/software/libiconv/

101−

102149The shell scripts (described below) automatically convert the input

103150from the local encoding to UTF-8 before running them through `pandoc`,

104151then convert the output back to the local encoding.

105152

106−## LaTeX and UTF-8

107−

108−LaTeX sources produced by Pandoc use `ucs.sty`, which is included in many

109−LaTeX distributions. This allows LaTeX to process UTF-8 characters.

110−If your installation of LaTeX does not include `ucs.sty`, you will get an

111−error when you try to compile a LaTeX file produced by Pandoc, or when

112−you use the `markdown2pdf` script (described below). If this happens,

113−install the [unicode] package from [CTAN]. (Get the `unicode.zip`

114−file from CTAN, unpack it, and copy the whole `unicode` directory into

115−`~/texmf/tex/latex/`. You may also need to run `mktexlsr` or `texhash`

116−before the files can be found by TeX.)

153+Convenience programs and wrapper scripts

154+========================================

117155

118−[CTAN]: http://www.ctan.org

119−[unicode]: http://www.ctan.org/tex-archive/macros/latex/contrib/unicode/

156+For convenience, eight variant programs are included with Pandoc:

157+`markdown2html` (which is equivalent to `pandoc -w html`),

158+`markdown2latex` (equivalent to `pandoc -w latex`), `markdown2rst`

159+(equivalent to `pandoc -w rst`), `markdown2rtf` (equivalent to

160+`pandoc -w rtf`), `markdown2s5` (equivalent to `pandoc -w s5`),

161+`html2markdown` (equivalent to `pandoc -r html -w markdown`),

162+`latex2markdown` (equivalent to `pandoc -r latex -w markdown`), and

163+`rst2markdown` (equivalent to `pandoc -r rst -w markdown`). These

164+programs take an appropriately restricted subset of `pandoc`'s

165+options. (Run them with the `-h` flag for a full list of allowed

166+options.)

120167

121−# The shell scripts

168+Like `pandoc`, all of these programs produce fragments by default.

169+If you want to produce a standalone file, complete with a header

170+and footer appropriate to the format, use the `-s` option:

122171

123−Five shell scripts have been included that make it easy to run

124−`pandoc` without worrying about character encodings, and without

125−remembering all the command-line options:

172+ markdown2latex -s sample.txt > sample.tex

126173

127−- `markdown2html` converts markdown-formatted text to HTML

128−- `markdown2latex` converts markdown-formatted text to LaTeX

129−- `markdown2pdf` produces a PDF file from markdown-formatted

130− text, using `pdflatex`.

131−- `html2markdown` converts HTML to markdown-formatted text

132−- `latex2markdown` converts LaTeX to markdown-formatted text

174+Two shell scripts have also been included:

133175

134−All of the scripts use `iconv` (if available) to convert to and from

135−the local character encoding. All of the scripts presuppose that

136−`pandoc` is in the path, and some have additional requirements. (For

137−example, `html2markdown` uses `tidy`, and `markdown2pdf` uses

138−`pdflatex`.)

176+1. `markdown2pdf` produces a PDF file from markdown-formatted

177+ text, using `markdown2latex` and `pdflatex`. The default

178+ behavior of `markdown2pdf` is to create a file with the same

179+ base name as the first argument and the extension `pdf`; thus,

180+ for example,

139181

140−When no arguments are specified, text will be read from standard

141−input. Arguments specify input files (limited to one in the case of

142−`latex2markdown` and `html2markdown`; the other scripts accept any number

143−of arguments). `html2markdown` may take a URL as argument instead of

144−a filename; in this case, `curl`, `wget`, or an available text-based

145−browser will be used to fetch the contents of the URL. (The `-n` option

146−inhibits this behavior; the `-g` option allows the user to specify a

147−custom command that will be used to fetch from a URL.)

182+ markdown2pdf sample.txt endnotes.txt

148183

149−With the exception of `markdown2pdf`, the scripts write to standard output.

150−Output can be sent to a file using shell output redirection:

184+ will produce `sample.pdf`. (If `sample.pdf` exists already,

185+ it will be backed up before being overwritten.) An output file

186+ name can be specified explicitly using the `-o` option:

151187

152− latex2markdown sample.tex > sample.txt

188+ markdown2pdf -o "My Book.pdf" chap1.txt chap2.txt chap3.txt

153189

154−The default behavior of `markdown2pdf` is to create a file with the same

155−base name as the first argument and the extension `pdf`; thus, for example,

190+ If no input file is specified, input will be taken from STDIN.

156191

157− markdown2pdf sample.txt endnotes.txt

192+2. `web2markdown` grabs a web page from a file or URL and converts

193+ it to markdown-formatted text, using `tidy` and `html2markdown`.

194+ Unless input is from STDIN, an attempt is made to determine the

195+ character encoding of the page from the "Content-type" meta tag.

196+ If this is not present, UTF-8 is assumed. Alternatively, a character

197+ encoding may be specified explicitly using the `-e` option.

158198

159−will produce `sample.pdf`. (If `sample.pdf` exists already, it will be

160−backed up before being overwritten.) An output file name can be specified

161−explicitly using the `-o` option:

199+ `web2markdown` searches for an available program (`wget`, `curl`,

200+ or a text-mode browser) to fetch the contents of a URL.

201+ Optionally, the `-g` command may be used to specify the command

202+ to be used:

162203

163− markdown2pdf -o "My Book.pdf" chap1.txt chap2.txt chap3.txt

204+ web2markdown -g 'wget --user=foo --password=bar' mysite.com

164205

165−Options specific to the scripts, like `-o`, `-g`, and `-n`, must

166−be specified *before* any command-line arguments (file names or URLs).

167−Any options specified *after* the command-line arguments will be

168−passed directly to `pandoc`. For example,

169−

170− markdown2html tusks.txt -S -T Elephants

171−

172−will convert `tusks.txt` to `tusks.html` using smart quotes, ellipses,

173−and dashes, with "Elephants" as the page title prefix. (For a

174−complete list of `pandoc` options, see below.) When there are no

175−command-line arguments (because input is from STDIN), `pandoc`

176−options must be preceded by ` -- `:

177−

178− cat tusks.txt | markdown2html -- -S -T Elephants

179−

180−The ` -- ` separator may optionally be used when there are command-line

181−arguments:

182−

183− markdown2html -- tusks.txt -S -T Elephants

184−

185−# Command-line options

206+Command-line options

207+====================

186208

187209Various command-line options can be used to customize the output.

188210For a complete list, type

@@ -207,9 +229,11 @@ specified.)

207229complete with appropriate document headers. By default, `pandoc`

208230produces a fragment.

209231

210−`--custom-header` can be used to specify a custom document header. To

211−see the headers used by default, use the `-D` option: for example,

212−`pandoc -D html` prints the default HTML header.

232+`-o` or `--output-file` can be used to specify an output file.

233+

234+`-C` or `--custom-header` can be used to specify a custom document

235+header. To see the headers used by default, use the `-D` option:

236+for example, `pandoc -D html` prints the default HTML header.

213237

214238`-c` or `--css` allows the user to specify a custom stylesheet that

215239will be linked to in HTML and S5 output.

@@ -253,15 +277,38 @@ is for lists to be displayed all at once.

253277`-N` or `--number-sections` causes sections to be numbered in LaTeX

254278output. By default, sections are not numbered.

255279

256−# Pandoc's markdown vs. standard markdown

280+`-d` or `--debug` causes a debugging message to be written to STDERR.

281+The format of the message is as follows:

282+

283+ OUTPUT=foo

284+ INPUT=bar

285+ INPUT=Foo Baz

286+

287+Here `OUTPUT=` is followed by the name of the output file specified

288+using `-o`, if any. If no output file was specified, `OUTPUT=`

289+will appear with nothing following it. Lines beginning `INPUT=`

290+specify input files. If there are no input files, no `INPUT=` lines

291+will be printed. The `-d` option forces output to be written to

292+STDOUT, even if an output file was specified using the `-o` option.

293+(This option is provided to make it easier to write wrappers for

294+`pandoc`.)

295+

296+`-v` or `--version` prints the version number to STDERR.

297+

298+`-h` or `--help` prints a usage message to STDERR.

299+

300+Pandoc's markdown vs. standard markdown

301+=======================================

257302

258303In parsing markdown, Pandoc departs from and extends [standard markdown]

259304in a few respects. (To run Pandoc on the official

260305markdown test suite, type `make test-markdown`.)

261306

262307[standard markdown]: http://daringfireball.net/projects/markdown/syntax

308+ "Markdown syntax description"

263309

264−## Section Headings

310+Section Headings

311+----------------

265312

266313Pandoc creates an invisible anchor in front of every HTML section

267314heading. The ID of this anchor is derived from the section heading

@@ -281,7 +328,8 @@ example, just insert:

281328

282329 [Back to Aristotle](#Aristotle's_De_Anima)

283330

284−## Lists

331+Lists

332+-----

285333

286334Pandoc behaves differently from standard markdown on some "edge

287335cases" involving lists. Consider this source:

@@ -332,7 +380,8 @@ the example above:

332380 B) Fie

333381 C) Third

334382

335−## Literal quotes in titles

383+Literal quotes in titles

384+------------------------

336385

337386Standard markdown allows unescaped literal quotes in titles, as

338387in

@@ -343,7 +392,8 @@ Pandoc requires all quotes within titles to be escaped:

343392

344393 [foo]: "bar \"embedded\" baz"

345394

346−## Reference links

395+Reference links

396+---------------

347397

348398Pandoc allows implicit reference links in either of two styles:

349399

@@ -357,7 +407,8 @@ will appear as regular bracketed text. Note: even `[link][]` will

357407appear as `[link]` if there's no reference for `link`. If you want

358408`[link][]`, use a backslash escape: `\[link]\[]`.

359409

360−## Footnotes

410+Footnotes

411+---------

361412

362413Pandoc's markdown allows footnotes, using the following syntax:

363414

@@ -394,7 +445,8 @@ they cannot contain multiple paragraphs). The syntax is as follows:

394445

395446Inline and regular footnotes may be mixed freely.

396447

397−## Embedded HTML

448+Embedded HTML

449+-------------

398450

399451Pandoc treats embedded HTML in markdown a bit differently than

400452Markdown 1.0. While Markdown 1.0 leaves HTML blocks exactly as they

@@ -427,7 +479,8 @@ markdown with HTML block elements. For example, one can surround

427479a block of markdown text with `<div>` tags without preventing it

428480from being interpreted as markdown.

429481

430−## Title blocks

482+Title blocks

483+------------

431484

432485If the file begins with a title block

433486

@@ -460,7 +513,8 @@ If a title prefix is specified with `-T` and no title block appears

460513in the document, the title prefix will be used by itself as the

461514HTML title.

462515

463−## Box-style blockquotes

516+Box-style blockquotes

517+---------------------

464518

465519Pandoc supports emacs-style boxquote block quotes, in addition to

466520standard markdown (email-style) boxquotes:

@@ -469,7 +523,8 @@ standard markdown (email-style) boxquotes:

469523 | They look like this.

470524 `----

471525

472−## Inline LaTeX

526+Inline LaTeX

527+------------

473528

474529Anything between two $ characters will be parsed as LaTeX math. The

475530opening $ must have a character immediately to its right, while the

@@ -501,7 +556,8 @@ You can also use LaTeX environments. For example,

501556Note, however, that material between the begin and end tags will

502557be interpreted as raw LaTeX, not as markdown.

503558

504−## Custom headers

559+Custom headers

560+--------------

505561

506562When run with the "standalone" option (`-s`), `pandoc` creates a

507563standalone file, complete with an appropriate header. To see the

@@ -516,13 +572,14 @@ it and specify it on the command line as follows:

516572

517573 pandoc --header=MyHeaderFile

518574

519−# Producing S5 with Pandoc

575+Producing S5 with Pandoc

576+========================

520577

521−Producing an [S5] slide show with Pandoc is easy. A title page is

522−constructed automatically from the document's title block (see above).

523−Each section (with a level-one header) produces a single slide. (Note

524−that if the section is too big, the slide will not fit on the page; S5

525−is not smart enough to produce multiple pages.)

578+Producing an [S5] web-based slide show with Pandoc is easy. A title

579+page is constructed automatically from the document's title block (see

580+above). Each section (with a level-one header) produces a single slide.

581+(Note that if the section is too big, the slide will not fit on the page;

582+S5 is not smart enough to produce multiple pages.)

526583

527584Here's the markdown source for a simple slide show, `eating.txt`:

528585

debian/changelog+2

@@ -14,6 +14,8 @@ pandoc (0.22) unstable; urgency=low

1414

1515 * Refactored template processing (fillTemplates.pl).

1616

17+ * Modified wrapper scripts to make them more robust.

18+

1719 * Modified wrapper scripts to make them more robust and portable.

1820 To avoid code duplication and ensure consistency, wrappers are

1921 generated via a templating system from templates in src/wrappers.

man/man1/html2markdown.1+1−60

@@ -1,60 +1 @@

1−.TH HTML2MARKDOWN 1 "November 21, 2006" Pandoc "User Manuals"

2−.SH NAME

3−html2markdown \- converts HTML to markdown-formatted text

4−.SH SYNOPSIS

5−\fBhtml2markdown\fR [\fIoptions\fR] [\fIinput\-file\fR or \fIURL\fR]

6−[\fB\-\-\fR] [\fIpandoc\-opts\fR]

7−.SH DESCRIPTION

8−\fBhtml2markdown\fR converts \fIinput\-file\fR or \fIURL\fR (or text

9−from STDIN) from HTML to markdown\-formatted plain text.

10−If a URL is specified, \fBhtml2markdown\fR uses an available program

11−(e.g. wget, w3m, lynx or curl) to fetch its contents. Output is sent

12−to STDOUT.

13−.PP

14−\fBhtml2markdown\fR is a wrapper for \fBpandoc\fR.

15−.SH OPTIONS

16−.TP

17−.B \-h

18−Show usage message.

19−.TP

20−.B \-e \fIencoding\fR

21−Assume the character encoding \fIencoding\fR in reading the HTML.

22−(Note: \fIencoding\fR will be passed to \fBiconv\fR; a list of

23−available encodings may be obtained using `\fBiconv \-l\fR'.)

24−If the \fB\-e\fR option is not specified, the encoding will be

25−determined as follows: If input is from STDIN, the local encoding

26−will be assumed. Otherwise, \fBhtml2markdown\fR will try to

27−extract the character encoding from the "Content-type" meta tag.

28−If no character encoding is specified in this way, UTF-8 will be

29−assumed for a URL argument, and the local encoding will be assumed

30−for a file argument.

31−.TP

32−.B \-g \fIcommand\fR

33−Use \fIcommand\fR to fetch the contents of a URL. (By default,

34−\fBhtml2markdown\fR searches for an available program or text-based

35−browser to fetch the contents of a URL.) For example:

36−.IP

37−html2markdown \-g 'wget \-\-user=foo \-\-password=bar' mysite.com

38−.TP

39−.B \-n

40−Disable automatic fetching of contents when URLs are specified as

41−arguments.

42−.TP

43−.I pandoc\-opts

44−Any options appearing after \fIinput\-file\fR or \fIURL\fR on the

45−command line will be passed directly to \fBpandoc\fR. If no

46−\fIinput-file\fR or \fIURL\fR is specified, these options must

47−be preceded by ` \fB\-\-\fR '. (In other cases, ` \fB\-\-\fR ' is

48−optional.) See \fBpandoc\fR(1) for a list of options that may be used.

49−Example:

50−.IP

51−html2markdown input.txt \-\- \-R

52−.SH "SEE ALSO"

53−\fBpandoc\fR(1),

54−\fBmarkdown2html\fR(1),

55−\fBmarkdown2latex\fR(1),

56−\fBlatex2markdown\fR(1),

57−\fBmarkdown2pdf\fR(1),

58−\fBiconv\fR(1)

59−.SH AUTHOR

60−John MacFarlane and Recai Oktas

1+.so man1/pandoc.1

man/man1/latex2markdown.1+1−33

@@ -1,33 +1 @@

1−.TH LATEX2MARKDOWN 1 "November 21, 2006" Pandoc "User Manuals"

2−.SH NAME

3−latex2markdown \- converts LaTeX to markdown\-formatted text

4−.SH SYNOPSIS

5−\fBlatex2markdown\fR [\fIoptions\fR] [\fIinput\-file\fR]

6−[\fB\-\-\fR] [\fIpandoc\-opts\fR]

7−.SH DESCRIPTION

8−\fBlatex2markdown\fR converts \fIinput\-file\fR

9−(or text from STDIN) from LaTeX to markdown\-formatted plain text.

10−Output is sent to STDOUT.

11−.PP

12−\fBlatex2markdown\fR is a wrapper for \fBpandoc\fR.

13−.SH OPTIONS

14−.TP

15−.B \-h

16−Show usage message.

17−.TP

18−.I pandoc\-opts

19−Any options appearing after \fIinput\-file\fR on the command line

20−will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR

21−is specified, these options must be preceded by ` \fB\-\-\fR '.

22−(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1)

23−for a list of options that may be used. Example:

24−.IP

25−latex2markdown input.txt \-\- \-R

26−.SH "SEE ALSO"

27−\fBpandoc\fR(1),

28−\fBmarkdown2html\fR(1),

29−\fBhtml2markdown\fR(1),

30−\fBmarkdown2latex\fR(1),

31−\fBmarkdown2pdf\fR(1)

32−.SH AUTHOR

33−John MacFarlane and Recai Oktas

1+.so man1/pandoc.1

man/man1/markdown2html.1+1−34

@@ -1,34 +1 @@

1−.TH MARKDOWN2HTML 1 "November 21, 2006" Pandoc "User Manuals"

2−.SH NAME

3−markdown2html \- converts markdown\-formatted text to HTML

4−.SH SYNOPSIS

5−\fBmarkdown2html\fR [\fIoptions\fR] [\fIinput\-file\fR]...

6−[\fB\-\-\fR] [\fIpandoc\-opts\fR]

7−.SH DESCRIPTION

8−\fBmarkdown2html\fR converts \fIinput\-file\fR

9−(or text from STDIN) from markdown\-formatted plain text to HTML.

10−If multiple files are specified, they will be combined to make a single

11−HTML document. Output is sent to STDOUT.

12−.PP

13−\fBmarkdown2html\fR is a wrapper for \fBpandoc\fR.

14−.SH OPTIONS

15−.TP

16−.B \-h

17−Show usage message.

18−.TP

19−.I pandoc\-opts

20−Any options appearing after \fIinput\-file\fR... on the command line

21−will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR

22−is specified, these options must be preceded by ` \fB\-\-\fR '.

23−(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1)

24−for a list of options that may be used. Example:

25−.IP

26−markdown2html input.txt \-\- \-\-css=main.css \-S

27−.SH "SEE ALSO"

28−\fBpandoc\fR(1),

29−\fBhtml2markdown\fR(1),

30−\fBmarkdown2latex\fR(1),

31−\fBlatex2markdown\fR(1),

32−\fBmarkdown2pdf\fR(1)

33−.SH AUTHOR

34−John MacFarlane and Recai Oktas

1+.so man1/pandoc.1

man/man1/markdown2latex.1+1−34

@@ -1,34 +1 @@

1−.TH MARKDOWN2LATEX 1 "November 21, 2006" Pandoc "User Manuals"

2−.SH NAME

3−markdown2latex \- converts markdown-formatted text to LaTeX

4−.SH SYNOPSIS

5−\fBmarkdown2latex\fR [\fIoptions\fR] [\fIinput\-file\fR]...

6−[\fB\-\-\fR] [\fIpandoc\-opts\fR]

7−.SH DESCRIPTION

8−\fBmarkdown2latex\fR converts \fIinput\-file\fR (or text from STDIN)

9−from markdown\-formatted plain text to LaTeX. If multiple files are

10−specified, they will be combined to make a single LaTeX document.

11−Output is sent to STDOUT.

12−.PP

13−\fBmarkdown2latex\fR is a wrapper for \fBpandoc\fR.

14−.SH OPTIONS

15−.TP

16−.B \-h

17−Show usage message.

18−.TP

19−.I pandoc\-opts

20−Any options appearing after \fIinput\-file\fR... on the command line

21−will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR

22−is specified, these options must be preceded by ` \fB\-\-\fR '.

23−(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1)

24−for a list of options that may be used. Example:

25−.IP

26−markdown2latex input.txt \-\- \-\-custom\-header=letterhead.tex

27−.SH "SEE ALSO"

28−\fBpandoc\fR(1),

29−\fBmarkdown2html\fR(1),

30−\fBhtml2markdown\fR(1),

31−\fBlatex2markdown\fR(1),

32−\fBmarkdown2pdf\fR(1)

33−.SH AUTHOR

34−John MacFarlane and Recai Oktas

1+.so man1/pandoc.1

man/man1/markdown2pdf.1+57−29

@@ -1,43 +1,71 @@

1−.TH MARKDOWN2PDF 1 "November 21, 2006" Pandoc "User Manuals"

1+.TH MARKDOWN2PDF 1 "December 15, 2006" Pandoc "User Manuals"

22.SH NAME

33markdown2pdf \- converts markdown-formatted text to PDF, using pdflatex

44.SH SYNOPSIS

5−\fBmarkdown2pdf\fR [\fIoptions\fR] [\fB\-o\fR \fIoutput-file\fR]

6−[\fIinput-file\fR]... [\fB\-\-\fR] [\fIpandoc\-opts\fR]

5+\fBmarkdown2pdf\fR [\fIoptions\fR] [\fIinput-file\fR]...

76.SH DESCRIPTION

8−\fBmarkdown2pdf\fR converts \fIinput\-file\fR (or text from STDIN) from

9−markdown\-formatted plain text to PDF, using \fBpdflatex\fR. If no output

10−filename is specified, the name of the output file is derived from the

11−input file; thus, for example, if the input file is \fIhello.txt\fR,

12−the output file will be \fIhello.pdf\fR. If the input is read from STDIN

13−and no output filename is specified, the output file will be named

14−\fIstdin.pdf\fR. If multiple input files are specified, they will be

15−concatenated before conversion, and the name of the output file will be

16−derived from the first input file.

7+\fBmarkdown2pdf\fR converts \fIinput\-file\fR (or text from standard

8+input) from markdown\-formatted plain text to PDF, using \fBpdflatex\fR.

9+If no output filename is specified, the name of the output file is

10+derived from the input file; thus, for example, if the input file

11+is \fIhello.txt\fR, the output file will be \fIhello.pdf\fR. If

12+the input is read from STDIN and no output filename is

13+specified, the output file will be named \fIstdin.pdf\fR. If

14+multiple input files are specified, they will be concatenated before

15+conversion, and the name of the output file will be derived from

16+the first input file.

1717.PP

18−\fBmarkdown2pdf\fR is a wrapper for \fBpandoc\fR.

18+Input is assumed to be in the UTF\-8 character encoding. If your

19+local character encoding is not UTF\-8, you should pipe input and

20+output through \fBiconv\fR:

21+.IP

22+.B iconv \-t utf\-8 input.txt | pandoc | iconv \-f utf\-8

23+.PP

24+\fBmarkdown2pdf\fR assumes that the 'unicode' package

25+is in latex's search path. If this package is not included in your

26+latex setup, it can be obtained from <http://ctan.org>.

27+.PP

28+\fBmarkdown2pdf\fR is a wrapper around \fBmarkdown2latex\fR.

1929.SH OPTIONS

2030.TP

21−.B \-h

22−Show usage message.

31+.B \-o FILE, \-\-output=FILE

32+Write output to \fIFILE\fR.

2333.TP

24−.B \-o \fIoutput-file\fR

25−Specify name of output (PDF) file.

34+.B \-p, \-\-preserve-tabs

35+Preserve tabs instead of converting them to spaces.

2636.TP

27−.I pandoc\-opts

28−Any options appearing after \fIinput\-file\fR... on the command line

29−will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR

30−is specified, these options must be preceded by ` \fB\-\-\fR '.

31−(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1)

32−for a list of options that may be used. Example:

33−.IP

34−markdown2pdf input.txt \-\- \-\-custom\-header=letterhead.tex

37+.B \-\-tab-stop=\fITABSTOP\fB

38+Specify tab stop (default is 4).

39+.TP

40+.B \-R, \-\-parse-raw

41+Parse untranslatable LaTeX environments as raw LaTeX,

42+instead of ignoring them.

43+.TP

44+.B \-N, \-\-number-sections

45+Number section headings in LaTeX output. (Default is not to number them.)

46+.TP

47+.B \-H \fIFILE\fB, \-\-include-in-header=\fIFILE\fB

48+Include (LaTeX) contents of \fIFILE\fR at the end of the header. Implies

49+\fB\-s\fR.

50+.TP

51+.B \-B \fIFILE\fB, \-\-include-before-body=\fIFILE\fB

52+Include (LaTeX) contents of \fIFILE\fR at the beginning of the document body.

53+.TP

54+.B \-A \fIFILE\fB, \-\-include-after-body=\fIFILE\fB

55+Include (LaTeX) contents of \fIFILE\fR at the end of the document body.

56+.TP

57+.B \-C \fIFILE\fB, \-\-custom-header=\fIFILE\fB

58+Use contents of \fIFILE\fR

59+as the LaTeX document header (overriding the default header, which can be

60+printed using '\fBpandoc \-D latex\fR'). Implies \fB-s\fR.

61+.TP

62+.B \-v, \-\-version

63+Print version.

64+.TP

65+.B \-h, \-\-help

66+Show usage message.

3567.SH "SEE ALSO"

3668\fBpandoc\fR(1),

37−\fBmarkdown2html\fR(1),

38−\fBhtml2markdown\fR(1),

39−\fBmarkdown2latex\fR(1),

40−\fBlatex2markdown\fR(1),

4169\fBpdflatex\fR(1)

4270.SH AUTHOR

4371John MacFarlane and Recai Oktas

man/man1/markdown2rst.1new file+1

@@ -0,0 +1 @@

1+.so man1/pandoc.1

man/man1/markdown2rtf.1new file+1

@@ -0,0 +1 @@

1+.so man1/pandoc.1

man/man1/markdown2s5.1new file+1

@@ -0,0 +1 @@

1+.so man1/pandoc.1

man/man1/pandoc.1+75−51

@@ -1,18 +1,23 @@

1−.TH PANDOC 1 "November 21, 2006" Pandoc "User Manuals"

1+.TH PANDOC 1 "December 15, 2006" Pandoc "User Manuals"

22.SH NAME

3−pandoc \- general markup converter

3+pandoc, markdown2html, markdown2latex, markdown2rst, markdown2rtf,

4+markdown2s5, html2markdown2, latex2markdown, rst2markdown \- general

5+markup converter

46.SH SYNOPSIS

57\fBpandoc\fR [\fIoptions\fR] [\fIinput\-file\fR]...

68.SH DESCRIPTION

7−\fIPandoc\fR converts files from one markup format to another. It can

9+\fBPandoc\fR converts files from one markup format to another. It can

810read markdown and (subsets of) reStructuredText, HTML, and LaTeX, and

911it can write markdown, reStructuredText, HTML, LaTeX, RTF, and S5 HTML

1012slide shows.

1113.PP

12−If no \fIinput\-file\fR is specified, input is read from STDIN. Otherwise,

13−the \fIinput\-files\fR are concatenated (with a blank line between each)

14−and used as input. Output goes to STDOUT. If you want output to a file,

15−use shell redirection:

14+If no \fIinput\-file\fR is specified, input is read from STDIN.

15+Otherwise, the \fIinput\-files\fR are concatenated (with a blank

16+line between each) and used as input. Output goes to standard

17+output. If you want output to a file, use the \fB\-o\fR option or

18+shell redirection:

19+.IP

20+.B pandoc \-o output.html input.txt

1621.IP

1722.B pandoc input.txt > output.html

1823.PP

@@ -25,6 +30,19 @@ formats can be specified using command\-line options. For example,

2530converts \fIchap1.tex\fR from LaTeX to markdown\-formatted plain text.

2631See below for a detailed list of command\-line options.

2732.PP

33+For convenience, eight variant programs are available:

34+\fBmarkdown2html\fR (same as \fBpandoc \-w html\fR),

35+\fBmarkdown2latex\fR (same as \fBpandoc \-w latex\fR),

36+\fBmarkdown2rst\fR (same as \fBpandoc \-w rst\fR),

37+\fBmarkdown2rtf\fR (same as \fBpandoc \-w rtf\fR),

38+\fBmarkdown2s5\fR (same as \fBpandoc \-w s5\fR),

39+\fBhtml2markdown\fR (same as \fBpandoc \-r html \-w markdown\fR),

40+\fBlatex2markdown\fR (same as \fBpandoc \-r latex \-w markdown\fR),

41+and \fBrst2markdown\fR (same as \fBpandoc \-r rst \-w markdown\fR).

42+These programs take an appropriately restricted subset of \fBpandoc\fR's

43+options. (Run them with the \fB-h\fR flag for a full list of allowed

44+options.)

45+.PP

2846\fIPandoc\fR uses the UTF\-8 character encoding for both input and output.

2947If your local character encoding is not UTF\-8, you should pipe input

3048and output through \fBiconv\fR:

@@ -33,61 +51,58 @@ and output through \fBiconv\fR:

3351

3452.SH OPTIONS

3553.TP

36−.B \-v, \-\-version

37−Print version.

38−.TP

39−.B \-h, \-\-help

40−Show usage message.

41−.TP

42−.B \-f FORMAT, \-r FORMAT, \-\-from=FORMAT, \-\-read=FORMAT

54+.B \-f \fIFORMAT\fB, \-r \fIFORMAT\fB, \-\-from=\fIFORMAT\fB, \-\-read=\fIFORMAT\fB

4355Specify input format.

4456.I FORMAT

4557can be

46−.I native

58+.B native

4759(native Haskell),

48−.I markdown

60+.B markdown

4961(markdown or plain text),

50−.I rst

62+.B rst

5163(reStructuredText),

52−.I html

64+.B html

5365(HTML),

5466or

55−.I latex

67+.B latex

5668(LaTeX).

5769.TP

58−.B \-t FORMAT, \-w FORMAT, \-\-to=FORMAT, \-\-write=FORMAT

70+.B \-t \fIFORMAT\fB, \-w \fIFORMAT\fB, \-\-to=\fIFORMAT\fB, \-\-write=\fIFORMAT\fB

5971Specify output format.

6072.I FORMAT

6173can be

62−.I native

74+.B native

6375(native Haskell),

64−.I markdown

76+.B markdown

6577(markdown or plain text),

66−.I rst

78+.B rst

6779(reStructuredText),

68−.I html

80+.B html

6981(HTML),

70−.I latex

82+.B latex

7183(LaTeX),

72−.I s5

84+.B s5

7385(S5 HTML and javascript slide show),

7486or

75−.I rtf

87+.B rtf

7688(rich text format).

7789.TP

7890.B \-s, \-\-standalone

7991Produce output with an appropriate header and footer (e.g. a

8092standalone HTML, LaTeX, or RTF file, not a fragment).

8193.TP

94+.B \-o FILE, \-\-output=FILE

95+Write output to \fIFILE\fR instead of STDOUT.

96+.TP

8297.B \-p, \-\-preserve-tabs

8398Preserve tabs instead of converting them to spaces.

8499.TP

85−.B \-\-tab-stop=TABSTOP

100+.B \-\-tab-stop=\fITABSTOP\fB

86101Specify tab stop (default is 4).

87102.TP

88103.B \-R, \-\-parse-raw

89−Parse untranslatable HTML codes and LaTeX environments as raw HTML or

90−LaTeX, instead of ignoring them.

104+Parse untranslatable HTML codes and LaTeX environments as raw HTML

105+or LaTeX, instead of ignoring them.

91106.TP

92107.B \-S, \-\-smartypants

93108Use smart quotes, dashes, and ellipses in HTML output.

@@ -99,41 +114,50 @@ Use ASCIIMathML to display embedded LaTeX math in HTML output.

99114Make list items in S5 display incrementally (one by one).

100115.TP

101116.B \-N, \-\-number-sections

102−Number section headings in LaTeX output. (Default is not to number them.)

117+Number section headings in LaTeX output. (Default is not to number

118+them.)

103119.TP

104−.B \-c CSS, \-\-css=CSS

120+.B \-c \fICSS\fB, \-\-css=\fICSS\fB

105121Link to a CSS style sheet.

106122.I CSS

107123is the pathname of the style sheet.

108124.TP

109−.B \-H FILENAME, \-\-include-in-header=FILENAME

110−Include contents of \fIFILENAME\fR at the end of the header. Implies

125+.B \-H \fIFILE\fB, \-\-include-in-header=\fIFILE\fB

126+Include contents of \fIFILE\fR at the end of the header. Implies

111127\fB\-s\fR.

112128.TP

113−.B \-B FILENAME, \-\-include-before-body=FILENAME

114−Include contents of \fIFILENAME\fR at the beginning of the document body.

129+.B \-B \fIFILE\fB, \-\-include-before-body=\fIFILE\fB

130+Include contents of \fIFILE\fR at the beginning of the document

131+body.

115132.TP

116−.B \-A FILENAME, \-\-include-after-body=FILENAME

117−Include contents of \fIFILENAME\fR at the end of the document body.

133+.B \-A \fIFILE\fB, \-\-include-after-body=\fIFILE\fB

134+Include contents of \fIFILE\fR at the end of the document body.

118135.TP

119−.B \-\-custom-header=FILENAME

120−Use contents of \fIFILENAME\fR

121−as the document header (overriding the default header, which can be

122−printed by using the \fB\-D\fR option). Implies

123−\fB-s\fR.

136+.B \-C \fIFILE\fB, \-\-custom-header=\fIFILE\fB

137+Use contents of \fIFILE\fR as the document header (overriding the

138+default header, which can be printed by using the \fB\-D\fR option).

139+Implies \fB-s\fR.

124140.TP

125−.B \-D FORMAT, \-\-print-default-header=FORMAT

126−Print the default header for \fIFORMAT\fR

127−(\fIhtml, s5, latex, markdown, rst, rtf\fR).

141+.B \-D \fIFORMAT\fB, \-\-print-default-header=\fIFORMAT\fB

142+Print the default header for \fIFORMAT\fR (\fIhtml, s5, latex,

143+markdown, rst, rtf\fR).

128144.TP

129−.B \-T STRING, \-\-title-prefix=STRING

145+.B \-T \fISTRING\fB, \-\-title-prefix=\fISTRING\fB

130146Specify \fISTRING\fR as a prefix to the HTML window title.

147+.TP

148+.B \-d, \-\-debug

149+Print debugging information (names of input and output files) to

150+STDERR. Write output to STDOUT, even if an output file was specified

151+using the \fB\-o\fR option.

152+.TP

153+.B \-v, \-\-version

154+Print version.

155+.TP

156+.B \-h, \-\-help

157+Show usage message.

131158

132159.SH "SEE ALSO"

133−\fBmarkdown2html\fR(1),

134−\fBhtml2markdown\fR(1),

135−\fBmarkdown2latex\fR(1),

136−\fBlatex2markdown\fR(1),

160+\fBweb2markdown\fR(1),

137161\fBmarkdown2pdf\fR(1),

138162\fBiconv\fR(1)

139163

man/man1/rst2markdown.1new file+1

@@ -0,0 +1 @@

1+.so man1/pandoc.1

man/man1/web2markdown.1new file+82

@@ -0,0 +1,82 @@

1+.TH WEB2MARKDOWN 1 "December 15, 2006" Pandoc "User Manuals"

2+.SH NAME

3+web2markdown \- converts HTML to markdown-formatted text

4+.SH SYNOPSIS

5+\fBweb2markdown\fR [\fIoptions\fR] [\fIinput\-file\fR or \fIURL\fR]

6+.SH DESCRIPTION

7+\fBweb2markdown\fR converts \fIinput\-file\fR or \fIURL\fR (or text

8+from STDIN) from HTML to markdown\-formatted plain text.

9+If a URL is specified, \fBweb2markdown\fR uses an available program

10+(e.g. wget, w3m, lynx or curl) to fetch its contents. Output is sent

11+to STDOUT unless an output file is specified using the \fB\-o\fR

12+option.

13+.PP

14+\fBweb2markdown\fR uses the character encoding specified in the

15+"Content-type" meta tag. If this is not present, or if input comes

16+from STDIN, UTF-8 is assumed. A character encoding may be specified

17+explicitly using the \fB\-e\fR option.

18+.PP

19+\fBweb2markdown\fR is a wrapper for \fBhtml2markdown\fR.

20+.SH OPTIONS

21+.TP

22+.B \-s, \-\-standalone

23+Include title, author, and date information (if present) at the

24+top of markdown output.

25+.TP

26+.B \-o FILE, \-\-output=FILE

27+Write output to \fIFILE\fR instead of STDOUT.

28+.TP

29+.B \-p, \-\-preserve-tabs

30+Preserve tabs instead of converting them to spaces.

31+.TP

32+.B \-\-tab-stop=\fITABSTOP\fB

33+Specify tab stop (default is 4).

34+.TP

35+.B \-R, \-\-parse-raw

36+Parse untranslatable HTML codes as raw HTML.

37+.TP

38+.B \-H \fIFILE\fB, \-\-include-in-header=\fIFILE\fB

39+Include contents of \fIFILE\fR at the end of the header. Implies

40+\fB\-s\fR.

41+.TP

42+.B \-B \fIFILE\fB, \-\-include-before-body=\fIFILE\fB

43+Include contents of \fIFILE\fR at the beginning of the document body.

44+.TP

45+.B \-A \fIFILE\fB, \-\-include-after-body=\fIFILE\fB

46+Include contents of \fIFILE\fR at the end of the document body.

47+.TP

48+.B \-C \fIFILE\fB, \-\-custom-header=\fIFILE\fB

49+Use contents of \fIFILE\fR

50+as the document header (overriding the default header, which can be

51+printed using '\fBpandoc \-D markdown\fR'). Implies

52+\fB-s\fR.

53+.TP

54+.B \-v, \-\-version

55+Print version.

56+.TP

57+.B \-h, \-\-help

58+Show usage message.

59+.TP

60+.B \-e \fIencoding\fR

61+Assume the character encoding \fIencoding\fR in reading HTML.

62+(Note: \fIencoding\fR will be passed to \fBiconv\fR; a list of

63+available encodings may be obtained using `\fBiconv \-l\fR'.)

64+If the \fB\-e\fR option is not specified and input is not from

65+STDIN, \fBweb2markdown\fR will try to extract the character encoding

66+from the "Content-type" meta tag. If no character encoding is

67+specified in this way, or if input is from STDIN, UTF-8 will be

68+assumed.

69+.TP

70+.B \-g \fIcommand\fR

71+Use \fIcommand\fR to fetch the contents of a URL. (By default,

72+\fBweb2markdown\fR searches for an available program or text-based

73+browser to fetch the contents of a URL.) For example:

74+.IP

75+web2markdown \-g 'wget \-\-user=foo \-\-password=bar' mysite.com

76+

77+.SH "SEE ALSO"

78+\fBpandoc\fR(1),

79+\fBhtml2markdown\fR(1),

80+\fBiconv\fR(1)

81+.SH AUTHOR

82+John MacFarlane and Recai Oktas

src/Main.hs+116−26

@@ -45,6 +45,7 @@ import Text.Pandoc.Writers.DefaultHeaders ( defaultHtmlHeader,

4545 defaultRTFHeader, defaultS5Header, defaultLaTeXHeader )

4646import Text.Pandoc.Definition

4747import Text.Pandoc.Shared

48+import Text.Regex ( mkRegex, splitRegex )

4849import System ( exitWith, getArgs, getProgName )

4950import System.Exit

5051import System.Console.GetOpt

@@ -57,6 +58,9 @@ import Control.Monad ( (>>=) )

5758version :: String

5859version = "0.3"

5960

61+copyrightMessage :: String

62+copyrightMessage = "\nCopyright (C) 2006 John MacFarlane\nWeb: http://sophos.berkeley.edu/macfarlane/pandoc\nThis is free software; see the source for copying conditions. There is no\nwarranty, not even for merchantability or fitness for a particular purpose."

63+

6064-- | Association list of formats and readers.

6165readers :: [(String, ParserState -> String -> Pandoc)]

6266readers = [("native" , readPandoc)

@@ -101,10 +105,13 @@ data Opt = Opt

101105 , optCustomHeader :: String -- ^ Custom header to use, or "DEFAULT"

102106 , optDefaultHeader :: String -- ^ Default header

103107 , optTitlePrefix :: String -- ^ Optional prefix for HTML title

108+ , optOutputFile :: String -- ^ Name of output file

104109 , optNumberSections :: Bool -- ^ If @True@, number sections in LaTeX

105110 , optIncremental :: Bool -- ^ If @True@, incremental lists in S5

106111 , optSmart :: Bool -- ^ If @True@, use smart typography

107112 , optASCIIMathML :: Bool -- ^ If @True@, use ASCIIMathML in HTML

113+ , optShowUsage :: Bool -- ^ If @True@, show usage message

114+ , optDebug :: Bool -- ^ If @True@, output debug messages

108115 }

109116

110117-- | Defaults for command-line options.

@@ -123,32 +130,20 @@ startOpt = Opt

123130 , optCustomHeader = "DEFAULT"

124131 , optDefaultHeader = defaultHtmlHeader

125132 , optTitlePrefix = ""

133+ , optOutputFile = "" -- null for stdout

126134 , optNumberSections = False

127135 , optIncremental = False

128136 , optSmart = False

129137 , optASCIIMathML = False

138+ , optShowUsage = False

139+ , optDebug = False

130140 }

131141

132142-- | A list of functions, each transforming the options data structure in response

133143-- to a command-line option.

134−options :: [OptDescr (Opt -> IO Opt)]

135−options =

136− [ Option "v" ["version"]

137− (NoArg

138− (\_ -> do

139− hPutStrLn stderr ("Version " ++ version)

140− exitWith ExitSuccess))

141− "Print version"

142−

143− , Option "h" ["help"]

144− (NoArg

145− (\_ -> do

146− prg <- getProgName

147− hPutStrLn stderr (usageInfo (prg ++ " [OPTIONS] [FILES] - convert FILES from one markup format to another\nIf no OPTIONS specified, converts from markdown to html.\nIf no FILES specified, input is read from STDIN.\nOptions:") options)

148− exitWith ExitSuccess))

149− "Show help"

150−

151− , Option "fr" ["from","read"]

144+allOptions :: [OptDescr (Opt -> IO Opt)]

145+allOptions =

146+ [ Option "fr" ["from","read"]

152147 (ReqArg

153148 (\arg opt -> case (lookup (map toLower arg) readers) of

154149 Just reader -> return opt { optReader = reader }

@@ -172,6 +167,13 @@ options =

172167 (\opt -> return opt { optStandalone = True }))

173168 "Include needed header and footer on output"

174169

170+ , Option "o" ["output"]

171+ (ReqArg

172+ (\arg opt -> do

173+ return opt { optOutputFile = arg })

174+ "FILENAME")

175+ "Name of output file"

176+

175177 , Option "p" ["preserve-tabs"]

176178 (NoArg

177179 (\opt -> return opt { optPreserveTabs = True }))

@@ -241,7 +243,7 @@ options =

241243 "FILENAME")

242244 "File to include after document body"

243245

244− , Option "" ["custom-header"]

246+ , Option "C" ["custom-header"]

245247 (ReqArg

246248 (\arg opt -> do

247249 text <- readFile arg

@@ -263,18 +265,87 @@ options =

263265 let header = case (lookup arg writers) of

264266 Just (writer, head) -> head

265267 Nothing -> error ("Unknown reader: " ++ arg)

266− hPutStrLn stdout header

268+ hPutStr stdout header

267269 exitWith ExitSuccess)

268270 "FORMAT")

269271 "Print default header for FORMAT"

272+

273+ , Option "d" ["debug"]

274+ (NoArg

275+ (\opt -> return opt { optDebug = True }))

276+ "Print debug messages to stderr, output to stdout"

277+

278+ , Option "v" ["version"]

279+ (NoArg

280+ (\_ -> do

281+ prg <- getProgName

282+ hPutStrLn stderr (prg ++ " " ++ version ++

283+ copyrightMessage)

284+ exitWith $ ExitFailure 2))

285+ "Print version"

286+

287+ , Option "h" ["help"]

288+ (NoArg

289+ (\opt -> return opt { optShowUsage = True }))

290+ "Show help"

270291 ]

292+

293+-- parse name of calling program and return default reader and writer descriptions

294+parseProgName name =

295+ case (splitRegex (mkRegex "2") (map toLower name)) of

296+ [from, to] -> (from, to)

297+ _ -> ("markdown", "html")

298+

299+-- set default options based on reader and writer descriptions; start is starting options

300+setDefaultOpts from to start =

301+ case ((lookup from readers), (lookup to writers)) of

302+ (Just reader, Just (writer, header)) -> start {optReader = reader,

303+ optWriter = writer,

304+ optDefaultHeader = header}

305+ _ -> start

306+

307+-- True if single-letter option is in option list

308+inOptList :: [Char] -> OptDescr (Opt -> IO Opt) -> Bool

309+inOptList list desc =

310+ let (Option letters _ _ _) = desc in

311+ any (\x -> x `elem` list) letters

312+

313+-- Reformat usage message so it doesn't wrap illegibly

314+reformatUsageInfo = gsub " *--" " --" .

315+ gsub "(-[A-Za-z0-9]) *--" "\\1, --" .

316+ gsub " *([^- ])" "\n\t\\1"

317+

271318main = do

272319

320+ name <- getProgName

321+ let (from, to) = parseProgName name

322+

323+ let irrelevantOptions = if not ('2' `elem` name)

324+ then ""

325+ else "frtwD" ++

326+ (if (to /= "html" && to /= "s5") then "SmcT" else "") ++

327+ (if (to /= "latex") then "N" else "") ++

328+ (if (to /= "s5") then "i" else "") ++

329+ (if (from /= "html" && from /= "latex") then "R" else "")

330+

331+ let options = filter (not . inOptList irrelevantOptions) allOptions

332+

333+ let defaultOpts = setDefaultOpts from to startOpt

334+

273335 args <- getArgs

274− let (actions, sources, errors) = getOpt RequireOrder options args

336+ let (actions, sources, errors) = getOpt Permute options args

337+

338+ if (not (null errors))

339+ then do

340+ mapM (\e -> hPutStrLn stderr e) errors

341+ hPutStrLn stderr (reformatUsageInfo $

342+ usageInfo (name ++ " [OPTIONS] [FILES]") options)

343+ exitWith $ ExitFailure 2

344+ else

345+ return ()

275346

276347 -- thread option data structure through all supplied option actions

277− opts <- foldl (>>=) (return startOpt) actions

348+ opts <- foldl (>>=) (return defaultOpts) actions

278349

279350 let Opt { optPreserveTabs = preserveTabs

280351 , optTabStop = tabStop

@@ -289,12 +360,31 @@ main = do

289360 , optCustomHeader = customHeader

290361 , optDefaultHeader = defaultHeader

291362 , optTitlePrefix = titlePrefix

363+ , optOutputFile = outputFile

292364 , optNumberSections = numberSections

293365 , optIncremental = incremental

294366 , optSmart = smart

295367 , optASCIIMathML = asciiMathML

368+ , optShowUsage = showUsage

369+ , optDebug = debug

296370 } = opts

297371

372+ if showUsage

373+ then do

374+ hPutStr stderr (reformatUsageInfo $ usageInfo (name ++ " [OPTIONS] [FILES]") options)

375+ exitWith $ ExitFailure 2

376+ else return ()

377+

378+ output <- if ((null outputFile) || debug)

379+ then return stdout

380+ else openFile outputFile WriteMode

381+

382+ if debug

383+ then do

384+ hPutStrLn stderr ("OUTPUT=" ++ outputFile)

385+ hPutStr stderr $ concatMap (\s -> "INPUT=" ++ s ++ "\n") sources

386+ else return ()

387+

298388 let writingS5 = (defaultHeader == defaultS5Header)

299389 let tabFilter = if preserveTabs then id else (tabsToSpaces tabStop)

300390 let addBlank str = str ++ "\n\n"

@@ -323,13 +413,13 @@ main = do

323413 writerIncludeBefore = includeBefore,

324414 writerIncludeAfter = includeAfter }

325415

326− (readSources sources) >>= (putStr . encodeUTF8 . (writer writerOptions) .

416+ (readSources sources) >>= (hPutStr output . encodeUTF8 .

417+ (writer writerOptions) .

327418 (reader startParserState) . filter .

328− decodeUTF8 . (joinWithSep "\n"))

419+ decodeUTF8 . (joinWithSep "\n")) >> hClose output

329420

330421 where

331422 readSources [] = mapM readSource ["-"]

332423 readSources sources = mapM readSource sources

333− readSource "-" = getContents

424+ readSource "-" = getContents

334425 readSource source = readFile source

335−

src/wrappers/checkin.shdeleted−7

@@ -1,7 +0,0 @@

1−# Check if input files exist.

2−for f; do

3− if [ -n "$f" ] && ! [ -f "$f" ]; then

4− err "File '$f' not found."

5− exit 1

6− fi

7−done

src/wrappers/common.sh−27

@@ -8,22 +8,6 @@ WRAPPEE_ARGS=

88err () { echo "$*" | fold -s -w ${COLUMNS:-110} >&2; }

99errn () { printf "$*" | fold -s -w ${COLUMNS:-110} >&2; }

1010

11−usage () {

12− synopsis="$@"

13− err "Usage: $THIS $synopsis"

14− err "See $THIS(1) man file for details."

15−}

16−

17−runpandoc () {

18− if [ -n "$WRAPPEE_ARGS" ]; then

19− # Unpack arguments that will be passed to pandoc.

20− oldifs="$IFS"; IFS="$NEWLINE"; set -- $WRAPPEE_ARGS "$@"; IFS="$oldifs"

21− case "$1" in --) shift;; esac # tolerate the existence of a leading '--'

22− fi

23−

24− pandoc "$@"

25−}

26−

2711# Portable which(1).

2812pathfind () {

2913 oldifs="$IFS"; IFS=':'

@@ -37,17 +21,6 @@ pathfind () {

3721 return 1

3822}

3923

40−HAVE_ICONV=

41−if pathfind iconv; then

42− HAVE_ICONV=1

43− alias to_utf8='iconv -t utf-8'

44− alias from_utf8='iconv -f utf-8'

45−else

46− err "Warning: iconv not present. Assuming UTF-8 character encoding."

47− alias to_utf8='cat'

48− alias from_utf8='cat'

49−fi

50−

5124for p in pandoc $REQUIRED; do

5225 pathfind $p || {

5326 err "You need '$p' to use this program!"

src/wrappers/getopts.shdeleted−12

@@ -1,12 +0,0 @@

1−if [ -z "$SYNOPSIS" ]; then

2− SYNOPSIS="[-h] [input_file]"

3− [ -n "$THIS_NARG" ] || SYNOPSIS="${SYNOPSIS}..."

4−fi

5−

6−while getopts h opt; do

7− case $opt in

8− h|?) usage "$SYNOPSIS"; exit 2 ;;

9− esac

10−done

11−

12−shift $(($OPTIND - 1))

src/wrappers/html2markdown.indeleted−134

@@ -1,134 +0,0 @@

1−#!/bin/sh -e

2−# converts html to markdown

3−# uses an available program to fetch URL and tidy to normalize it first

4−

5−REQUIRED=tidy

6−

7−### common.sh

8−

9−grab_url_with () {

10− url="${1:?internal error: grab_url_with: url required}"

11−

12− shift

13− cmdline="$@"

14−

15− prog=

16− prog_opts=

17− if [ -n "$cmdline" ]; then

18− eval "set -- $cmdline"

19− prog=$1

20− shift

21− prog_opts="$@"

22− fi

23−

24− if [ -z "$prog" ]; then

25− # Locate a sensible web grabber (note the order).

26− for p in wget lynx w3m curl links w3c; do

27− if pathfind $p; then

28− prog=$p

29− break

30− fi

31− done

32−

33− [ -n "$prog" ] || {

34− errn "$THIS: Couldn't find a program to fetch the file from URL "

35− err "(e.g. wget, w3m, lynx, w3c, or curl)."

36− return 1

37− }

38− else

39− pathfind "$prog" || {

40− err "$THIS: No such web grabber '$prog' found; aborting."

41− return 1

42− }

43− fi

44−

45− # Setup proper base options for known grabbers.

46− base_opts=

47− case "$prog" in

48− wget) base_opts="-O-" ;;

49− lynx) base_opts="-source" ;;

50− w3m) base_opts="-dump_source" ;;

51− curl) base_opts="" ;;

52− links) base_opts="-source" ;;

53− w3c) base_opts="-n -get" ;;

54− *) err "$THIS: unhandled web grabber '$prog'; hope it succeeds."

55− esac

56−

57− err "$THIS: invoking '$prog $base_opts $prog_opts $url'..."

58− eval "set -- $base_opts $prog_opts"

59− $prog "$@" "$url"

60−}

61−

62−encoding=

63−grabber=

64−nograb=

65−while getopts e:g:nh opt; do

66− case $opt in

67− e) encoding="$OPTARG" ;;

68− g) grabber="$OPTARG" ;;

69− n) nograb=1 ;;

70− h|?)

71− usage "[-e encoding] [-g grabber_command] [-n] [-h] [input_file|url]"

72− exit 2 ;;

73− esac

74−done

75−

76−shift $(($OPTIND - 1))

77−

78−### postopts.sh

79−

80−### singlearg.sh

81−

82−inurl=

83−if [ -n "$1" ] && ! [ -f "$1" ]; then

84− if [ -n "$nograb" ]; then

85− err "'$1' not found; refusing to treat input as URL."

86− exit 1

87− fi

88− # Treat given argument as an URL.

89− inurl="$1"

90−fi

91−

92−if [ -n "$inurl" ]; then

93− err "Attempting to fetch file from '$inurl'..."

94−

95− ### tempdir.sh

96−

97− grabber_out=$THIS_TEMPDIR/grabber.out

98− grabber_log=$THIS_TEMPDIR/grabber.log

99− if ! grab_url_with "$inurl" "$grabber" 1>$grabber_out \

100− 2>$grabber_log; then

101− errn "grab_url_with failed"

102− if [ -f $grabber_log ]; then

103− err " with the following error log."

104− err

105− cat >&2 $grabber_log

106− else

107− err .

108− fi

109− exit 1

110− fi

111−

112− set -- $grabber_out

113−fi

114−

115−if [ -z "$encoding" ] && [ "x$@" != "x" ]; then

116− # Try to determine character encoding unless not specified

117− # and input is STDIN.

118− encoding=$(

119− head "$@" |

120− LC_ALL=C tr 'A-Z' 'a-z' |

121− sed -ne '/<meta .*content-type.*charset=/ {

122− s/.*charset=["'\'']*$[-a-zA-Z0-9]*$.*["'\'']*/\1/p

123− }'

124− )

125−fi

126−

127−if [ -n "$encoding" ] && [ -n "$HAVE_ICONV" ]; then

128− alias to_utf8='iconv -f "$encoding" -t utf-8'

129−elif [ -n "$inurl" ]; then # assume web pages are UTF-8

130− alias to_utf8='cat'

131−fi # else just use local encoding

132−

133−to_utf8 "$@" | tidy -utf8 2>/dev/null |

134−runpandoc -r html -w markdown -s | from_utf8

src/wrappers/latex2markdown.indeleted−14

@@ -1,14 +0,0 @@

1−#!/bin/sh -e

2−# runs pandoc to convert latex to markdown

3−

4−### common.sh

5−

6−### getopts.sh

7−

8−### postopts.sh

9−

10−### singlearg.sh

11−

12−### checkin.sh

13−

14−to_utf8 "$@" | runpandoc -r latex -w markdown -s | from_utf8

src/wrappers/markdown2html.indeleted−12

@@ -1,12 +0,0 @@

1−#!/bin/sh -e

2−# converts markdown to HTML

3−

4−### common.sh

5−

6−### getopts.sh

7−

8−### postopts.sh

9−

10−### checkin.sh

11−

12−to_utf8 "$@" | runpandoc | from_utf8

src/wrappers/markdown2latex.indeleted−12

@@ -1,12 +0,0 @@

1−#!/bin/sh -e

2−# converts markdown to latex

3−

4−### common.sh

5−

6−### getopts.sh

7−

8−### postopts.sh

9−

10−### checkin.sh

11−

12−to_utf8 "$@" | runpandoc -w latex -s | from_utf8

src/wrappers/markdown2pdf.in+29−39

@@ -1,64 +1,54 @@

11#!/bin/sh -e

2−# converts markdown to latex, then uses latex to make a PDF

4−REQUIRED=pdflatex

3+REQUIRED="markdown2latex pdflatex"

65### common.sh

8−outfile=

9−while getopts o:h opt; do

10− case $opt in

11− o) outfile="$OPTARG" ;;

12− h|?) usage "[-o output_file] [-h] [input_file]..."; exit 2 ;;

13− esac

14−done

15−

16−shift $(($OPTIND - 1))

17−

18−### postopts.sh

7+### tempdir.sh

198

20−### checkin.sh

9+texname=output

10+logfile=$THIS_TEMPDIR/log

2111

22−if [ -z "$outfile" ]; then

23− if [ -n "$1" ]; then

24− outfile="${1%.*}"

25− else

26− outfile="stdin" # input is STDIN, since no argument given

27− fi

12+if ! markdown2latex -s -d "$@" >$THIS_TEMPDIR/$texname.tex 2>$logfile; then

13+ [ -f $logfile ] && sed -e 's/markdown2latex/markdown2pdf/g' \

14+ -e '/^INPUT=/d' -e '/^OUTPUT=/d' $logfile >&2

15+ exit 1

2816fi

29−case "$outfile" in

30−*.*) ;; # skip appending extension if one is already present

31−*) outfile="${outfile%.*}.pdf";;

32−esac

3317

34−### tempdir.sh

35−

36−# We should use a filename without white spaces for pdflatex.

37−TEXNAME=$THIS

18+outfile="$(sed -ne 's/^OUTPUT=//p' $logfile)"

19+IFS="$NEWLINE"

20+set -- $(sed -ne 's/^INPUT=//p' $logfile)

21+firstinfilebase="${1%.*}"

22+defaultdest="${firstinfilebase:-stdin}.pdf"

23+destname="${outfile:-$defaultdest}"

3824

39−to_utf8 "$@" | runpandoc -w latex -s >$THIS_TEMPDIR/$TEXNAME.tex

4025(

4126 cd $THIS_TEMPDIR

42− if ! pdflatex -interaction=batchmode $TEXNAME.tex >/dev/null 2>&1; then

27+ if ! pdflatex -interaction=batchmode $texname.tex >/dev/null 2>&1; then

4328 err "LaTeX errors:"

44− from_utf8 $TEXNAME.log | sed -ne '/^!/,/^ *$/p' >&2

45− if grep -q "File \`ucs.sty' not found" $TEXNAME.log; then

46− err "Please install the 'unicode' package from ctan.org."

29+ sed -ne '/^!/,/^ *$/p' $texname.log >&2

30+ if grep -q "File \`ucs.sty' not found" $texname.log; then

31+ err "Please install the 'unicode' package from CTAN:"

32+ err "http://www.ctan.org/tex-archive/macros/latex/contrib/unicode/"

33+ fi

34+ if grep -q "File \`fancyvrb.sty' not found" $texname.log; then

35+ err "Please install the 'fancyvrb' package from CTAN:"

36+ err "http://www.ctan.org/tex-archive/macros/latex/contrib/fancyvrb/"

4737 fi

4838 exit 1

4939 fi

50−)

40+) || exit $?

5141

5242is_target_exists=

53−if [ -f "$outfile" ]; then

43+if [ -f "$destname" ]; then

5444 is_target_exists=1

55− mv -f "$outfile" "$outfile~"

45+ mv "$destname" "$destname~"

5646fi

5747

58−mv -f $THIS_TEMPDIR/$TEXNAME.pdf "$outfile"

48+mv -f $THIS_TEMPDIR/$texname.pdf "$destname"

5949

60−errn "Created '$outfile'"

50+errn "Created $destname"

6151[ -z "$is_target_exists" ] || {

62− errn " (previous file has been backed up as '$outfile~')"

52+ errn " (previous file has been backed up as $destname~)"

6353}

6454err .

src/wrappers/postopts.shdeleted−17

@@ -1,17 +0,0 @@

1−# Parse wrapper and wrappee (pandoc) arguments by taking

2−# into account that they may have space or tab characters.

3−pick="WRAPPER_ARGS"

4−while [ $# -gt 0 ]; do

5− if [ "$pick" = "WRAPPER_ARGS" ]; then

6− case "$1" in

7− -*) pick="WRAPPEE_ARGS" ;;

8− esac

9− fi

10− # Pack args with NEWLINE to preserve spaces,

11− # and put them into the picked variable.

12− eval "$pick=\"\$${pick}${NEWLINE}${1}\""

13− shift

14−done

15−

16−# Unpack filename arguments. Now "$@" will hold the filenames.

17−oldifs="$IFS"; IFS="$NEWLINE"; set -- $WRAPPER_ARGS; IFS="$oldifs"

src/wrappers/singlearg.shdeleted−7

@@ -1,7 +0,0 @@

1−# Ensure to work with a single argument.

2−if [ $# -gt 1 ]; then

3− first_arg="$1"

4− shift

5− err "Warning: extra arguments '$@' will be ignored."

6− set -- $first_arg

7−fi

src/wrappers/testwrapper.indeleted−141

@@ -1,141 +0,0 @@

1−#!/bin/sh

2−

3−THIS=$1

4−

5−ASH="ash -s"

6−BASH="bash --posix -s"

7−DASH="dash -s"

8−KSH="ksh -s"

9−POSH="posh -s"

10−ZSH="zsh -s"

11−

12−ERROR=""

13−

14−wrapper () {

15− $SH -- "$@" <<-'EOF'

16−### common.sh

17−

18−outfile=

19−while getopts o: opt; do

20− case $opt in

21− o) outfile="$OPTARG" ;;

22− esac

23−done

24−

25−shift $(($OPTIND - 1))

26−

27−### postopts.sh

28−

29−echo "Options passed to wrapper:"

30−[ -z "$outfile" ] || echo "|$outfile|"

31−

32−echo "Arguments passed to wrapper:"

33−for arg; do

34− echo "|$arg|"

35−done

36−

37−pandoc () {

38− echo "Arguments passed to wrappee:"

39− for arg; do

40− echo "|$arg|"

41− done

42−}

43−runpandoc

44−EOF

45−}

46−

47−# Portable which(1).

48−pathfind () {

49− oldifs="$IFS"; IFS=':'

50− for _p in $PATH; do

51− if [ -x "$_p/$*" ] && [ -f "$_p/$*" ]; then

52− IFS="$oldifs"

53− return 0

54− fi

55− done

56− IFS="$oldifs"

57− return 1

58−}

59−

60−check_results () {

61− if [ "$1" = "$2" ]; then

62− echo >&2 ok

63− return 0

64− else

65− echo >&2 failed

66− sed "s/^/\t/" >&2 <<EOF

67−Command line: '$3'

68−===> Expected:

69−$2

70−<=== Got:

71−$1

72−EOF

73− return 1

74− fi

75−}

76−

77−for SH in "$BASH" "$DASH" "$KSH" "$ZSH"; do

78− CMD=${SH%% *}

79− echo >&2 " Testing with $CMD..."

80− if pathfind "$CMD"; then

81− if [ "$CMD" = "zsh" ]; then

82− # Zsh needs to be called as 'sh' to enable POSIX mode.

83− ln -s $(which zsh) ./sh

84− SH="./sh ${SH#* }"

85− trap 'err=$?; rm -f ./sh; exit $err' 0 1 2 3 13 15

86− fi

87−

88− set -e

89−

90− # Test 1

91− printf >&2 " test case 1... "

92− actual=$(wrapper -o "output file" "foo bar" -A "quux baz" -B)

93− expected=$(cat <<'EOF'

94−Options passed to wrapper:

95−|output file|

96−Arguments passed to wrapper:

97−|foo bar|

98−Arguments passed to wrappee:

99−|-A|

100−|quux baz|

101−|-B|

102−EOF

103−)

104− check_results "$actual" "$expected" \

105− 'wrapper -o "output file" "foo bar" -A "quux baz" -B'

106−

107− # Test 2

108− printf >&2 " test case 2... "

109− actual=$(wrapper -- -A "foo bar")

110− expected=$(cat <<'EOF'

111−Options passed to wrapper:

112−Arguments passed to wrapper:

113−Arguments passed to wrappee:

114−|-A|

115−|foo bar|

116−EOF

117−)

118− check_results "$actual" "$expected" 'wrapper -- -A "foo bar"'

119−

120− # Test 3 (Test 1 with a redundant '--')

121− printf >&2 " test case 3... "

122− actual=$(wrapper -o "output file" "foo bar" -- -A "quux baz" -B)

123− expected=$(cat <<'EOF'

124−Options passed to wrapper:

125−|output file|

126−Arguments passed to wrapper:

127−|foo bar|

128−Arguments passed to wrappee:

129−|-A|

130−|quux baz|

131−|-B|

132−EOF

133−)

134− check_results "$actual" "$expected" \

135− 'wrapper -o "output file" "foo bar" -- -A "quux baz" -B'

136− else

137− echo >&2 "Warning: cannot verify correctness with $CMD; shell not available"

138− fi

139−done

140−

141−exit 0

src/wrappers/web2markdown.innew file+173

@@ -0,0 +1,173 @@

1+#!/bin/sh -e

2+# converts HTML from a URL, file, or stdin to markdown

3+# uses an available program to fetch URL and tidy to normalize it first

5+REQUIRED="tidy html2markdown"

7+### common.sh

9+grab_url_with () {

10+ url="${1:?internal error: grab_url_with: url required}"

11+

12+ shift

13+ cmdline="$@"

14+

15+ prog=

16+ prog_opts=

17+ if [ -n "$cmdline" ]; then

18+ eval "set -- $cmdline"

19+ prog=$1

20+ shift

21+ prog_opts="$@"

22+ fi

23+

24+ if [ -z "$prog" ]; then

25+ # Locate a sensible web grabber (note the order).

26+ for p in wget lynx w3m curl links w3c; do

27+ if pathfind $p; then

28+ prog=$p

29+ break

30+ fi

31+ done

32+

33+ [ -n "$prog" ] || {

34+ errn "$THIS: Couldn't find a program to fetch the file from URL "

35+ err "(e.g. wget, w3m, lynx, w3c, or curl)."

36+ return 1

37+ }

38+ else

39+ pathfind "$prog" || {

40+ err "$THIS: No such web grabber '$prog' found; aborting."

41+ return 1

42+ }

43+ fi

44+

45+ # Setup proper base options for known grabbers.

46+ base_opts=

47+ case "$prog" in

48+ wget) base_opts="-O-" ;;

49+ lynx) base_opts="-source" ;;

50+ w3m) base_opts="-dump_source" ;;

51+ curl) base_opts="" ;;

52+ links) base_opts="-source" ;;

53+ w3c) base_opts="-n -get" ;;

54+ *) err "$THIS: unhandled web grabber '$prog'; hope it succeeds."

55+ esac

56+

57+ err "$THIS: invoking '$prog $base_opts $prog_opts $url'..."

58+ eval "set -- $base_opts $prog_opts"

59+ $prog "$@" "$url"

60+}

61+

62+add_option () {

63+ options="$options$NEWLINE$1"

64+}

65+

66+options=

67+argument=

68+encoding=

69+grabber=

70+

71+# Parse command-line arguments

72+while [ $# -gt 0 ]; do

73+ case "$1" in

74+ -h|--help)

75+ html2markdown -h 2>&1 | sed -e 's/html2markdown/web2markdown/' 1>&2

76+ err " -e ENCODING, --encoding=ENCODING"

77+ err " Specify character encoding of input"

78+ err " -g COMMAND, --grabber=COMMAND"

79+ err " Specify command to be used to grab contents of URL"

80+ exit 0 ;;

81+ -v|--version)

82+ html2markdown -v

83+ exit 0 ;;

84+ -e)

85+ shift

86+ encoding=$1 ;;

87+ --encoding=*)

88+ wholeopt=$1

89+ # extract encoding from after =

90+ encoding=${wholeopt#*=} ;;

91+ -g)

92+ shift

93+ grabber=$1 ;;

94+ --grabber=*)

95+ wholeopt=$1

96+ # extract encoding from after =

97+ grabber=${wholeopt#*=} ;;

98+ -o|--output|-b|--tab-stop|-H|--include-in-header| \

99+ -A|--include-after-body|-C|-B|--include-before-body| \

100+ -C|--custom-header|-T|--title-prefix)

101+ add_option $1

102+ shift

103+ add_option $1 ;;

104+ -*) add_option $1 ;;

105+ *)

106+ if [ -z "$argument" ]; then

107+ argument=$1

108+ else

109+ err "Warning: extra argument '$1' will be ignored."

110+ fi ;;

111+ esac

112+ shift

113+done

114+

115+# Unpack options. Now "$@" will hold the html2markdown options.

116+oldifs="$IFS"; IFS="$NEWLINE"; set -- $options; IFS="$oldifs"

117+

118+inurl=

119+if [ -n "$argument" ] && ! [ -f "$argument" ]; then

120+ # Treat given argument as an URL.

121+ inurl="$argument"

122+fi

123+

124+if [ -n "$inurl" ]; then

125+ err "Attempting to fetch file from '$inurl'..."

126+

127+ ### tempdir.sh

128+

129+ grabber_out=$THIS_TEMPDIR/grabber.out

130+ grabber_log=$THIS_TEMPDIR/grabber.log

131+ if ! grab_url_with "$inurl" "$grabber" 1>$grabber_out 2>$grabber_log; then

132+ errn "grab_url_with failed"

133+ if [ -f $grabber_log ]; then

134+ err " with the following error log."

135+ err

136+ cat >&2 $grabber_log

137+ else

138+ err .

139+ fi

140+ exit 1

141+ fi

142+

143+ argument="$grabber_out"

144+fi

145+

146+if [ -z "$encoding" ] && [ "x$argument" != "x" ]; then

147+ # Try to determine character encoding if not specified

148+ # and input is not STDIN.

149+ encoding=$(

150+ head "$argument" |

151+ LC_ALL=C tr 'A-Z' 'a-z' |

152+ sed -ne '/<meta .*content-type.*charset=/ {

153+ s/.*charset=["'\'']*$[-a-zA-Z0-9]*$.*["'\'']*/\1/p

154+ }'

155+ )

156+fi

157+

158+if [ -n "$encoding" ] && pathfind iconv; then

159+ alias to_utf8='iconv -f "$encoding" -t utf-8'

160+else # assume UTF-8

161+ alias to_utf8='cat'

162+fi

163+

164+if [ -z "$argument" ]; then

165+ tidy -utf8 2>/dev/null | html2markdown "$@"

166+else

167+ if [ -f "$argument" ]; then

168+ to_utf8 "$argument" | tidy -utf8 2>/dev/null | html2markdown "$@"

169+ else

170+ err "File '$argument' not found."

171+ exit 1

172+ fi

173+fi