diff options
| -rw-r--r-- | Makefile | 37 | ||||
| -rw-r--r-- | README | 168 | ||||
| -rw-r--r-- | html2markdown | 69 | ||||
| -rw-r--r-- | latex2markdown | 32 | ||||
| -rw-r--r-- | man/man1/html2markdown.1 | 63 | ||||
| -rw-r--r-- | man/man1/latex2markdown.1 | 34 | ||||
| -rw-r--r-- | man/man1/markdown2html.1 | 35 | ||||
| -rw-r--r-- | man/man1/markdown2latex.1 | 32 | ||||
| -rw-r--r-- | man/man1/markdown2pdf.1 | 41 | ||||
| -rw-r--r-- | man/man1/pandoc.1 | 29 | ||||
| -rw-r--r-- | markdown2html | 31 | ||||
| -rw-r--r-- | markdown2latex | 31 | ||||
| -rw-r--r-- | markdown2pdf | 113 | ||||
| -rw-r--r-- | src/wrappers/checkin.sh | 7 | ||||
| -rw-r--r-- | src/wrappers/common.sh | 56 | ||||
| -rw-r--r-- | src/wrappers/getopts.sh | 12 | ||||
| -rw-r--r-- | src/wrappers/html2markdown.in | 134 | ||||
| -rw-r--r-- | src/wrappers/latex2markdown.in | 14 | ||||
| -rw-r--r-- | src/wrappers/markdown2html.in | 12 | ||||
| -rw-r--r-- | src/wrappers/markdown2latex.in | 12 | ||||
| -rw-r--r-- | src/wrappers/markdown2pdf.in | 64 | ||||
| -rw-r--r-- | src/wrappers/postopts.sh | 17 | ||||
| -rw-r--r-- | src/wrappers/singlearg.sh | 7 | ||||
| -rw-r--r-- | src/wrappers/tempdir.sh | 10 | ||||
| -rw-r--r-- | src/wrappers/testwrapper.in | 141 |
25 files changed, 735 insertions, 466 deletions
@@ -6,6 +6,7 @@ CABAL := Pandoc.cabal SRCDIR := src MANDIR := man +TESTDIR := tests BUILDDIR := dist BUILDCONF := .setup-config BUILDCMD := runhaskell Setup.hs @@ -24,7 +25,8 @@ MAIN := $(firstword $(EXECS)) #------------------------------------------------------------------------------- # Install targets #------------------------------------------------------------------------------- -PROGS := $(EXECS) html2markdown markdown2html latex2markdown markdown2latex markdown2pdf +WRAPPERS := html2markdown latex2markdown markdown2html markdown2latex markdown2pdf +PROGS := $(EXECS) $(WRAPPERS) DOCS := README.html README BUGS TODO #------------------------------------------------------------------------------- @@ -86,6 +88,27 @@ templates: $(SRCDIR)/templates $(SRCDIR)/templates: $(MAKE) -C $(SRCDIR)/templates +define generate-shell-script +echo "Generating $@..."; \ +awk ' \ + /^[ \t]*###+ / { \ + lead = $$0; sub(/[^ \t].*$$/, "", lead); \ + t = "$(dir $<)/"$$2; \ + while (getline line < t > 0) \ + print lead line; \ + next; \ + } \ + { print } \ +' <$< >$@ +endef + +.PHONY: wrappers +wrappers: $(WRAPPERS) +wrapper_deps := $(wildcard $(SRCDIR)/wrappers/*.sh) +cleanup_files+=$(WRAPPERS) +%: $(SRCDIR)/wrappers/%.in $(wrapper_deps) + @$(generate-shell-script) + cleanup_files+=$(CABAL) $(CABAL): cabalize $(CABAL).in ./cabalize <$(CABAL).in >$(CABAL) @@ -104,7 +127,7 @@ build: templates configure $(BUILDCMD) build .PHONY: build-exec -build-exec: $(EXECS) +build-exec: $(PROGS) cleanup_files+=$(EXECS) $(EXECS): build for f in $@; do \ @@ -250,11 +273,15 @@ osx-dmg: ../$(osx_dmg_name) -rm -f $(osx_dmg_name) mv $(osx_udzo_name) ../$(osx_dmg_name) -.PHONY: test test-markdown +.PHONY: test test-markdown test-wrapper test: $(MAIN) - @cd tests && perl runtests.pl -s $(PWD)/$(MAIN) + @cd $(TESTDIR) && perl runtests.pl -s $(PWD)/$(MAIN) test-markdown: $(MAIN) - @cd tests/MarkdownTest_1.0.3 && perl MarkdownTest.pl -s $(PWD)/$(MAIN) -tidy + @cd $(TESTDIR)/MarkdownTest_1.0.3 && perl MarkdownTest.pl -s $(PWD)/$(MAIN) -tidy +cleanup_files+=testwrapper +test-wrappers: testwrapper + @echo "Running $<..." + @sh testwrapper # Stolen and slightly improved from a GPLed Makefile. Credits to John Meacham. src_all:=$(shell find $(SRCDIR) -type f -name '*hs' | egrep -v '^\./(_darcs|lib|test)/') @@ -39,24 +39,24 @@ STDIN. If you run it with file names as arguments, it will take input from those files. It accepts several command-line options. For a list, type - pandoc -h + pandoc -h The most important options specify the format of the source file and the output. The default reader is markdown; the default writer is HTML. So if you don't specify a reader or writer, `pandoc` will convert markdown to HTML. For example, - pandoc hello.txt + pandoc hello.txt will convert `hello.txt` from markdown to HTML. For other conversions, you must specify a reader and/or a writer using the `-r` and `-w` flags. To convert markdown to LaTeX, you would write: - pandoc -w latex hello.txt + pandoc -w latex hello.txt To convert html to markdown: - pandoc -r html -w markdown hello.txt + pandoc -r html -w markdown hello.txt Supported writers include `markdown`, `latex`, `html`, `rtf` (rich text format), `rst` (reStructuredText), and `s5` (which produces an HTML @@ -120,72 +120,74 @@ before the files can be found by TeX.) # The shell scripts -For convenience, five shell scripts have been included that make it -easy to run `pandoc` without remembering all the command-line options. -All of the scripts use `iconv` to convert to and from the local -character encoding. All of the scripts presuppose that `pandoc` -is in the path, and some have additional requirements. (For example, -`html2markdown` uses `tidy`, and `markdown2pdf` uses `pdflatex`.) +Five shell scripts have been included that make it easy to run +`pandoc` without worrying about character encodings, and without +remembering all the command-line options: -1. `markdown2html` converts markdown to HTML. (This can be used - as a replacement for `Markdown.pl`.) +- `markdown2html` converts markdown-formatted text to HTML +- `markdown2latex` converts markdown-formatted text to LaTeX +- `markdown2pdf` produces a PDF file from markdown-formatted + text, using `pdflatex`. +- `html2markdown` converts HTML to markdown-formatted text +- `latex2markdown` converts LaTeX to markdown-formatted text -2. `html2markdown` can take either a filename or a URL as argument. If - it is given a URL, it uses `curl`, `wget`, or an available text-based - browser to fetch the contents of the specified URL, then filters this - through `tidy` to straighten up the HTML, and finally passes - this HTML to `pandoc` to produce markdown text: +All of the scripts use `iconv` (if available) to convert to and from +the local character encoding. All of the scripts presuppose that +`pandoc` is in the path, and some have additional requirements. (For +example, `html2markdown` uses `tidy`, and `markdown2pdf` uses +`pdflatex`.) - html2markdown http://www.fsf.org +When no arguments are specified, text will be read from standard +input. Arguments specify input files (limited to one in the case of +`latex2markdown` and `html2markdown`; the other scripts accept any number +of arguments). `html2markdown` may take a URL as argument instead of +a filename; in this case, `curl`, `wget`, or an available text-based +browser will be used to fetch the contents of the URL. (The `-n` option +inhibits this behavior; the `-g` option allows the user to specify a +custom command that will be used to fetch from a URL.) - html2markdown www.fsf.org +With the exception of `markdown2pdf`, the scripts write to standard output. +Output can be sent to a file using shell output redirection: - html2markdown subdir/mylocalfile.html + latex2markdown sample.tex > sample.txt -3. `latex2markdown` converts a LaTeX file to markdown: +The default behavior of `markdown2pdf` is to create a file with the same +base name as the first argument and the extension `pdf`; thus, for example, - latex2markdown mytexfile.tex + markdown2pdf sample.txt endnotes.txt -4. `markdown2latex` converts markdown to LaTeX: +will produce `sample.pdf`. (If `sample.pdf` exists already, it will be +backed up before being overwritten.) An output file name can be specified +explicitly using the `-o` option: - markdown2latex mytextfile.txt + markdown2pdf -o "My Book.pdf" chap1.txt chap2.txt chap3.txt -5. `markdown2pdf` converts markdown to PDF using `pdflatex`: +Options specific to the scripts, like `-o`, `-g`, and `-n`, must +be specified *before* any command-line arguments (file names or URLs). +Any options specified *after* the command-line arguments will be +passed directly to `pandoc`. For example, - markdown2pdf mytextfile.txt + markdown2html tusks.txt -S -T Elephants - creates a file `mytextfile.pdf`. To specify a different - name for the output file, use the `-o` option: +will convert `tusks.txt` to `tusks.html` using smart quotes, ellipses, +and dashes, with "Elephants" as the page title prefix. (For a +complete list of `pandoc` options, see below.) When there are no +command-line arguments (because input is from STDIN), `pandoc` +options must be preceded by ` -- `: - markdown2pdf -o final-draft.pdf mytextfile.txt + cat tusks.txt | markdown2html -- -S -T Elephants -If you want to use `pandoc`'s command-line options in these scripts, -put the options in the environment variable `PANDOC_OPTS`. For -example, to convert `tusks.txt` to `tusks.html` using smart quotes, -ellipses, and dashes, with "Elephants" as the page title prefix: +The ` -- ` separator may optionally be used when there are command-line +arguments: - PANDOC_OPTS="-S -T Elephants" markdown2html tusks.txt > tusks.html - -To make these options persistent, use `export`: - - export PANDOC_OPTS="-S -T Elephants" - markdown2html tusks.txt > tusks.html - markdown2html trunk.txt > trunk.html - -You may also specify options on the command line, separating them -from the arguments with '` -- `': - - markdown2html tusks.txt -- -S -T Elephants - -Options specified in this way will override any options set in -`PANDOC_OPTS`. + markdown2html -- tusks.txt -S -T Elephants # Command-line options Various command-line options can be used to customize the output. For a complete list, type - pandoc --help + pandoc --help `-p` or `--preserve-tabs` causes tabs in the source text to be preserved, rather than converted to spaces (the default). @@ -264,13 +266,13 @@ markdown test suite, type `make test-markdown`.) Pandoc behaves differently from standard markdown on some "edge cases" involving lists. Consider this source: - 1. First - 2. Second: - - Fee - - Fie - - Foe + 1. First + 2. Second: + - Fee + - Fie + - Foe - 3. Third + 3. Third Pandoc transforms this into a "compact list" (with no `<p>` tags around "First", "Second", or "Third"), while markdown puts `<p>` @@ -286,20 +288,20 @@ is followed by a blank line is irrelevant. Standard markdown allows unescaped literal quotes in titles, as in - [foo]: "bar "embedded" baz" + [foo]: "bar "embedded" baz" Pandoc requires all quotes within titles to be escaped: - [foo]: "bar \"embedded\" baz" + [foo]: "bar \"embedded\" baz" ## Reference links Pandoc allows implicit reference links in either of two styles: - 1. Here's my [link] - 2. Here's my [link][] + 1. Here's my [link] + 2. Here's my [link][] - [link]: linky.com + [link]: linky.com If there's no corresponding reference, the implicit reference link will appear as regular bracketed text. Note: even `[link][]` will @@ -313,18 +315,18 @@ Pandoc's markdown allows footnotes, using the following syntax: here is a footnote reference,^(1) and another.^(longnote) ^(1) Here is the footnote. It can go anywhere in the document, - except in embedded contexts like block quotes or lists. + except in embedded contexts like block quotes or lists. ^(longnote) Here's the other note. This one contains multiple blocks. ^ ^ Caret characters are used to indicate that the blocks all belong - to a single footnote (as with block quotes). + to a single footnote (as with block quotes). ^ ^ If you want, you can use a caret at the beginning of every line, - ^ as with blockquotes, but all that you need is a caret at the - ^ beginning of the first line of the block and any preceding - ^ blank lines. + ^ as with blockquotes, but all that you need is a caret at the + ^ beginning of the first line of the block and any preceding + ^ blank lines. Footnote references may not contain spaces, tabs, or newlines. @@ -335,21 +337,21 @@ Markdown 1.0. While Markdown 1.0 leaves HTML blocks exactly as they are, Pandoc treats text between HTML tags as markdown. Thus, for example, Pandoc will turn - <table> - <tr> - <td>*one*</td> - <td>[a link](http://google.com)</td> - </tr> - </table> + <table> + <tr> + <td>*one*</td> + <td>[a link](http://google.com)</td> + </tr> + </table> into - <table> - <tr> - <td><em>one</em></td> - <td><a href="http://google.com">a link</a></td> - </tr> - </table> + <table> + <tr> + <td><em>one</em></td> + <td><a href="http://google.com">a link</a></td> + </tr> + </table> whereas Markdown 1.0 will preserve it as is. @@ -426,10 +428,10 @@ include BibTeX citations: You can also use LaTeX environments. For example, \begin{tabular}{|l|l|}\hline - Age & Frequency \\ \hline + Age & Frequency \\ \hline 18--25 & 15 \\ - 26--35 & 33 \\ - 36--45 & 22 \\ \hline + 26--35 & 33 \\ + 36--45 & 22 \\ \hline \end{tabular} Note, however, that material between the begin and end tags will @@ -441,14 +443,14 @@ When run with the "standalone" option (`-s`), `pandoc` creates a standalone file, complete with an appropriate header. To see the default headers used for html and latex, use the following commands: - pandoc -D html + pandoc -D html - pandoc -D latex + pandoc -D latex If you want to use a different header, just create a file containing it and specify it on the command line as follows: - pandoc --header=MyHeaderFile + pandoc --header=MyHeaderFile # Producing S5 with Pandoc diff --git a/html2markdown b/html2markdown deleted file mode 100644 index 53ea05c08..000000000 --- a/html2markdown +++ /dev/null @@ -1,69 +0,0 @@ -#!/bin/sh -e -# converts html to markdown -# uses an available program to fetch URL and tidy to normalize it first - -pathfind () { # portable which(1), code taken from Debian Developer's Reference - OLDIFS="$IFS" - IFS=: - for _p in $PATH; do - if [ -x "$_p/$*" ]; then - IFS="$OLDIFS" - return 0 - fi - done - IFS="$OLDIFS" - return 1 -} - -for p in pandoc tidy; do - pathfind $p || { - echo >&2 "You need '$p' to use this program!" - exit 1 - } -done - -ALL="$*" -ARGS=${ALL%% -- *} # only the part before ' -- ' delimiters is relevant -set -- $ARGS - -REST=${ALL#$ARGS}; REST=${REST# -- } -PANDOC_OPTS=${REST:-$PANDOC_OPTS} - -infile=$1 - -if [ $# -gt 1 ]; then - shift - echo >&2 "Warning: extra arguments '$@' will be ignored!" -fi - -if [ -z "$infile" ] || [ -f $infile ]; then - tidy -utf8 $infile 2>/dev/null | \ - pandoc $PANDOC_OPTS -r html -w markdown -s | \ - iconv -f utf-8 -else - # Treat given argument as an URL. Locate a - # sensible text based browser (note the order). - for p in wget lynx w3m curl links w3c; do - if pathfind $p; then - DUMPER=$p - break - fi - done - # Setup proper options. - case "$DUMPER" in - wget) OPT="-O-" ;; - lynx) OPT="-source" ;; - w3m) OPT="-dump_source" ;; - curl) OPT="" ;; - links) OPT="-source" ;; - w3c) OPT="-n -get" ;; - "") printf "Needs a program to fetch the URL " >&2 - printf "(e.g. wget, w3m, lynx, w3c, or curl)." >&2 - exit 1 ;; - esac - # Fetch and feed to pandoc. - $DUMPER $OPT $infile 2>/dev/null | \ - tidy -utf8 2>/dev/null | \ - pandoc $PANDOC_OPTS -r html -w markdown -s | \ - iconv -f utf-8 -fi diff --git a/latex2markdown b/latex2markdown deleted file mode 100644 index 3974f279a..000000000 --- a/latex2markdown +++ /dev/null @@ -1,32 +0,0 @@ -#!/bin/sh -e -# runs pandoc to convert latex to markdown - -pathfind () { # portable which(1), code taken from Debian Developer's Reference - OLDIFS="$IFS" - IFS=: - for _p in $PATH; do - if [ -x "$_p/$*" ]; then - IFS="$OLDIFS" - return 0 - fi - done - IFS="$OLDIFS" - return 1 -} - -for p in pandoc; do - pathfind $p || { - echo >&2 "You need '$p' to use this program!" - exit 1 - } -done - -ALL="$*" -ARGS=${ALL%% -- *} # only the part before ' -- ' delimiters is relevant -set -- $ARGS - -REST=${ALL#$ARGS}; REST=${REST# -- } -PANDOC_OPTS=${REST:-$PANDOC_OPTS} - -iconv -t utf-8 $* | pandoc $PANDOC_OPTS -r latex -w markdown -s | \ - iconv -f utf-8 diff --git a/man/man1/html2markdown.1 b/man/man1/html2markdown.1 index b24f340ab..6cdba595c 100644 --- a/man/man1/html2markdown.1 +++ b/man/man1/html2markdown.1 @@ -1,33 +1,60 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH HTML2MARKDOWN 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME html2markdown \- converts HTML to markdown-formatted text .SH SYNOPSIS -\fBhtml2markdown\fR [ \fIinput-file\fR or \fIURL\fR ] +\fBhtml2markdown\fR [\fIoptions\fR] [\fIinput\-file\fR or \fIURL\fR] +[\fB\-\-\fR] [\fIpandoc\-opts\fR] .SH DESCRIPTION -\fBhtml2markdown\fR converts \fIinput-file\fR or \fIURL\fR -(or text from STDIN) from HTML to markdown-formatted plain text. -It uses an available program (e.g. wget, w3m, lynx or curl) to fetch -the contents of the URL. +\fBhtml2markdown\fR converts \fIinput\-file\fR or \fIURL\fR (or text +from STDIN) from HTML to markdown\-formatted plain text. +If a URL is specified, \fBhtml2markdown\fR uses an available program +(e.g. wget, w3m, lynx or curl) to fetch its contents. Output is sent +to STDOUT. .PP \fBhtml2markdown\fR is a wrapper for \fBpandoc\fR. .SH OPTIONS .TP -.B \-\- -Any options appearing after ` \-\- ' on the command line will be passed -directly to \fBpandoc\fR. See \fBpandoc\fR(1) for a list of options -that may be used. Options specified in this way will override -PANDOC_OPTS (see below). Example: +.B \-h +Show usage message. +.TP +.B \-e \fIencoding\fR +Assume the character encoding \fIencoding\fR in reading the HTML. +(Note: \fIencoding\fR will be passed to \fBiconv\fR; a list of +available encodings may be obtained using `\fBiconv \-l\fR'.) +If the \fB\-e\fR option is not specified, the encoding will be +determined as follows: If input is from STDIN, the local encoding +will be assumed. Otherwise, \fBhtml2markdown\fR will try to +extract the character encoding from the "Content-type" meta tag. +If no character encoding is specified in this way, UTF-8 will be +assumed for a URL argument, and the local encoding will be assumed +for a file argument. +.TP +.B \-g \fIcommand\fR +Use \fIcommand\fR to fetch the contents of a URL. (By default, +\fBhtml2markdown\fR searches for an available program or text-based +browser to fetch the contents of a URL.) For example: +.IP +html2markdown \-g 'wget \-\-user=foo \-\-password=bar' mysite.com +.TP +.B \-n +Disable automatic fetching of contents when URLs are specified as +arguments. +.TP +.I pandoc\-opts +Any options appearing after \fIinput\-file\fR or \fIURL\fR on the +command line will be passed directly to \fBpandoc\fR. If no +\fIinput-file\fR or \fIURL\fR is specified, these options must +be preceded by ` \fB\-\-\fR '. (In other cases, ` \fB\-\-\fR ' is +optional.) See \fBpandoc\fR(1) for a list of options that may be used. +Example: .IP -html2markdown input.txt -- -R -.SH ENVIRONMENT -Any command-line options contained in the PANDOC_OPTS environment variable -will be passed directly to \fBpandoc\fR. See \fBpandoc\fR(1) -for a list of options that may be used. +html2markdown input.txt \-\- \-R .SH "SEE ALSO" \fBpandoc\fR(1), \fBmarkdown2html\fR(1), \fBmarkdown2latex\fR(1), \fBlatex2markdown\fR(1), -\fBmarkdown2pdf\fR(1) +\fBmarkdown2pdf\fR(1), +\fBiconv\fR(1) .SH AUTHOR -John MacFarlane +John MacFarlane and Recai Oktas diff --git a/man/man1/latex2markdown.1 b/man/man1/latex2markdown.1 index 086cb4272..6e7e9e033 100644 --- a/man/man1/latex2markdown.1 +++ b/man/man1/latex2markdown.1 @@ -1,26 +1,28 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH LATEX2MARKDOWN 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME -latex2markdown \- converts LaTeX to markdown-formatted text +latex2markdown \- converts LaTeX to markdown\-formatted text .SH SYNOPSIS -\fBlatex2markdown\fR [ \fIinput-file\fR ] +\fBlatex2markdown\fR [\fIoptions\fR] [\fIinput\-file\fR] +[\fB\-\-\fR] [\fIpandoc\-opts\fR] .SH DESCRIPTION -\fBlatex2markdown\fR converts \fIinput-file\fR -(or text from STDIN) from LaTeX to markdown-formatted plain text. +\fBlatex2markdown\fR converts \fIinput\-file\fR +(or text from STDIN) from LaTeX to markdown\-formatted plain text. +Output is sent to STDOUT. .PP \fBlatex2markdown\fR is a wrapper for \fBpandoc\fR. .SH OPTIONS .TP -.B \-\- -Any options appearing after ` \-\- ' on the command line will be passed -directly to \fBpandoc\fR. See \fBpandoc\fR(1) for a list of options -that may be used. Options specified in this way will override -PANDOC_OPTS (see below). Example: +.B \-h +Show usage message. +.TP +.I pandoc\-opts +Any options appearing after \fIinput\-file\fR on the command line +will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR +is specified, these options must be preceded by ` \fB\-\-\fR '. +(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1) +for a list of options that may be used. Example: .IP -latex2markdown input.txt -- -R -.SH ENVIRONMENT -Any command-line options contained in the PANDOC_OPTS environment variable -will be passed directly to \fBpandoc\fR. See \fBpandoc\fR(1) -for a list of options that may be used. +latex2markdown input.txt \-\- \-R .SH "SEE ALSO" \fBpandoc\fR(1), \fBmarkdown2html\fR(1), @@ -28,4 +30,4 @@ for a list of options that may be used. \fBmarkdown2latex\fR(1), \fBmarkdown2pdf\fR(1) .SH AUTHOR -John MacFarlane +John MacFarlane and Recai Oktas diff --git a/man/man1/markdown2html.1 b/man/man1/markdown2html.1 index d6243182a..33d063321 100644 --- a/man/man1/markdown2html.1 +++ b/man/man1/markdown2html.1 @@ -1,28 +1,29 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH MARKDOWN2HTML 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME -markdown2html \- converts markdown-formatted text to HTML +markdown2html \- converts markdown\-formatted text to HTML .SH SYNOPSIS -\fBmarkdown2html\fR [ \fIinput-files\fR ] +\fBmarkdown2html\fR [\fIoptions\fR] [\fIinput\-file\fR]... +[\fB\-\-\fR] [\fIpandoc\-opts\fR] .SH DESCRIPTION -\fBmarkdown2html\fR converts \fIinput-files\fR -(or text from STDIN) from markdown-formatted plain text to HTML. +\fBmarkdown2html\fR converts \fIinput\-file\fR +(or text from STDIN) from markdown\-formatted plain text to HTML. If multiple files are specified, they will be combined to make a single -HTML document. +HTML document. Output is sent to STDOUT. .PP \fBmarkdown2html\fR is a wrapper for \fBpandoc\fR. .SH OPTIONS .TP -.B \-\- -Any options appearing after ` \-\- ' on the command line will be passed -directly to \fBpandoc\fR. See \fBpandoc\fR(1) for a list of options -that may be used. Options specified in this way will override -PANDOC_OPTS (see below). Example: +.B \-h +Show usage message. +.TP +.I pandoc\-opts +Any options appearing after \fIinput\-file\fR... on the command line +will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR +is specified, these options must be preceded by ` \fB\-\-\fR '. +(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1) +for a list of options that may be used. Example: .IP -markdown2html input.txt -- --css=main.css -S -.SH ENVIRONMENT -Any command-line options contained in the PANDOC_OPTS environment variable -will be passed directly to \fBpandoc\fR. See \fBpandoc\fR(1) -for a list of options that may be used. +markdown2html input.txt \-\- \-\-css=main.css \-S .SH "SEE ALSO" \fBpandoc\fR(1), \fBhtml2markdown\fR(1), @@ -30,4 +31,4 @@ for a list of options that may be used. \fBlatex2markdown\fR(1), \fBmarkdown2pdf\fR(1) .SH AUTHOR -John MacFarlane +John MacFarlane and Recai Oktas diff --git a/man/man1/markdown2latex.1 b/man/man1/markdown2latex.1 index 136d6ae84..3039192d1 100644 --- a/man/man1/markdown2latex.1 +++ b/man/man1/markdown2latex.1 @@ -1,27 +1,29 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH MARKDOWN2LATEX 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME markdown2latex \- converts markdown-formatted text to LaTeX .SH SYNOPSIS -\fBmarkdown2latex\fR [ \fIinput-files\fR ] +\fBmarkdown2latex\fR [\fIoptions\fR] [\fIinput\-file\fR]... +[\fB\-\-\fR] [\fIpandoc\-opts\fR] .SH DESCRIPTION -\fBmarkdown2latex\fR converts \fIinput-files\fR (or text from STDIN) -from markdown-formatted plain text to LaTeX. If multiple files are +\fBmarkdown2latex\fR converts \fIinput\-file\fR (or text from STDIN) +from markdown\-formatted plain text to LaTeX. If multiple files are specified, they will be combined to make a single LaTeX document. +Output is sent to STDOUT. .PP \fBmarkdown2latex\fR is a wrapper for \fBpandoc\fR. .SH OPTIONS .TP -.B \-\- -Any options appearing after ` \-\- ' on the command line will be passed -directly to \fBpandoc\fR. See \fBpandoc\fR(1) for a list of options -that may be used. Options specified in this way will override -PANDOC_OPTS (see below). Example: +.B \-h +Show usage message. +.TP +.I pandoc\-opts +Any options appearing after \fIinput\-file\fR... on the command line +will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR +is specified, these options must be preceded by ` \fB\-\-\fR '. +(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1) +for a list of options that may be used. Example: .IP -markdown2latex input.txt -- --custom-header=letterhead.tex -.SH ENVIRONMENT -Any command-line options contained in the PANDOC_OPTS environment variable -will be passed directly to \fBpandoc\fR. See \fBpandoc\fR(1) -for a list of options that may be used. +markdown2latex input.txt \-\- \-\-custom\-header=letterhead.tex .SH "SEE ALSO" \fBpandoc\fR(1), \fBmarkdown2html\fR(1), @@ -29,4 +31,4 @@ for a list of options that may be used. \fBlatex2markdown\fR(1), \fBmarkdown2pdf\fR(1) .SH AUTHOR -John MacFarlane +John MacFarlane and Recai Oktas diff --git a/man/man1/markdown2pdf.1 b/man/man1/markdown2pdf.1 index 4254e7119..99aa50a28 100644 --- a/man/man1/markdown2pdf.1 +++ b/man/man1/markdown2pdf.1 @@ -1,38 +1,37 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH MARKDOWN2PDF 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME markdown2pdf \- converts markdown-formatted text to PDF, using pdflatex .SH SYNOPSIS -\fBmarkdown2pdf\fR [ \fB-o|--output\fR \fIoutput-file\fR ] -[ \fIinput-file\fR ] +\fBmarkdown2pdf\fR [\fIoptions\fR] [\fB\-o\fR \fIoutput-file\fR] +[\fIinput-file\fR]... [\fB\-\-\fR] [\fIpandoc\-opts\fR] .SH DESCRIPTION -\fBmarkdown2pdf\fR converts \fIinput-file\fR (or text from STDIN) from -markdown-formatted plain text to PDF, using \fBpdflatex\fR. If no output +\fBmarkdown2pdf\fR converts \fIinput\-file\fR (or text from STDIN) from +markdown\-formatted plain text to PDF, using \fBpdflatex\fR. If no output filename is specified, the name of the output file is derived from the input file; thus, for example, if the input file is \fIhello.txt\fR, the output file will be \fIhello.pdf\fR. If the input is read from STDIN and no output filename is specified, the output file will be named -\fIstdin.pdf\fR. +\fIstdin.pdf\fR. If multiple input files are specified, they will be +concatenated before conversion, and the name of the output file will be +derived from the first input file. .PP \fBmarkdown2pdf\fR is a wrapper for \fBpandoc\fR. .SH OPTIONS .TP -.B \-o FILENAME, \-\-output=FILENAME -Specify name of output (PDF) file. -.TP -.B \-h, \-\-help +.B \-h Show usage message. .TP -.B \-\- -Any options appearing after ` \-\- ' on the command line will be passed -directly to \fBpandoc\fR. See \fBpandoc\fR(1) for a list of options -that may be used. Options specified in this way will override -PANDOC_OPTS (see below). Example: +.B \-o \fIoutput-file\fR +Specify name of output (PDF) file. +.TP +.I pandoc\-opts +Any options appearing after \fIinput\-file\fR... on the command line +will be passed directly to \fBpandoc\fR. If no \fIinput-file\fR +is specified, these options must be preceded by ` \fB\-\-\fR '. +(In other cases, ` \fB\-\-\fR ' is optional.) See \fBpandoc\fR(1) +for a list of options that may be used. Example: .IP -markdown2pdf input.txt -- --custom-header=letterhead.tex -.SH ENVIRONMENT -Any command-line options contained in the PANDOC_OPTS environment variable -will be passed directly to \fBpandoc\fR. See \fBpandoc\fR(1) -for a list of options that may be used. +markdown2pdf input.txt \-\- \-\-custom\-header=letterhead.tex .SH "SEE ALSO" \fBpandoc\fR(1), \fBmarkdown2html\fR(1), @@ -41,4 +40,4 @@ for a list of options that may be used. \fBlatex2markdown\fR(1), \fBpdflatex\fR(1) .SH AUTHOR -John MacFarlane +John MacFarlane and Recai Oktas diff --git a/man/man1/pandoc.1 b/man/man1/pandoc.1 index 0d0c1b09f..5f632de90 100644 --- a/man/man1/pandoc.1 +++ b/man/man1/pandoc.1 @@ -1,36 +1,35 @@ -.TH PANDOC 1 "November 1, 2006" Linux "User Manuals" +.TH PANDOC 1 "November 21, 2006" Pandoc "User Manuals" .SH NAME pandoc \- general markup converter .SH SYNOPSIS -\fBpandoc\fR [ \fIoptions\fR ] [ \fIinput-files\fR ] - +\fBpandoc\fR [\fIoptions\fR] [\fIinput\-file\fR]... .SH DESCRIPTION \fIPandoc\fR converts files from one markup format to another. It can read markdown and (subsets of) reStructuredText, HTML, and LaTeX, and it can write markdown, reStructuredText, HTML, LaTeX, RTF, and S5 HTML slide shows. .PP -If no \fIinput-files\fR are specified, input is read from STDIN. Otherwise, -the \fIinput-files\fR are concatenated (with a blank line between each) +If no \fIinput\-file\fR is specified, input is read from STDIN. Otherwise, +the \fIinput\-files\fR are concatenated (with a blank line between each) and used as input. Output goes to STDOUT. If you want output to a file, use shell redirection: .IP .B pandoc input.txt > output.html .PP The default behavior of \fIPandoc\fR is to convert the input from -markdown-formatted plain text to HTML. Different input and output -formats can be specified using command-line options. For example, +markdown\-formatted plain text to HTML. Different input and output +formats can be specified using command\-line options. For example, .IP -.B pandoc -f latex -t markdown chap1.tex > chap1.txt +.B pandoc \-f latex \-t markdown chap1.tex > chap1.txt .PP -converts \fIchap1.tex\fR from LaTeX to markdown-formatted plain text. -See below for a detailed list of command-line options. +converts \fIchap1.tex\fR from LaTeX to markdown\-formatted plain text. +See below for a detailed list of command\-line options. .PP -\fIPandoc\fR uses the UTF-8 character encoding for both input and output. -If your local character encoding is not UTF-8, you should pipe input +\fIPandoc\fR uses the UTF\-8 character encoding for both input and output. +If your local character encoding is not UTF\-8, you should pipe input and output through \fBiconv\fR: .IP -.B iconv -t utf-8 input.txt | pandoc | iconv -f utf-8 +.B iconv \-t utf\-8 input.txt | pandoc | iconv \-f utf\-8 .SH OPTIONS .TP @@ -40,7 +39,7 @@ Print version. .B \-h, \-\-help Show usage message. .TP -.B \-f FORMAT, \-r FORMAT, --from=FORMAT, --read=FORMAT +.B \-f FORMAT, \-r FORMAT, \-\-from=FORMAT, \-\-read=FORMAT Specify input format. .I FORMAT can be @@ -56,7 +55,7 @@ or .I latex (LaTeX). .TP -.B \-t FORMAT, \-w FORMAT, --to=FORMAT, --write=FORMAT +.B \-t FORMAT, \-w FORMAT, \-\-to=FORMAT, \-\-write=FORMAT Specify output format. .I FORMAT can be diff --git a/markdown2html b/markdown2html deleted file mode 100644 index 285dd866a..000000000 --- a/markdown2html +++ /dev/null @@ -1,31 +0,0 @@ -#!/bin/sh -e -# converts markdown to HTML - -pathfind () { # portable which(1), code taken from Debian Developer's Reference - OLDIFS="$IFS" - IFS=: - for _p in $PATH; do - if [ -x "$_p/$*" ]; then - IFS="$OLDIFS" - return 0 - fi - done - IFS="$OLDIFS" - return 1 -} - -for p in pandoc; do - pathfind $p || { - echo >&2 "You need '$p' to use this program!" - exit 1 - } -done - -ALL="$*" -ARGS=${ALL%% -- *} # only the part before ' -- ' delimiters is relevant -set -- $ARGS - -REST=${ALL#$ARGS}; REST=${REST# -- } -PANDOC_OPTS=${REST:-$PANDOC_OPTS} - -iconv -t utf-8 $* | pandoc $PANDOC_OPTS | iconv -f utf-8 diff --git a/markdown2latex b/markdown2latex deleted file mode 100644 index 9fafbafa5..000000000 --- a/markdown2latex +++ /dev/null @@ -1,31 +0,0 @@ -#!/bin/sh -e -# converts markdown to latex - -pathfind () { # portable which(1), code taken from Debian Developer's Reference - OLDIFS="$IFS" - IFS=: - for _p in $PATH; do - if [ -x "$_p/$*" ]; then - IFS="$OLDIFS" - return 0 - fi - done - IFS="$OLDIFS" - return 1 -} - -for p in pandoc; do - pathfind $p || { - echo >&2 "You need '$p' to use this program!" - exit 1 - } -done - -ALL="$*" -ARGS=${ALL%% -- *} # only the part before ' -- ' delimiters is relevant -set -- $ARGS - -REST=${ALL#$ARGS}; REST=${REST# -- } -PANDOC_OPTS=${REST:-$PANDOC_OPTS} - -iconv -t utf-8 $* | pandoc $PANDOC_OPTS -w latex -s | iconv -f utf-8 diff --git a/markdown2pdf b/markdown2pdf deleted file mode 100644 index 7799b1077..000000000 --- a/markdown2pdf +++ /dev/null @@ -1,113 +0,0 @@ -#!/bin/sh -# converts markdown to latex, then uses latex to make a PDF - -pathfind () { # portable which(1), code taken from Debian Developer's Reference - OLDIFS="$IFS" - IFS=: - for _p in $PATH; do - if [ -x "$_p/$*" ]; then - IFS="$OLDIFS" - return 0 - fi - done - IFS="$OLDIFS" - return 1 -} - -for p in pandoc pdflatex; do - pathfind $p || { - echo >&2 "You need '$p' to use this program!" - exit 1 - } -done - -ALL="$*" -ARGS=${ALL%% -- *} # only the part before ' -- ' delimiters is relevant -set -- $ARGS - -REST=${ALL#$ARGS}; REST=${REST# -- } -PANDOC_OPTS=${REST:-$PANDOC_OPTS} - -outfile= -for option; do - if [ -n "$prev" ]; then - eval "$prev=\$option" - prev= - shift - continue - fi - optarg=$(expr "x$option" : 'x[^=]*=\(.*\)') - case $option in - -h | --h | --help ) - help=1 - shift ;; - -o | --o | --output ) - prev=outfile - shift ;; - -o=* | --o=* | --output=* ) - outfile=$optarg - shift ;; - -* ) echo >&2 "$0: unknown option: $option; aborting" - exit 1 ;; - * ) break ;; - esac -done - -[ -z "$help" ] || { - echo >&2 "Usage: $0 [-o|--output output-file] [input-file]" - exit 0 -} - -infile=$1 - -if ! [ -r $infile ]; then - echo >&2 "'$infile' not found" - exit 1 -fi - -if [ $# -gt 1 ]; then - shift - echo >&2 "Warning: extra arguments '$@' will be ignored!" -fi - -if [ -z "$outfile" ]; then - if [ -n "$infile" ]; then - outfile=${infile%.*}.pdf - else - outfile='stdin.pdf' # input is STDIN, since no argument given - fi -fi - -BASE=${outfile##*/} -BASE=${BASE%.*} - -set -e - -TEMP=${TMPDIR-/tmp}/markdown2pdf.$$ -trap "status=$?; rm -rf $TEMP; exit $status" 0 INT -mkdir -p $TEMP - -iconv -t utf-8 $infile | \ -pandoc $PANDOC_OPTS -w latex -s | \ -iconv -f utf-8 > $TEMP/$BASE.tex && ( - cd $TEMP - if ! pdflatex -interaction=batchmode $BASE.tex >/dev/null 2>&1; then - echo >&2 "LaTeX errors:" - sed -ne '/^!/,/^ *$/p' $BASE.log >&2 - exit 1 - fi -) || exit $? - -is_target_exists= -if [ -f $outfile ]; then - is_target_exists=1 - mv -f $outfile $outfile~ -fi - -mv -f $TEMP/$BASE.pdf $outfile - -printf "Created $outfile" >&2 -[ -z "$is_target_exists" ] || { - printf " (previous file has been backed up as '$outfile~')" >&2 -} -echo >&2 . diff --git a/src/wrappers/checkin.sh b/src/wrappers/checkin.sh new file mode 100644 index 000000000..c9c564a23 --- /dev/null +++ b/src/wrappers/checkin.sh @@ -0,0 +1,7 @@ +# Check if input files exist. +for f; do + if [ -n "$f" ] && ! [ -f "$f" ]; then + err "File '$f' not found." + exit 1 + fi +done diff --git a/src/wrappers/common.sh b/src/wrappers/common.sh new file mode 100644 index 000000000..99a83be50 --- /dev/null +++ b/src/wrappers/common.sh @@ -0,0 +1,56 @@ +THIS=${0##*/} + +NEWLINE=' +' +WRAPPER_ARGS= +WRAPPEE_ARGS= + +err () { echo "$*" | fold -s -w ${COLUMNS:-110} >&2; } +errn () { printf "$*" | fold -s -w ${COLUMNS:-110} >&2; } + +usage () { + synopsis="$@" + err "Usage: $THIS $synopsis" + err "See $THIS(1) man file for details." +} + +runpandoc () { + if [ -n "$WRAPPEE_ARGS" ]; then + # Unpack arguments that will be passed to pandoc. + oldifs="$IFS"; IFS="$NEWLINE"; set -- $WRAPPEE_ARGS "$@"; IFS="$oldifs" + case "$1" in --) shift;; esac # tolerate the existence of a leading '--' + fi + + pandoc "$@" +} + +# Portable which(1). +pathfind () { + oldifs="$IFS"; IFS=':' + for _p in $PATH; do + if [ -x "$_p/$*" ] && [ -f "$_p/$*" ]; then + IFS="$oldifs" + return 0 + fi + done + IFS="$oldifs" + return 1 +} + +HAVE_ICONV= +if pathfind iconv; then + HAVE_ICONV=1 + alias to_utf8='iconv -t utf-8' + alias from_utf8='iconv -f utf-8' +else + err "Warning: iconv not present. Assuming UTF-8 character encoding." + alias to_utf8='cat' + alias from_utf8='cat' +fi + +for p in pandoc $REQUIRED; do + pathfind $p || { + err "You need '$p' to use this program!" + exit 1 + } +done diff --git a/src/wrappers/getopts.sh b/src/wrappers/getopts.sh new file mode 100644 index 000000000..263263c07 --- /dev/null +++ b/src/wrappers/getopts.sh @@ -0,0 +1,12 @@ +if [ -z "$SYNOPSIS" ]; then + SYNOPSIS="[-h] [input_file]" + [ -n "$THIS_NARG" ] || SYNOPSIS="${SYNOPSIS}..." +fi + +while getopts h opt; do + case $opt in + h|?) usage "$SYNOPSIS"; exit 2 ;; + esac +done + +shift $(($OPTIND - 1)) diff --git a/src/wrappers/html2markdown.in b/src/wrappers/html2markdown.in new file mode 100644 index 000000000..0fece3ccd --- /dev/null +++ b/src/wrappers/html2markdown.in @@ -0,0 +1,134 @@ +#!/bin/sh -e +# converts html to markdown +# uses an available program to fetch URL and tidy to normalize it first + +REQUIRED=tidy + +### common.sh + +grab_url_with () { + url="${1:?internal error: grab_url_with: url required}" + + shift + cmdline="$@" + + prog= + prog_opts= + if [ -n "$cmdline" ]; then + eval "set -- $cmdline" + prog=$1 + shift + prog_opts="$@" + fi + + if [ -z "$prog" ]; then + # Locate a sensible web grabber (note the order). + for p in wget lynx w3m curl links w3c; do + if pathfind $p; then + prog=$p + break + fi + done + + [ -n "$prog" ] || { + errn "$THIS: Couldn't find a program to fetch the file from URL " + err "(e.g. wget, w3m, lynx, w3c, or curl)." + return 1 + } + else + pathfind "$prog" || { + err "$THIS: No such web grabber '$prog' found; aborting." + return 1 + } + fi + + # Setup proper base options for known grabbers. + base_opts= + case "$prog" in + wget) base_opts="-O-" ;; + lynx) base_opts="-source" ;; + w3m) base_opts="-dump_source" ;; + curl) base_opts="" ;; + links) base_opts="-source" ;; + w3c) base_opts="-n -get" ;; + *) err "$THIS: unhandled web grabber '$prog'; hope it succeeds." + esac + + err "$THIS: invoking '$prog $base_opts $prog_opts $url'..." + eval "set -- $base_opts $prog_opts" + $prog "$@" "$url" +} + +encoding= +grabber= +nograb= +while getopts e:g:nh opt; do + case $opt in + e) encoding="$OPTARG" ;; + g) grabber="$OPTARG" ;; + n) nograb=1 ;; + h|?) + usage "[-e encoding] [-g grabber_command] [-n] [-h] [input_file|url]" + exit 2 ;; + esac +done + +shift $(($OPTIND - 1)) + +### postopts.sh + +### singlearg.sh + +inurl= +if [ -n "$1" ] && ! [ -f "$1" ]; then + if [ -n "$nograb" ]; then + err "'$1' not found; refusing to treat input as URL." + exit 1 + fi + # Treat given argument as an URL. + inurl="$1" +fi + +if [ -n "$inurl" ]; then + err "Attempting to fetch file from '$inurl'..." + + ### tempdir.sh + + grabber_out=$THIS_TEMPDIR/grabber.out + grabber_log=$THIS_TEMPDIR/grabber.log + if ! grab_url_with "$inurl" "$grabber" 1>$grabber_out \ + 2>$grabber_log; then + errn "grab_url_with failed" + if [ -f $grabber_log ]; then + err " with the following error log." + err + cat >&2 $grabber_log + else + err . + fi + exit 1 + fi + + set -- $grabber_out +fi + +if [ -z "$encoding" ] && [ "x$@" != "x" ]; then + # Try to determine character encoding unless not specified + # and input is STDIN. + encoding=$( + head "$@" | + LC_ALL=C tr 'A-Z' 'a-z' | + sed -ne '/<meta .*content-type.*charset=/ { + s/.*charset=["'\'']*\([-a-zA-Z0-9]*\).*["'\'']*/\1/p + }' + ) +fi + +if [ -n "$encoding" ] && [ -n "$HAVE_ICONV" ]; then + alias to_utf8='iconv -f "$encoding" -t utf-8' +elif [ -n "$inurl" ]; then # assume web pages are UTF-8 + alias to_utf8='cat' +fi # else just use local encoding + +to_utf8 "$@" | tidy -utf8 2>/dev/null | +runpandoc -r html -w markdown -s | from_utf8 diff --git a/src/wrappers/latex2markdown.in b/src/wrappers/latex2markdown.in new file mode 100644 index 000000000..e8cde8a97 --- /dev/null +++ b/src/wrappers/latex2markdown.in @@ -0,0 +1,14 @@ +#!/bin/sh -e +# runs pandoc to convert latex to markdown + +### common.sh + +### getopts.sh + +### postopts.sh + +### singlearg.sh + +### checkin.sh + +to_utf8 "$@" | runpandoc -r latex -w markdown -s | from_utf8 diff --git a/src/wrappers/markdown2html.in b/src/wrappers/markdown2html.in new file mode 100644 index 000000000..e255398d2 --- /dev/null +++ b/src/wrappers/markdown2html.in @@ -0,0 +1,12 @@ +#!/bin/sh -e +# converts markdown to HTML + +### common.sh + +### getopts.sh + +### postopts.sh + +### checkin.sh + +to_utf8 "$@" | runpandoc | from_utf8 diff --git a/src/wrappers/markdown2latex.in b/src/wrappers/markdown2latex.in new file mode 100644 index 000000000..c532b2f99 --- /dev/null +++ b/src/wrappers/markdown2latex.in @@ -0,0 +1,12 @@ +#!/bin/sh -e +# converts markdown to latex + +### common.sh + +### getopts.sh + +### postopts.sh + +### checkin.sh + +to_utf8 "$@" | runpandoc -w latex -s | from_utf8 diff --git a/src/wrappers/markdown2pdf.in b/src/wrappers/markdown2pdf.in new file mode 100644 index 000000000..838767224 --- /dev/null +++ b/src/wrappers/markdown2pdf.in @@ -0,0 +1,64 @@ +#!/bin/sh -e +# converts markdown to latex, then uses latex to make a PDF + +REQUIRED=pdflatex + +### common.sh + +outfile= +while getopts o:h opt; do + case $opt in + o) outfile="$OPTARG" ;; + h|?) usage "[-o output_file] [-h] [input_file]..."; exit 2 ;; + esac +done + +shift $(($OPTIND - 1)) + +### postopts.sh + +### checkin.sh + +if [ -z "$outfile" ]; then + if [ -n "$1" ]; then + outfile="${1%.*}" + else + outfile="stdin" # input is STDIN, since no argument given + fi +fi +case "$outfile" in +*.*) ;; # skip appending extension if one is already present +*) outfile="${outfile%.*}.pdf";; +esac + +### tempdir.sh + +# We should use a filename without white spaces for pdflatex. +TEXNAME=$THIS + +to_utf8 "$@" | runpandoc -w latex -s >$THIS_TEMPDIR/$TEXNAME.tex +( + cd $THIS_TEMPDIR + if ! pdflatex -interaction=batchmode $TEXNAME.tex >/dev/null 2>&1; then + err "LaTeX errors:" + from_utf8 $TEXNAME.log | sed -ne '/^!/,/^ *$/p' >&2 + if grep -q "File \`ucs.sty' not found" $TEXNAME.log; then + err "Please install the 'unicode' package from ctan.org." + fi + exit 1 + fi +) + +is_target_exists= +if [ -f "$outfile" ]; then + is_target_exists=1 + mv -f "$outfile" "$outfile~" +fi + +mv -f $THIS_TEMPDIR/$TEXNAME.pdf "$outfile" + +errn "Created '$outfile'" +[ -z "$is_target_exists" ] || { + errn " (previous file has been backed up as '$outfile~')" +} +err . diff --git a/src/wrappers/postopts.sh b/src/wrappers/postopts.sh new file mode 100644 index 000000000..e0d015f41 --- /dev/null +++ b/src/wrappers/postopts.sh @@ -0,0 +1,17 @@ +# Parse wrapper and wrappee (pandoc) arguments by taking +# into account that they may have space or tab characters. +pick="WRAPPER_ARGS" +while [ $# -gt 0 ]; do + if [ "$pick" = "WRAPPER_ARGS" ]; then + case "$1" in + -*) pick="WRAPPEE_ARGS" ;; + esac + fi + # Pack args with NEWLINE to preserve spaces, + # and put them into the picked variable. + eval "$pick=\"\$${pick}${NEWLINE}${1}\"" + shift +done + +# Unpack filename arguments. Now "$@" will hold the filenames. +oldifs="$IFS"; IFS="$NEWLINE"; set -- $WRAPPER_ARGS; IFS="$oldifs" diff --git a/src/wrappers/singlearg.sh b/src/wrappers/singlearg.sh new file mode 100644 index 000000000..f742d1383 --- /dev/null +++ b/src/wrappers/singlearg.sh @@ -0,0 +1,7 @@ +# Ensure to work with a single argument. +if [ $# -gt 1 ]; then + first_arg="$1" + shift + err "Warning: extra arguments '$@' will be ignored." + set -- $first_arg +fi diff --git a/src/wrappers/tempdir.sh b/src/wrappers/tempdir.sh new file mode 100644 index 000000000..8d2754b1c --- /dev/null +++ b/src/wrappers/tempdir.sh @@ -0,0 +1,10 @@ +# As a security measure refuse to proceed if mktemp is not available. +pathfind mktemp || { err "Couldn't find 'mktemp'; aborting."; exit 1; } + +THIS_TEMPDIR= +THIS_TEMPDIR="$(mktemp -d -t $THIS.XXXXXXXX)" || exit 1 +readonly THIS_TEMPDIR + +trap 'exitcode=$? + [ -z "$THIS_TEMPDIR" ] || rm -rf "$THIS_TEMPDIR" + exit $exitcode' 0 1 2 3 13 15 diff --git a/src/wrappers/testwrapper.in b/src/wrappers/testwrapper.in new file mode 100644 index 000000000..b9e10b450 --- /dev/null +++ b/src/wrappers/testwrapper.in @@ -0,0 +1,141 @@ +#!/bin/sh + +THIS=$1 + +ASH="ash -s" +BASH="bash --posix -s" +DASH="dash -s" +KSH="ksh -s" +POSH="posh -s" +ZSH="zsh -s" + +ERROR="" + +wrapper () { + $SH -- "$@" <<-'EOF' +### common.sh + +outfile= +while getopts o: opt; do + case $opt in + o) outfile="$OPTARG" ;; + esac +done + +shift $(($OPTIND - 1)) + +### postopts.sh + +echo "Options passed to wrapper:" +[ -z "$outfile" ] || echo "|$outfile|" + +echo "Arguments passed to wrapper:" +for arg; do + echo "|$arg|" +done + +pandoc () { + echo "Arguments passed to wrappee:" + for arg; do + echo "|$arg|" + done +} +runpandoc +EOF +} + +# Portable which(1). +pathfind () { + oldifs="$IFS"; IFS=':' + for _p in $PATH; do + if [ -x "$_p/$*" ] && [ -f "$_p/$*" ]; then + IFS="$oldifs" + return 0 + fi + done + IFS="$oldifs" + return 1 +} + +check_results () { + if [ "$1" = "$2" ]; then + echo >&2 ok + return 0 + else + echo >&2 failed + sed "s/^/\t/" >&2 <<EOF +Command line: '$3' +===> Expected: +$2 +<=== Got: +$1 +EOF + return 1 + fi +} + +for SH in "$BASH" "$DASH" "$KSH" "$ZSH"; do + CMD=${SH%% *} + echo >&2 " Testing with $CMD..." + if pathfind "$CMD"; then + if [ "$CMD" = "zsh" ]; then + # Zsh needs to be called as 'sh' to enable POSIX mode. + ln -s $(which zsh) ./sh + SH="./sh ${SH#* }" + trap 'err=$?; rm -f ./sh; exit $err' 0 1 2 3 13 15 + fi + + set -e + + # Test 1 + printf >&2 " test case 1... " + actual=$(wrapper -o "output file" "foo bar" -A "quux baz" -B) + expected=$(cat <<'EOF' +Options passed to wrapper: +|output file| +Arguments passed to wrapper: +|foo bar| +Arguments passed to wrappee: +|-A| +|quux baz| +|-B| +EOF +) + check_results "$actual" "$expected" \ + 'wrapper -o "output file" "foo bar" -A "quux baz" -B' + + # Test 2 + printf >&2 " test case 2... " + actual=$(wrapper -- -A "foo bar") + expected=$(cat <<'EOF' +Options passed to wrapper: +Arguments passed to wrapper: +Arguments passed to wrappee: +|-A| +|foo bar| +EOF +) + check_results "$actual" "$expected" 'wrapper -- -A "foo bar"' + + # Test 3 (Test 1 with a redundant '--') + printf >&2 " test case 4... " + actual=$(wrapper -o "output file" "foo bar" -- -A "quux baz" -B) + expected=$(cat <<'EOF' +Options passed to wrapper: +|output file| +Arguments passed to wrapper: +|foo bar| +Arguments passed to wrappee: +|-A| +|quux baz| +|-B| +EOF +) + check_results "$actual" "$expected" \ + 'wrapper -o "output file" "foo bar" -- -A "quux baz" -B' + else + echo >&2 "Warning: cannot verify correctness with $CMD; shell not available" + fi +done + +exit 0 |