aboutsummaryrefslogtreecommitdiff
path: root/src/wrappers/html2markdown.in
blob: 740d695885489d723172eaf3c51e716dd1f656df (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
#!/bin/sh -e
# converts HTML from a URL, file, or stdin 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"
}

add_option () {
    options="$options$NEWLINE$1"
}

options=
argument=
encoding=
grabber=

# Parse command-line arguments
while [ $# -gt 0 ]; do
    case "$1" in
        -h|--help)
            pandoc -h 2>&1 | sed -e 's/pandoc/html2markdown/' \
            -e '/^[[:space:]]*\(-f\|-t\|-S\|-N\|-m\|-i\|-c\|-T\|-D\|-d\)/,/./d'\
            1>&2 
            err "  -e ENCODING, --encoding=ENCODING"
            err "        Specify character encoding of input"
            err "  -g COMMAND, --grabber=COMMAND"
            err "        Specify command to be used to grab contents of URL"
            exit 0 ;;
        -v|--version) 
            pandoc -v 2>&1 | sed -e 's/pandoc/html2markdown/' 1>&2
            exit 0 ;;
        -e)
            shift
            encoding=$1 ;; 
        --encoding=*)
            wholeopt=$1
            # extract encoding from after =
            encoding=${wholeopt#*=} ;;
        -g)
            shift
            grabber=$1 ;; 
        --grabber=*)
            wholeopt=$1
            # extract encoding from after =
            grabber=${wholeopt#*=} ;;
        -o|--output|-b|--tab-stop|-H|--include-in-header| \
        -A|--include-after-body|-C|-B|--include-before-body| \
        -C|--custom-header|-T|--title-prefix) 
            add_option $1
            shift
            add_option $1 ;;
        -*) add_option $1 ;;
        *) 
            if [ -z "$argument" ]; then
                argument=$1
            else
                err "Warning: extra argument '$1' will be ignored."
            fi ;;            
    esac
    shift
done

# Unpack options.  Now "$@" will hold the pandoc options.
oldifs="$IFS"; IFS="$NEWLINE"; set -- $options; IFS="$oldifs"

inurl=
if [ -n "$argument" ] && ! [ -f "$argument" ]; then
    # Treat given argument as an URL.
    inurl="$argument"
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

    argument="$grabber_out"
fi

if [ -z "$encoding" ] && [ "x$argument" != "x" ]; then
    # Try to determine character encoding if not specified
    # and input is not STDIN.
    encoding=$(
        head "$argument" |
        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 "$encoding" ] && pathfind iconv; then
    alias to_utf8='iconv -f "$encoding" -t utf-8'
else # assume UTF-8
    alias to_utf8='cat'
fi 

if [ -z "$argument" ]; then
    tidy -utf8 2>/dev/null | pandoc -r html -w markdown "$@"
else
    if [ -f "$argument" ]; then
        to_utf8 "$argument" | 
        tidy -utf8 2>/dev/null | pandoc -r html -w markdown "$@"
    else
        err "File '$argument' not found."
        exit 1
    fi
fi