viewvc.cgi/safari/get_book.sh

#!/bin/sh

# proxy settings (same as in firebird)
fping proxy && export http_proxy=http://proxy:8080
# user agent (same as in firebird)
ua="Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031225 Firebird/0.7"

# wait between pages
export wait=120


if [ -z "$1" ] ; then
        echo "Usage: $0 ISBN"
        exit 1
fi

if [ -e orig ] ; then
        echo "orig directory found. Resume download? [Y/n]"
        read ans
        if [ "$ans" = "n" ] ; then
                exit 1;
        fi
        mv orig/* .
        rm -Rf orig

        grep -l 'promo.asp' * | xargs -i rm {}
        grep -l 'This is only a preview of the full book' * | xargs -i rm {}
fi

isbn=$1;

isbn2=`echo $isbn | sed 's/-//g'`

function mirror() {

        url="$1"

        file=`echo $url | sed -e s,http://[^?]*?,index.html?, -e s,#.*$,, -e s,/,%2F,g`
        if [ -e "$file" ] ; then
#               echo "skip $url"
                return
        fi

        if echo $url | grep '/index' >/dev/null ; then
                cookies=""
                echo -n "no login (index) "
        elif echo $url | grep 'mode=toc' >/dev/null ; then
                cookies=""
                echo -n "no login (toc) "
        else
                cookies="--load-cookies=/home/dpavlin/.phoenix/default/g6b45nt6.slt/cookies.txt"
                echo -n "with login "
        fi
        echo $url

        wget -p -nH -nc -k \
                -t 1 -U "$ua" \
                $cookies $url

#       -D safari.oreilly.com \
#       -A 0-201-41975-0 \

        perl -e '$t=rand($ENV{wait} || 30);print "sleep for $t sec.\n"; sleep($t);'
}

function geturl() {
        hindent -s $1 | grep -i href | grep mode=[st][eo]c | \
        sed -e 's/^.*<a.*href="//i' \
                -e 's/".*//' -e 's/amp;//g' \
                -e 's,^[^\?]*\?,http://safari.oreilly.com/,' \
                -e 's/#$//' \
                -e 's/\&srchText=//' \
                -e 's/open=false/open=true/' | \
                grep '&s=1&b=1&f=1&t=1&c=1&u=1&r=&o=1' | \
                grep $2 | \
                grep -v "$2/[0-9][0-9][0-9][0-9][0-9][0-9][0-9]" | \
                sort -u >> in
}

function uniqurl() {
        mv in in.tmp
        grep -v 'view=[A-Z]' in.tmp | sort -u > in
        grep 'view=[A-Z].*/index' in.tmp | sort -u >> in
}

function mirror_in() {
        cat in | while read url ; do
                mirror "$url"
                #sleep $wait

                if grep 'promo.asp' `ls -t index.html* | head -3` >/dev/null ; then
                        echo "WARNING: safari seems to logout you as user. Aborting."
                        exit 1
                fi

                if grep -i '>Account locked<' `ls -t index.html* | head -3` >/dev/null ; then
                        echo "WARNING: your safari account is locked. Aborting."
                        exit 1
                fi

                if grep -i 'session disabled' `ls -t index.html* | head -3` >/dev/null ; then
                        echo "WARNING: your safari session is disabled. Aborting."
                        exit 1
                fi

                if grep -i 'This is only a preview of the full book' `ls -t index.html* | head -3` >/dev/null ; then
                        echo "WARNING: you didn't add this book to your bookshelf!"
                        exit 1
                fi
        done
}

echo -n > in
mirror "http://safari.oreilly.com/?XmlId=$isbn"

echo "extract URLs from first page..."
geturl "index.html?XmlId=$isbn" $isbn
uniqurl

mirror_in

echo -n "extracting URLs [1]"
ls index.html* | while read file ; do
        echo -n "."
        geturl $file $isbn
done
echo

uniqurl

mirror_in

echo -n > in
echo -n "extracting URLs [2]"
ls index.html* | while read file ; do
        echo -n "."
        geturl $file $isbn
done

uniqurl

mirror_in

# convert links in html
bn=`basename $0`
dir=`echo $0 | sed "s/$bn$//"`
ls index.html* | xargs -i $dir/filter.pl {}
mkdir orig
mv index.html* orig/

1	#!/bin/sh
2
3	# proxy settings (same as in firebird)
4	fping proxy && export http_proxy=http://proxy:8080
5	# user agent (same as in firebird)
6	ua="Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031225 Firebird/0.7"
7
8	# wait between pages
9	export wait=120
10
11
12	if [ -z "$1" ] ; then
13	echo "Usage: $0 ISBN"
14	exit 1
15	fi
16
17	if [ -e orig ] ; then
18	echo "orig directory found. Resume download? [Y/n]"
19	read ans
20	if [ "$ans" = "n" ] ; then
21	exit 1;
22	fi
23	mv orig/* .
24	rm -Rf orig
25
26	grep -l 'promo.asp' * \| xargs -i rm {}
27	grep -l 'This is only a preview of the full book' * \| xargs -i rm {}
28	fi
29
30	isbn=$1;
31
32	isbn2=`echo $isbn \| sed 's/-//g'`
33
34	function mirror() {
35
36	url="$1"
37
38	file=`echo $url \| sed -e s,http://[^?]?,index.html?, -e s,#.$,, -e s,/,%2F,g`
39	if [ -e "$file" ] ; then
40	# echo "skip $url"
41	return
42	fi
43
44	if echo $url \| grep '/index' >/dev/null ; then
45	cookies=""
46	echo -n "no login (index) "
47	elif echo $url \| grep 'mode=toc' >/dev/null ; then
48	cookies=""
49	echo -n "no login (toc) "
50	else
51	cookies="--load-cookies=/home/dpavlin/.phoenix/default/g6b45nt6.slt/cookies.txt"
52	echo -n "with login "
53	fi
54	echo $url
55
56	wget -p -nH -nc -k \
57	-t 1 -U "$ua" \
58	$cookies $url
59
60	# -D safari.oreilly.com \
61	# -A 0-201-41975-0 \
62
63	perl -e '$t=rand($ENV{wait} \|\| 30);print "sleep for $t sec.\n"; sleep($t);'
64	}
65
66	function geturl() {
67	hindent -s $1 \| grep -i href \| grep mode=[st][eo]c \| \
68	sed -e 's/^.<a.href="//i' \
69	-e 's/".*//' -e 's/amp;//g' \
70	-e 's,^[^\?]*\?,http://safari.oreilly.com/,' \
71	-e 's/#$//' \
72	-e 's/\&srchText=//' \
73	-e 's/open=false/open=true/' \| \
74	grep '&s=1&b=1&f=1&t=1&c=1&u=1&r=&o=1' \| \
75	grep $2 \| \
76	grep -v "$2/[0-9][0-9][0-9][0-9][0-9][0-9][0-9]" \| \
77	sort -u >> in
78	}
79
80	function uniqurl() {
81	mv in in.tmp
82	grep -v 'view=[A-Z]' in.tmp \| sort -u > in
83	grep 'view=[A-Z].*/index' in.tmp \| sort -u >> in
84	}
85
86	function mirror_in() {
87	cat in \| while read url ; do
88	mirror "$url"
89	#sleep $wait
90
91	if grep 'promo.asp' `ls -t index.html* \| head -3` >/dev/null ; then
92	echo "WARNING: safari seems to logout you as user. Aborting."
93	exit 1
94	fi
95
96	if grep -i '>Account locked<' `ls -t index.html* \| head -3` >/dev/null ; then
97	echo "WARNING: your safari account is locked. Aborting."
98	exit 1
99	fi
100
101	if grep -i 'session disabled' `ls -t index.html* \| head -3` >/dev/null ; then
102	echo "WARNING: your safari session is disabled. Aborting."
103	exit 1
104	fi
105
106	if grep -i 'This is only a preview of the full book' `ls -t index.html* \| head -3` >/dev/null ; then
107	echo "WARNING: you didn't add this book to your bookshelf!"
108	exit 1
109	fi
110	done
111	}
112
113	echo -n > in
114	mirror "http://safari.oreilly.com/?XmlId=$isbn"
115
116	echo "extract URLs from first page..."
117	geturl "index.html?XmlId=$isbn" $isbn
118	uniqurl
119
120	mirror_in
121
122	echo -n "extracting URLs [1]"
123	ls index.html* \| while read file ; do
124	echo -n "."
125	geturl $file $isbn
126	done
127	echo
128
129	uniqurl
130
131	mirror_in
132
133	echo -n > in
134	echo -n "extracting URLs [2]"
135	ls index.html* \| while read file ; do
136	echo -n "."
137	geturl $file $isbn
138	done
139
140	uniqurl
141
142	mirror_in
143
144	# convert links in html
145	bn=`basename $0`
146	dir=`echo $0 \| sed "s/$bn$//"`
147	ls index.html* \| xargs -i $dir/filter.pl {}
148	mkdir orig
149	mv index.html* orig/
150