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"
                echo -n "."
                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
        links=`wc -l in | cut -d" " -f1`
        echo "found $links unique links"
}

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
}

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

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	dpavlin	1.1	#!/bin/sh
2
3	dpavlin	1.5	# proxy settings (same as in firebird)
4	dpavlin	1.7	fping proxy && export http_proxy=http://proxy:8080
5	dpavlin	1.5	# 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	dpavlin	1.1
12	dpavlin	1.2	if [ -z "$1" ] ; then
13			echo "Usage: $0 ISBN"
14			exit 1
15			fi
16
17	dpavlin	1.5	if [ -e orig ] ; then
18			echo "orig directory found. Resume download? [Y/n]"
19			read ans
20			if [ "$ans" = "n" ] ; then
21			exit 1;
22			fi
23			mv orig/* .
24			rm -Rf orig
25
26	dpavlin	1.7	grep -l 'promo.asp' * \| xargs -i rm {}
27			grep -l 'This is only a preview of the full book' * \| xargs -i rm {}
28	dpavlin	1.5	fi
29
30	dpavlin	1.2	isbn=$1;
31	dpavlin	1.1
32			isbn2=`echo $isbn \| sed 's/-//g'`
33
34			function mirror() {
35	dpavlin	1.5
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	dpavlin	1.8	echo -n "."
42	dpavlin	1.5	return
43			fi
44
45			if echo $url \| grep '/index' >/dev/null ; then
46			cookies=""
47			echo -n "no login (index) "
48			elif echo $url \| grep 'mode=toc' >/dev/null ; then
49			cookies=""
50			echo -n "no login (toc) "
51			else
52			cookies="--load-cookies=/home/dpavlin/.phoenix/default/g6b45nt6.slt/cookies.txt"
53			echo -n "with login "
54			fi
55			echo $url
56
57	dpavlin	1.1	wget -p -nH -nc -k \
58	dpavlin	1.5	-t 1 -U "$ua" \
59			$cookies $url
60	dpavlin	1.1
61			# -D safari.oreilly.com \
62			# -A 0-201-41975-0 \
63	dpavlin	1.5
64			perl -e '$t=rand($ENV{wait} \|\| 30);print "sleep for $t sec.\n"; sleep($t);'
65	dpavlin	1.1	}
66
67			function geturl() {
68	dpavlin	1.2	hindent -s $1 \| grep -i href \| grep mode=[st][eo]c \| \
69	dpavlin	1.1	sed -e 's/^.<a.href="//i' \
70			-e 's/".*//' -e 's/amp;//g' \
71			-e 's,^[^\?]*\?,http://safari.oreilly.com/,' \
72	dpavlin	1.8	-e 's/#.*$//' \
73	dpavlin	1.2	-e 's/\&srchText=//' \
74			-e 's/open=false/open=true/' \| \
75	dpavlin	1.1	grep '&s=1&b=1&f=1&t=1&c=1&u=1&r=&o=1' \| \
76	dpavlin	1.2	grep $2 \| \
77	dpavlin	1.5	grep -v "$2/[0-9][0-9][0-9][0-9][0-9][0-9][0-9]" \| \
78	dpavlin	1.1	sort -u >> in
79			}
80
81	dpavlin	1.2	function uniqurl() {
82			mv in in.tmp
83			grep -v 'view=[A-Z]' in.tmp \| sort -u > in
84			grep 'view=[A-Z].*/index' in.tmp \| sort -u >> in
85	dpavlin	1.8	links=`wc -l in \| cut -d" " -f1`
86			echo "found $links unique links"
87	dpavlin	1.2	}
88
89	dpavlin	1.4	function mirror_in() {
90			cat in \| while read url ; do
91			mirror "$url"
92	dpavlin	1.5	#sleep $wait
93	dpavlin	1.4
94			if grep 'promo.asp' `ls -t index.html* \| head -3` >/dev/null ; then
95	dpavlin	1.5	echo "WARNING: safari seems to logout you as user. Aborting."
96			exit 1
97			fi
98
99			if grep -i '>Account locked<' `ls -t index.html* \| head -3` >/dev/null ; then
100			echo "WARNING: your safari account is locked. Aborting."
101			exit 1
102			fi
103
104			if grep -i 'session disabled' `ls -t index.html* \| head -3` >/dev/null ; then
105			echo "WARNING: your safari session is disabled. Aborting."
106	dpavlin	1.4	exit 1
107			fi
108	dpavlin	1.7
109			if grep -i 'This is only a preview of the full book' `ls -t index.html* \| head -3` >/dev/null ; then
110			echo "WARNING: you didn't add this book to your bookshelf!"
111			exit 1
112			fi
113	dpavlin	1.4	done
114	dpavlin	1.8	echo
115	dpavlin	1.3	}
116
117	dpavlin	1.4	echo -n > in
118	dpavlin	1.2	mirror "http://safari.oreilly.com/?XmlId=$isbn"
119	dpavlin	1.8	echo
120	dpavlin	1.1
121			echo "extract URLs from first page..."
122			geturl "index.html?XmlId=$isbn" $isbn
123	dpavlin	1.2	uniqurl
124	dpavlin	1.1
125	dpavlin	1.4	mirror_in
126	dpavlin	1.1
127			echo -n "extracting URLs [1]"
128			ls index.html* \| while read file ; do
129			echo -n "."
130			geturl $file $isbn
131			done
132			echo
133
134	dpavlin	1.2	uniqurl
135	dpavlin	1.1
136	dpavlin	1.4	mirror_in
137	dpavlin	1.1
138	dpavlin	1.4	echo -n > in
139	dpavlin	1.1	echo -n "extracting URLs [2]"
140			ls index.html* \| while read file ; do
141			echo -n "."
142			geturl $file $isbn
143			done
144
145	dpavlin	1.2	uniqurl
146
147	dpavlin	1.4	mirror_in
148	dpavlin	1.2
149			# convert links in html
150			bn=`basename $0`
151			dir=`echo $0 \| sed "s/$bn$//"`
152			ls index.html* \| xargs -i $dir/filter.pl {}
153			mkdir orig
154			mv index.html* orig/
155	dpavlin	1.1