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:1.6) Gecko/20040506 Firefox/0.8"
cookie_file="cookie.txt"

# 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=`echo $1 | 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

        cookie=" "
        if echo $url | grep '/index' >/dev/null ; then
                echo -n "no login (index) "
        elif echo $url | grep 'mode=toc' >/dev/null ; then
                echo -n "no login (toc) "
        else
                if [ ! -e $cookie_file ] ; then
                        echo "cookies file $cookie_file doesn't exits! Please create it."
                        echo "It should be in format:"
                        echo "Cookie: Site=UICode=&Portal=oreilly&GUID=..."
                        exit 1
                fi
                read cookie < $cookie_file
                if [ -z "$cookie" ] ; then
                        echo "Empty cookie file $cookie_file !"
                        exit 1
                fi
        fi

        if [ -z "$cookie" ] ; then
                echo "$url [no cookie]"
        else
                echo "$url [with cookie]"
        fi

        wget -q -p -nH -nc -k -t 1 -U "$ua" --cookies=off --header="$cookie" $url

        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	dpavlin	1.9	ua="Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040506 Firefox/0.8"
7			cookie_file="cookie.txt"
8	dpavlin	1.5
9			# wait between pages
10			export wait=120
11
12	dpavlin	1.1
13	dpavlin	1.2	if [ -z "$1" ] ; then
14			echo "Usage: $0 ISBN"
15			exit 1
16			fi
17
18	dpavlin	1.5	if [ -e orig ] ; then
19			echo "orig directory found. Resume download? [Y/n]"
20			read ans
21			if [ "$ans" = "n" ] ; then
22			exit 1;
23			fi
24			mv orig/* .
25			rm -Rf orig
26
27	dpavlin	1.7	grep -l 'promo.asp' * \| xargs -i rm {}
28			grep -l 'This is only a preview of the full book' * \| xargs -i rm {}
29	dpavlin	1.5	fi
30
31	dpavlin	1.9	isbn=`echo $1 \| sed 's/-//g'`
32	dpavlin	1.1
33			function mirror() {
34	dpavlin	1.5
35			url="$1"
36
37			file=`echo $url \| sed -e s,http://[^?]?,index.html?, -e s,#.$,, -e s,/,%2F,g`
38			if [ -e "$file" ] ; then
39			# echo "skip $url"
40	dpavlin	1.8	echo -n "."
41	dpavlin	1.5	return
42			fi
43
44	dpavlin	1.10	cookie=" "
45	dpavlin	1.5	if echo $url \| grep '/index' >/dev/null ; then
46			echo -n "no login (index) "
47			elif echo $url \| grep 'mode=toc' >/dev/null ; then
48			echo -n "no login (toc) "
49			else
50	dpavlin	1.9	if [ ! -e $cookie_file ] ; then
51			echo "cookies file $cookie_file doesn't exits! Please create it."
52			echo "It should be in format:"
53			echo "Cookie: Site=UICode=&Portal=oreilly&GUID=..."
54			exit 1
55			fi
56			read cookie < $cookie_file
57			if [ -z "$cookie" ] ; then
58			echo "Empty cookie file $cookie_file !"
59			exit 1
60			fi
61	dpavlin	1.5	fi
62
63	dpavlin	1.9	if [ -z "$cookie" ] ; then
64			echo "$url [no cookie]"
65			else
66			echo "$url [with cookie]"
67			fi
68	dpavlin	1.1
69	dpavlin	1.9	wget -q -p -nH -nc -k -t 1 -U "$ua" --cookies=off --header="$cookie" $url
70	dpavlin	1.5
71			perl -e '$t=rand($ENV{wait} \|\| 30);print "sleep for $t sec.\n"; sleep($t);'
72	dpavlin	1.1	}
73
74			function geturl() {
75	dpavlin	1.2	hindent -s $1 \| grep -i href \| grep mode=[st][eo]c \| \
76	dpavlin	1.1	sed -e 's/^.<a.href="//i' \
77			-e 's/".*//' -e 's/amp;//g' \
78			-e 's,^[^\?]*\?,http://safari.oreilly.com/,' \
79	dpavlin	1.8	-e 's/#.*$//' \
80	dpavlin	1.2	-e 's/\&srchText=//' \
81			-e 's/open=false/open=true/' \| \
82	dpavlin	1.1	grep '&s=1&b=1&f=1&t=1&c=1&u=1&r=&o=1' \| \
83	dpavlin	1.2	grep $2 \| \
84	dpavlin	1.5	grep -v "$2/[0-9][0-9][0-9][0-9][0-9][0-9][0-9]" \| \
85	dpavlin	1.1	sort -u >> in
86			}
87
88	dpavlin	1.2	function uniqurl() {
89			mv in in.tmp
90			grep -v 'view=[A-Z]' in.tmp \| sort -u > in
91			grep 'view=[A-Z].*/index' in.tmp \| sort -u >> in
92	dpavlin	1.8	links=`wc -l in \| cut -d" " -f1`
93			echo "found $links unique links"
94	dpavlin	1.2	}
95
96	dpavlin	1.4	function mirror_in() {
97			cat in \| while read url ; do
98			mirror "$url"
99	dpavlin	1.5	#sleep $wait
100	dpavlin	1.4
101			if grep 'promo.asp' `ls -t index.html* \| head -3` >/dev/null ; then
102	dpavlin	1.5	echo "WARNING: safari seems to logout you as user. Aborting."
103			exit 1
104			fi
105
106			if grep -i '>Account locked<' `ls -t index.html* \| head -3` >/dev/null ; then
107			echo "WARNING: your safari account is locked. Aborting."
108			exit 1
109			fi
110
111			if grep -i 'session disabled' `ls -t index.html* \| head -3` >/dev/null ; then
112			echo "WARNING: your safari session is disabled. Aborting."
113	dpavlin	1.4	exit 1
114			fi
115	dpavlin	1.7
116			if grep -i 'This is only a preview of the full book' `ls -t index.html* \| head -3` >/dev/null ; then
117			echo "WARNING: you didn't add this book to your bookshelf!"
118			exit 1
119			fi
120	dpavlin	1.4	done
121	dpavlin	1.8	echo
122	dpavlin	1.3	}
123
124	dpavlin	1.4	echo -n > in
125	dpavlin	1.2	mirror "http://safari.oreilly.com/?XmlId=$isbn"
126	dpavlin	1.8	echo
127	dpavlin	1.1
128			echo "extract URLs from first page..."
129			geturl "index.html?XmlId=$isbn" $isbn
130	dpavlin	1.2	uniqurl
131	dpavlin	1.1
132	dpavlin	1.4	mirror_in
133	dpavlin	1.1
134			echo -n "extracting URLs [1]"
135			ls index.html* \| while read file ; do
136			echo -n "."
137			geturl $file $isbn
138			done
139			echo
140
141	dpavlin	1.2	uniqurl
142	dpavlin	1.1
143	dpavlin	1.4	mirror_in
144	dpavlin	1.1
145	dpavlin	1.4	echo -n > in
146	dpavlin	1.1	echo -n "extracting URLs [2]"
147			ls index.html* \| while read file ; do
148			echo -n "."
149			geturl $file $isbn
150			done
151
152	dpavlin	1.2	uniqurl
153
154	dpavlin	1.4	mirror_in
155	dpavlin	1.2
156			# convert links in html
157			bn=`basename $0`
158			dir=`echo $0 \| sed "s/$bn$//"`
159			ls index.html* \| xargs -i $dir/filter.pl {}
160			mkdir orig
161			mv index.html* orig/
162	dpavlin	1.1