自动下载一个blog的脚本

一念行者 的blog，内容很好，保存之，脚本如下：
这个脚本把所有的页面html文件取下来。

#!/bin/bash
#set -x

START_PAGE=$1

declare -A G_MAP_BLOG_NAME_ADDR
declare -A G_MAP_PAGE_NAME_ADDR
declare -A G_MAP_NAME_PARSE_STAT
declare -A G_MAP_NAME_IDX
declare -a G_MAP_IDX_NAME

G_NUM_HTML=0
G_THIS_INC=0
G_LINKS_IN_THIS_PAGE=

function get_links_in_page() {
    page=$1
    G_LINKS_IN_THIS_PAGE=
    if [ ! -e $page ]
    then
        echo "File '$page' not exist"
        return 1
    fi
    articles=` cat $1 |sed 's/href=/\nhref=/g' |grep "href=\"http:\/\/blog\.sina\.com\.cn\/s\/blog_[a-z0-9]\+\.html\"" |sed -n '/href=\"/ p' |sed 's/.*href=\"\([^"]\+\).*/\1/g'`

    pagelinks=`cat $1 |sed 's/href=/\nhref=/g' |grep "href=\"http:\/\/blog\.sina\.com\.cn\/s\/articlelist_[0-9a-z_]\+\.html\"" |sed -n '/href=\"/ p' |sed 's/.*href=\"\([^"]\+\).*/\1/g'`

    G_LINKS_IN_THIS_PAGE="$articles $pagelinks"
    return 0
}

function push_html() {
    if [ "aa$1" == "aa" ]
    then
        echo "Must provide a parameter for 'push_html()'"
        return 1
    fi
    html_link=$1
    html_file=${html_link##*\/}
    if [ "${html_file:(-4)}" != "html" ]
    then
        echo "Invalid parameter(not html) for 'push_html()': $html_link"
        return 1
    fi

    if [ -e $html_file ]
    then
        # already get the file
        return 999
    fi

    if [ "${html_file:0:4}" == "arti" ]
    then
        G_MAP_PAGE_NAME_ADDR[$html_file]=$html_link
    elif [ "${html_file:0:4}" == "blog" ]
    then
        G_MAP_BLOG_NAME_ADDR[$html_file]=$html_link
    else
        echo "Invlid parameter(not blog or article) for 'push_html()': $html_link"
        return 1
    fi

    G_MAP_NAME_PARSE_STAT[$html_file]=NOT_PARSED
    G_MAP_NAME_IDX[$html_file]=$G_NUM_HTML
    G_MAP_IDX_NAME[$G_NUM_HTML]=$html_file

    (( G_NUM_HTML ++))
    (( G_THIS_INC ++ ))

}
function is_local_link() {
    blogtitle">http://blog.sina.com.cn/xnfm
    if [ "aa$1" == "aa" ]
    then
        echo "In 'is_local_link()', you must provide a parameter"
        return 0
    fi
    if [ ! -e $1 ]
    then
        echo "In 'is_local_link()', file '$1' not exist"
        return 0
    fi
    cat $1 |grep "blogtitle\">
    if [ $? -eq 0 ]
    then
        return 1
    else
        return 0
    fi
}

function parse_html() {
    idx=0
    while [ $idx -lt $G_NUM_HTML ] && [ "aa${G_MAP_IDX_NAME[$idx]}" != "aa" ]
    do
        html_name=${G_MAP_IDX_NAME[$idx]}
        if [ "${G_MAP_NAME_PARSE_STAT[$html_name]}" == "NOT_PARSED" ]
        then
            if [ "${html_name:0:4}" == "blog" ]
            then
                html_link=${G_MAP_BLOG_NAME_ADDR[$html_name]}
            elif [ "${html_name:0:4}" == "arti" ]
            then
                html_link=${G_MAP_PAGE_NAME_ADDR[$html_name]}
            else
                echo "Something must be error, get invalid html name from G_MAP_IDX_NAME"
                echo "idx: $idx; name: ${G_MAP_IDX_NAME[$idx]}"
            fi

            wget $html_link
            if [ $? -ne 0 ]
            then
                echo "get '$html_link' fail\n"
            else
                is_local_link $html_name
                if [ $? -eq 1 ] #local link
                then
                    get_links_in_page $html_name
                    if [ $? -eq 0 ]
                    then
                        G_MAP_NAME_PARSE_STAT[$html_name]="PARSED"
                        for xx in $G_LINKS_IN_THIS_PAGE
                        do
                            push_html $xx
                        done
                    else
                        echo "Fail 'get_links_in_page $html_name'"
                    fi
                fi
            fi
        fi

        (( idx ++ ))
    done
}

push_html $START_PAGE

while [ "1" == "1" ]
do
    G_THIS_INC=0
    parse_html
    if [ $G_THIS_INC -ne 0 ]
    then
        continue
    else
        echo "No more file"
        break
    fi
done

rm article*html
for ff in $.html
do
    is_local_link $ff
    if [ $? -ne 1 ]
    then
        rm $ff
    fi
done