echoprint

来源：互联网发布：linux 播放器剪辑编辑：程序博客网时间：2024/05/01 08:59

安装数据库：tokyotyrant/

下载地址：

wget http://cloud.github.com/downloads/xurenlu/mc_list_patch_4_tokyotyrant/tokyocabinet-1.4.41.tar.gz

wget http://cloud.github.com/downloads/xurenlu/mc_list_patch_4_tokyotyrant/tokyotyrant-1.1.37.tar.gz

tar -xzf tokyotyrant-1.1.37.tar.gz  327  cd tokyocabinet-1.4.41  328  ll  329   tar -xzf tokyocabinet-1.4.41.tar.gz  330  ll  331  cd tokyocabinet-1.4.41  332  ./configure  333   make && make install  334  ll  335  cd ../  340  cd tokyotyrant-1.1.37  341   ./configure  342  make && sudo make install  343  pwd  344  cd ../  345  ll  346  find / -name ttserver  347  mkdir -p /ttserver/  348  chmod 777 -R /ttserver/  349  /usr/local/bin/ttserver -host 127.0.0.1 -port 11111 -thnum 8 -dmn -pid /ttserver/ttserver.pid -log /ttserver/ttserver.log -le -ulog /ttserver/ -ulim 128m -sid 1 -rts /ttserver/ttserver.rts /ttserver/database.tch  350  curl -X PUT http://127.0.0.1:11111/my_key -d "this is value"    351  curl http://127.0.0.1:11111/my_key  352  curl -X DELETE http://127.0.0.1:11111/my_key

安装web.py,

1, http://webpy.org/static/web.py-0.37.tar.gz
2, tar xvfz web.py-0.37.tar.gz
3, cd web.py-0.37
4, sudo python setup.py install

启动数据库：
ttservctl start

启动solr:

# cd /usr/local/src/echoprint/echoprint-server/solr/solr/

# java -Dsolr.solr.home=/usr/local/src/echoprint/echoprint-server/solr/solr/solr/ -Djava.awt.headless=true -jar start.jar

solr管理地址：
http://192.168.1.3:8502/solr/fp/admin/

cd echoprint-server-master\API目录，测试数据写入读出：
>>> import fp
>>> fp.ingest({"track_id": "my_track_id", "fp": "123 40 123 60 123 80 123 90 123 110 123 130", "length": "120", "codever": "4.12"})
>>> fp.commit()
>>> r = fp.best_match_for_query("123 40 124 60 125 80 126 90 127 110 128 130 129 60 123 40 127 50")
>>> r.message()
'query code length is too small'
>>> example_code = "eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw"
>>> r = fp.best_match_for_query(example_code)
>>> r.message()
'OK (match type 3)'
>>> r.TRID
'my_track_id'

打开web.py服务，添加数据：
cd API
python api.py 8080

$ curl http://192.168.1.3:8080/ingest -d "fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw&track_id=thisone&length=300&codever=4.12"

后台启动：

nohup python code.py 8088 > /webpy.log 2>&1 &

查询数据
http://192.168.1.3:8080/query?fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw

开始装codegen
https://github.com/echonest/echoprint-codegen

（http://blog.163.com/yungang_z/blog/static/175153133201262445539767/）
安装# yum install zlib zlib-devel
# yum install ffmpeg boost taglib

# yum install ffmpeg-devel boost-devel taglib-devel

如果ffmpeg安装不成功，使用以下方法：

方法二、此方法有时不太好用

1.安装前准备

vi /etc/yum.repos.d/ffmpeg.repo

[dag]

name=Dag RPM Repository for Red Hat Enterprise Linux

baseurl=http://apt.sw.be/redhat/el\$releasever/en/\$basearch/dag

gpgcheck=0

enabled=1

2.安装ffmpeg

yum -y install ffmpeg ffmpeg-devel

开始安装codegen:

389 unzip echoprint-codegen-master.zip
390 ll
391 mv echoprint-codegen-master echoprint-codegen
392 cd echoprint-codegen
397 cd src/
398 ll
399 whereis boost
400 vi Makefile //edit BOOST_CFLAGS and other variables as necessary
401 make
402 cd ..
403 ll

# ./echoprint-codegen billie_jean.mp3 10 30

# cd /usr/local/src/echoprint/echoprint-server/util

440 mkdir /music

444 ll /music
445 find /music -name "*.mp3" > music_to_ingest
446 ./echoprint-codegen -s < music_to_ingest > allcodes.json
# python splitdata.py /usr/local/src/echoprint/echoprint-codegen/allcodes.json

/
(http://blog.sina.com.cn/s/blog_7985987f010196c1.html)
安装python-setuptools
wget –q http://peak.telecommunity.com/dist/ez_setup.py
python ez_setup.py
[root@MIO-03 pyechonest-master]# python setup.py install
pyechonest：
Your API Key: G1QCI9RB0AIJKXTJN

Your Consumer Key: 5cd12bae2c16f47e3e14c3eef93e037c

Your Shared Secret: +EniaamaQgiSd8nRQ07a2g

ps -ef | grep solr

操作手册

启动数据库：
ttservctl restart

启动solr:
# cd /usr/local/src/echoprint/echoprint-server/solr/solr/
# java -Dsolr.solr.home=/usr/local/src/echoprint/echoprint-server/solr/solr/solr/ -Djava.awt.headless=true -jar start.jar

solr管理地址：
http://172.16.146.129:8502/solr/fp/admin/

原生的数据操作：
tcrmgr list -port 1978 172.16.146.129
curl -X PUT http://172.16.146.129:1978/mykey -d "this is value"
curl http://172.16.146.129:1978/thisone2
curl -X DELETE http://172.16.146.129:1978/my_key

命令行测试数据写入读出：
cd /usr/local/src/echoprint/echoprint-server/API目录，
>>> import fp
>>> fp.ingest({"track_id": "my_track_id", "fp": "123 40 123 60 123 80 123 90 123 110 123 130", "length": "120", "codever": "4.12"})
>>> fp.delete(['my_track_id-0','thisfour-0'])
>>> fp.commit()
>>> r = fp.best_match_for_query("123 40 124 60 125 80 126 90 127 110 128 130 129 60 123 40 127 50")
>>> r.message()
'query code length is too small'
>>> example_code = "eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw"
>>> r = fp.best_match_for_query(example_code)
>>> r.message()
'OK (match type 3)'
>>> r.TRID
'my_track_id'

web测试数据读写：
打开web.py服务，开启8080端口，添加数据：
cd /usr/local/src/echoprint/echoprint-server/API
python api.py 8080

模拟post提交添加数据：

$ curl http://172.16.146.129:8080/ingest -d "fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw&track_id=thisone&length=300&codever=4.12"

$ curl http://172.16.146.129:8080/ingest -d "fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw&track_id=enya&length=300&codever=4.12"

查询数据
http://172.16.146.129:8080/query?fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw
123 40 124 60 125 80 126 90 127 110 128 130 129 60 123 40 127 50 127 60

solr返回json:&wt=json
http://172.16.146.129:8502/solr/fp/select/?q=123+40+124+60+125+80+126+90+127+110+128+130+129+60+123+40+127+50+127+60&version=2.2&start=0&rows=10&indent=on&wt=json

通过ID查询：
http://172.16.146.129:8502/solr/fp/select/?q=track_id:thisone-0&version=2.2&start=0&rows=10&indent=on&wt=json

转码
http://172.16.146.129:8080/decode?fp_code=eJwty7kNADAMw8BVNILl-Mv-iwWCU11D0g_CQA-USIwoXNEg5YBH3o3-0sil7AHIrAyw

到这里又可以用命令行去查数据，会现r.TRID变成了刚ingent的ID：
>>> r.TRID
'thisone'

通过codegen把mp3信息生成json文件，
cd /usr/local/src/echoprint/echoprint-codegen
# ./echoprint-codegen Kalimba.mp3 10 30
find /music -name "*.mp3" > music_to_ingest
./echoprint-codegen -s < music_to_ingest > allcodes.json

然后在server里把这个json数据添加进去；
cd /usr/local/src/echoprint/echoprint-server/util
python fastingest.py /usr/local/src/echoprint/echoprint-codegen/allcodes.json
python splitdata.py /usr/local/src/echoprint/echoprint-codegen/allcodes.json

0 0