Skip to content
Projects
Groups
Snippets
Help
This project
Loading...
Sign in / Register
Toggle navigation
S
searchhotcrawler
Overview
Overview
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
zhiwei
searchhotcrawler
Commits
ee793daf
Commit
ee793daf
authored
Jul 26, 2019
by
zhiwei
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
更新提交
parent
d4cb16b0
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
46 additions
and
25 deletions
+46
-25
src/main/java/com/zhiwei/searchhotcrawler/dao/HotSearchListDAO.java
+0
-2
src/main/java/com/zhiwei/searchhotcrawler/test/HotSearchListTest.java
+46
-23
No files found.
src/main/java/com/zhiwei/searchhotcrawler/dao/HotSearchListDAO.java
View file @
ee793daf
...
...
@@ -2,7 +2,6 @@ package com.zhiwei.searchhotcrawler.dao;
import
java.util.ArrayList
;
import
java.util.Calendar
;
import
java.util.Date
;
import
java.util.List
;
...
...
@@ -30,7 +29,6 @@ public class HotSearchListDAO extends MongoDBTemplate{
String
time
=
TimeParse
.
dateFormartString
(
new
Date
(),
"yyyy-MM-dd"
);
String
year
=
time
.
substring
(
0
,
4
);
String
month
=
time
.
substring
(
5
,
7
);
String
collName
=
Config
.
collName
+
year
+
"_"
+
month
;
super
.
setCollName
(
collName
);
}
...
...
src/main/java/com/zhiwei/searchhotcrawler/test/HotSearchListTest.java
View file @
ee793daf
...
...
@@ -36,7 +36,7 @@ public class HotSearchListTest{
Mongo
mongoNew
=
new
MongoClient
(
address
,
Arrays
.
asList
(
credentialNew
));
DB
dbNew
=
mongoNew
.
getDB
(
"hot_search_list"
);
Map
<
String
,
String
>
timLine
=
TimeParse
.
getTimeMap
(
"201
9-07-16 00:00:00"
,
"2019-07-16 23:59:59"
,
"HH
"
,
1
);
Map
<
String
,
String
>
timLine
=
TimeParse
.
getTimeMap
(
"201
8-02-01 00:00:00"
,
"2019-04-30 23:59:59"
,
"MM
"
,
1
);
timLine
.
forEach
((
start
,
end
)
->{
...
...
@@ -49,29 +49,52 @@ public class HotSearchListTest{
System
.
out
.
println
(
"collName=========="
+
collName
);
DBCollection
collNew
=
dbNew
.
getCollection
(
collName
);
DBObject
query
=
new
BasicDBObject
(
new
BasicDBObject
(
"time"
,
new
BasicDBObject
(
"$gte"
,
startDate
).
append
(
"$lte"
,
endDate
)));
DBCursor
cur
=
coll
.
find
(
query
);
System
.
out
.
println
(
query
+
"======="
+
cur
.
count
());
List
<
DBObject
>
dataList
=
new
ArrayList
<>();
int
i
=
0
;
while
(
cur
.
hasNext
())
{
DBObject
doc
=
cur
.
next
();
DBObject
zhihu
=
new
BasicDBObject
();
zhihu
.
put
(
"_id"
,
doc
.
get
(
"_id"
));
zhihu
.
put
(
"name"
,
doc
.
get
(
"name"
));
zhihu
.
put
(
"url"
,
doc
.
get
(
"url"
));
zhihu
.
put
(
"count"
,
doc
.
get
(
"count"
));
zhihu
.
put
(
"hot"
,
doc
.
get
(
"hot"
));
zhihu
.
put
(
"day"
,
doc
.
get
(
"day"
));
zhihu
.
put
(
"time"
,
doc
.
get
(
"time"
));
zhihu
.
put
(
"changeCount"
,
doc
.
get
(
"changeCount"
));
zhihu
.
put
(
"rank"
,
doc
.
get
(
"rank"
));
zhihu
.
put
(
"type"
,
HotSearchType
.
微博热搜
.
name
());
collNew
.
save
(
zhihu
);
dataList
.
add
(
zhihu
);
DBObject
countIndexDoc
=
new
BasicDBObject
();
countIndexDoc
.
put
(
"count"
,
-
1
);
DBObject
timeIndexDoc
=
new
BasicDBObject
();
timeIndexDoc
.
put
(
"time"
,
-
1
);
DBObject
rankIndexDoc
=
new
BasicDBObject
();
rankIndexDoc
.
put
(
"rank"
,
-
1
);
DBObject
nameIndexDoc
=
new
BasicDBObject
();
nameIndexDoc
.
put
(
"name"
,
-
1
);
DBObject
typeIndexDoc
=
new
BasicDBObject
();
typeIndexDoc
.
put
(
"type"
,
-
1
);
try
{
collNew
.
createIndex
(
countIndexDoc
,
new
BasicDBObject
(
"name"
,
"count_desc"
));
collNew
.
createIndex
(
timeIndexDoc
,
new
BasicDBObject
(
"name"
,
"time_desc"
));
collNew
.
createIndex
(
rankIndexDoc
,
new
BasicDBObject
(
"name"
,
"rank_desc"
));
collNew
.
createIndex
(
nameIndexDoc
,
new
BasicDBObject
(
"name"
,
"name_desc"
));
collNew
.
createIndex
(
typeIndexDoc
,
new
BasicDBObject
(
"name"
,
"type_desc"
));
}
catch
(
Exception
e
)
{
e
.
printStackTrace
();
}
// DBObject query = new BasicDBObject(new BasicDBObject("time",
// new BasicDBObject("$gte",startDate).append("$lte", endDate)));
// DBCursor cur = coll.find(query);
// System.out.println(query +"======="+ cur.count());
// List<DBObject> dataList = new ArrayList<>();
// int i = 0;
// while(cur.hasNext()) {
// DBObject doc = cur.next();
// DBObject zhihu = new BasicDBObject();
// zhihu.put("_id", doc.get("_id"));
// zhihu.put("name", doc.get("name"));
// zhihu.put("url", doc.get("url"));
// zhihu.put("count", doc.get("count"));
// zhihu.put("hot", doc.get("hot"));
// zhihu.put("day", doc.get("day"));
// zhihu.put("time", doc.get("time"));
// zhihu.put("changeCount", doc.get("changeCount"));
// zhihu.put("rank", doc.get("rank"));
// zhihu.put("type", HotSearchType.微博热搜.name());
//
// collNew.save(zhihu);
// dataList.add(zhihu);
// }
// System.out.println(collName +"数据量大小" +dataList.size());
// cur.close();
// if(!dataList.isEmpty()) {
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment