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
d68f3ced
Commit
d68f3ced
authored
Feb 05, 2021
by
chenweitao
Browse files
Options
Browse Files
Download
Plain Diff
Merge branch 'mlbWork' into 'master'
Mlb work See merge request
!74
parents
de80cbe7
529937a4
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
18 additions
and
24 deletions
+18
-24
src/main/java/com/zhiwei/searchhotcrawler/crawler/WeiboTopicCrawler.java
+18
-24
No files found.
src/main/java/com/zhiwei/searchhotcrawler/crawler/WeiboTopicCrawler.java
View file @
d68f3ced
...
...
@@ -132,8 +132,8 @@ public class WeiboTopicCrawler {
*/
public
static
List
<
HotSearchList
>
startCrawlerByPhone
(
Date
date
){
List
<
HotSearchList
>
topicList
=
new
ArrayList
<>();
for
(
int
page
=
1
;
page
<=
6
;
page
++){
String
pageUrl
=
"https://
m.weibo.cn/api/container/getIndex?containerid=231648_-_2
&page="
+
page
;
for
(
int
page
=
1
;
page
<=
3
;
page
++){
String
pageUrl
=
"https://
api.weibo.cn/2/page?st_bottom_bar_new_style_enable=1&c=android&s=34dc160d&from=10A9295010&gsid=_2A25NH7inDeRxGeNH4lUX9ifIzTWIHXVvjUtvrDV6PUJbkdANLRjfkWpNSk7RXJ9vYwBfAr66TNj0zcFmOBPKZDuI&containerid=231648_-_4
&page="
+
page
;
Request
request
=
RequestUtils
.
wrapGet
(
pageUrl
);
String
htmlBody
=
null
;
//重试三次
...
...
@@ -144,7 +144,7 @@ public class WeiboTopicCrawler {
log
.
error
(
"下载榜单列表页面时出现错误,错误为:{}"
,
e
);
continue
;
}
if
(
StringUtils
.
isNotBlank
(
htmlBody
)
&&
htmlBody
.
contains
(
"data"
)
)
{
if
(
StringUtils
.
isNotBlank
(
htmlBody
))
{
topicList
.
addAll
(
parseTopicHtml
(
htmlBody
,
date
));
break
;
}
else
{
...
...
@@ -158,9 +158,12 @@ public class WeiboTopicCrawler {
private
static
List
<
HotSearchList
>
parseTopicHtml
(
String
htmlBody
,
Date
date
)
{
try
{
JSONArray
jsonArray
=
JSONObject
.
parseObject
(
htmlBody
).
getJSON
Object
(
"data"
).
getJSON
Array
(
"cards"
);
JSONArray
jsonArray
=
JSONObject
.
parseObject
(
htmlBody
).
getJSONArray
(
"cards"
);
if
(
Objects
.
nonNull
(
jsonArray
)
&&
!
jsonArray
.
isEmpty
())
{
JSONArray
cards
=
jsonArray
.
getJSONObject
(
0
).
getJSONArray
(
"card_group"
);
for
(
int
j
=
0
;
j
<
jsonArray
.
size
();
j
++){
JSONObject
card
=
jsonArray
.
getJSONObject
(
j
);
if
(
card
.
containsKey
(
"card_group"
)){
JSONArray
cards
=
card
.
getJSONArray
(
"card_group"
);
List
<
HotSearchList
>
topicList
=
new
ArrayList
<>();
Integer
rank
=
null
;
String
topicName
=
null
;
...
...
@@ -184,32 +187,23 @@ public class WeiboTopicCrawler {
try
{
commentNum
=
TipsUtils
.
getHotCount
(
commentNumStr
);
readNum
=
TipsUtils
.
getHotCount
(
readNumStr
);
// if(commentNumStr.contains("万")){
// commentNumStr = commentNumStr.replaceAll("万", "");
// commentNum = (int)(Double.parseDouble(commentNumStr)*10000);
// }else if(commentNumStr.contains("亿")){
// commentNumStr = commentNumStr.replaceAll("亿", "");
// commentNum = (int)(Double.parseDouble(commentNumStr)*10000000);
// }else{
// commentNum = Integer.getInteger(commentNumStr);
// }
//
// if(readNumStr.contains("万")){
// readNumStr = readNumStr.replaceAll("万", "");
// readNum = (int)(Double.parseDouble(readNumStr)*10000);
// }else if(readNumStr.contains("亿")){
// readNumStr = readNumStr.replaceAll("亿", "");
// readNum = (int)(Double.parseDouble(readNumStr)*10000000);
// }else{
// readNum = Integer.getInteger(readNumStr);
// }
}
catch
(
Exception
e
){
e
.
printStackTrace
();
}
HotSearchList
topic
=
new
HotSearchList
(
url
,
topicName
,
readNum
,
rank
,
HotSearchType
.
微博话题
.
name
(),
commentNum
,
description
,
date
);
if
(
cardGroup
.
containsKey
(
"title_flag_pic"
)){
String
titlePic
=
cardGroup
.
getString
(
"title_flag_pic"
);
if
(
titlePic
.
contains
(
"new"
)){
topic
.
setIcon
(
"新"
);
}
else
if
(
titlePic
.
contains
(
"hot"
)){
topic
.
setIcon
(
"热"
);
}
}
topicList
.
add
(
topic
);
}
return
topicList
;
}
}
}
else
{
// log.info("html:{}",htmlBody);
}
...
...
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