eaglexmw/novel_dev_team_novel
1
0
Fork 0
mirror of https://gitee.com/novel_dev_team/novel synced 2026-07-20 17:52:45 +08:00
Code Issues Packages Projects Releases Wiki Activity

爬虫优化

Browse Source
This commit is contained in:
xiongxiaoyang
2019-11-15 09:13:22 +08:00
parent 5f7c29ce26
commit 6a051a3c9f
2 changed files with 7 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
novel-admin/src/main/java/com/java2nb/books/service/impl/BookCrawlServiceImpl.java
View File

@@ -152,7 +152,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
parseBiquTaBook(bookPatten, forObject, i, baseUrl);
while (currentPage < totalPage) {
if(isInteruptBiquTaCrawl){
break;
return;
}
catBookListUrl = catBookListUrlBase + i + "/" + (currentPage + 1) + ".html";
@@ -196,7 +196,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
System.out.println("匹配分数" + scoreFind);
while (isFind && scoreFind && isBookNameMatch && isFindAuthor) {
if(isInteruptBiquTaCrawl){
break;
return;
}
try {
@@ -274,7 +274,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
while (isFindIndex) {
if(isInteruptBiquTaCrawl){
break;
return;
}
if (!hasIndexNum.contains(indexNum)) {
@@ -371,7 +371,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
while (currentPage < totalPage) {
if(isInteruptBiquDaoCrawl){
break;
return;
}
catBookListUrl = catBookListUrlBase + i + "/" + (currentPage + 1) + ".html";
@@ -419,7 +419,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
try {
if(isInteruptBiquDaoCrawl){
break;
return;
}
@@ -497,7 +497,7 @@ public class BookCrawlServiceImpl implements BookCrawlService {
while (isFindIndex) {
if(isInteruptBiquDaoCrawl){
break;
return;
}
if (!hasIndexNum.contains(indexNum)) {