67 lines
1.8 KiB
YAML
67 lines
1.8 KiB
YAML
name: Milovana
|
|
sceneByURL:
|
|
- action: scrapeXPath
|
|
url:
|
|
- milovana.com
|
|
scraper: sceneScraper
|
|
sceneByName:
|
|
action: scrapeXPath
|
|
queryURL: https://milovana.com/forum/search.php?keywords={}&terms=all&author=&fid%5B%5D=25&sc=1&sf=titleonly&sr=topics&sk=t&sd=d&st=0&ch=300&t=0&submit=Search
|
|
scraper: search
|
|
sceneByQueryFragment:
|
|
action: scrapeXPath
|
|
queryURL: "{url}"
|
|
scraper: sceneScraper
|
|
|
|
xPathScrapers:
|
|
sceneScraper:
|
|
common:
|
|
$pagebody: //*[@id="page-body"]/h2[@class="topic-title"]/a
|
|
$postbody: //div[@class="postbody"]//h3[@class="first"]/..
|
|
scene:
|
|
Title:
|
|
selector: $pagebody
|
|
postProcess:
|
|
- replace:
|
|
- regex: (\s*[\[\(].*?[\]\)]\s*)
|
|
with: ""
|
|
Date:
|
|
selector: $postbody//p[@class="author"]//time/@datetime
|
|
postProcess:
|
|
- replace:
|
|
- regex: (.+)(T.*)
|
|
with: $1
|
|
- parseDate: 2006-01-02 #2021-07-25T14:36:40+00:00
|
|
Image:
|
|
selector: $postbody//img[@class="postimage"]/@src
|
|
postProcess:
|
|
- replace:
|
|
- regex: ^./
|
|
with: https://milovana.com/forum/
|
|
Studio:
|
|
Name: $postbody//p[@class="author"]//a[@class="username"]
|
|
URL:
|
|
selector: $pagebody/@href
|
|
postProcess:
|
|
- replace:
|
|
- regex: ^./
|
|
with: https://milovana.com/forum/
|
|
- regex: (http.+)&sid=\w*(.*)
|
|
with: $1$2
|
|
search:
|
|
common:
|
|
$card: //ul[@class="topiclist topics"]/li/dl/dt/div/a
|
|
scene:
|
|
Title: $card
|
|
URL:
|
|
selector: $card/@href
|
|
postProcess:
|
|
- replace:
|
|
- regex: ^./
|
|
with: https://milovana.com/forum/
|
|
driver:
|
|
headers:
|
|
- Key: User-Agent
|
|
Value: stash-scraper/1.0.0
|
|
# Last Updated March 07, 2024
|