go-shiori/shiori
1
1
Fork 0
mirror of https://github.com/go-shiori/shiori.git synced 2025-02-22 23:14:02 +08:00
Code Issues Projects Releases Packages Wiki Activity

Fix: some CSS resources not loaded

Browse source
This commit is contained in:
Radhi Fadlillah 2019-06-10 08:44:47 +07:00
parent 2077d8fa2c
commit a794dd8ea2
2 changed files with 21 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
pkg/warc/internal/archiver/processor.go
View file

@ -404,6 +404,7 @@ func processCSS(input io.Reader, baseURL *nurl.URL) (string, []ResourceURL) {
// Save the CSS URL and replace it with archival URL
res := ToResourceURL(cssURL, baseURL)
if res.ArchivalURL == "" {
buffer.Write(bt)
continue
}

22
pkg/warc/internal/archiver/resource-url.go
View file

@ -31,12 +31,30 @@ func ToResourceURL(uri string, base *nurl.URL) ResourceURL {
return ResourceURL{}
}
// Create archive URL
// Create download URL
downloadURL := toAbsoluteURI(uri, base)
downloadURL = rxTrailingSlash.ReplaceAllString(downloadURL, "")
downloadURL = strings.ReplaceAll(downloadURL, " ", "+")
archivalURL := strings.Replace(downloadURL, "://", "/", 1)
// Create archival URL
archivalURL := downloadURL
// Some URL have its query escaped.
// For example, Wikipedia's stylesheet looks like this :
// load.php?lang=en&modules=ext.3d.styles%7Cext.cite.styles%7Cext.uls.interlanguage
// However, when browser download it, it will be registered as unescaped query :
// load.php?lang=en&modules=ext.3d.styles|ext.cite.styles|ext.uls.interlanguage
// So, for archival URL, we need to unescape the query first.
tmp, err := nurl.Parse(downloadURL)
if err == nil {
newQuery, _ := nurl.QueryUnescape(tmp.RawQuery)
if newQuery != "" {
tmp.RawQuery = newQuery
archivalURL = tmp.String()
}
}
archivalURL = strings.Replace(archivalURL, "://", "/", 1)
archivalURL = strings.ReplaceAll(archivalURL, "?", "-")
archivalURL = strings.ReplaceAll(archivalURL, "#", "-")
archivalURL = strings.ReplaceAll(archivalURL, "/", "-")