Andreas/siyuan
1
0
Fork 0
mirror of https://github.com/siyuan-note/siyuan.git synced 2025-12-17 07:00:12 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

🎨 Limit memory usage of PDF parsing https://github.com/siyuan-note/siyuan/pull/9051

Browse source
This commit is contained in:
Daniel 2023-08-27 17:46:23 +08:00
parent 06f6534402
commit f755daee1e
No known key found for this signature in database
GPG key ID: 86211BA83DF03017
1 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
kernel/model/asset_content.go
View file

@ -846,10 +846,10 @@ func (parser *PdfAssetParser) Parse(absPath string) (ret *AssetParseResult) {
// finally fetch the PDF page text results // finally fetch the PDF page text results
// Note: some workers will process pages faster than other workers depending on the page contents // Note: some workers will process pages faster than other workers depending on the page contents
// the order of returned PDF text pages is random and must be sorted using the pageNo index // the order of returned PDF text pages is random and must be sorted using the pageNo index
pagetext := make([]string, pc.PageCount) pageText := make([]string, pc.PageCount)
for p := 0; p < pc.PageCount; p++ { for p := 0; p < pc.PageCount; p++ {
res := <-results res := <-results
pagetext[res.pageNo] = res.text pageText[res.pageNo] = res.text
if nil != res.err { if nil != res.err {
logging.LogErrorf("convert [%s] of page %d failed: [%s]", tmp, res.pageNo, err) logging.LogErrorf("convert [%s] of page %d failed: [%s]", tmp, res.pageNo, err)
} }
@ -862,7 +862,7 @@ func (parser *PdfAssetParser) Parse(absPath string) (ret *AssetParseResult) {
// loop through ordered PDF text pages and join content for asset parse DB result // loop through ordered PDF text pages and join content for asset parse DB result
content := "" content := ""
for _, pt := range pagetext { for _, pt := range pageText {
content += " " + normalizeNonTxtAssetContent(pt) content += " " + normalizeNonTxtAssetContent(pt)
} }
ret = &AssetParseResult{ ret = &AssetParseResult{