#409 closed defect (fixed)
Google Books translator broken after site update
| Reported by: | dstillman | Owned by: | stakats |
|---|---|---|---|
| Priority: | critical | Milestone: | 1.0 Beta 3 |
| Component: | ingester | Version: | 1.0 |
| Keywords: | Cc: | stakats |
Description
Assigning to Simon, but whoever has time to fix this should Accept it.
Change History (6)
comment:1 Changed 10 years ago by dcohen
comment:2 Changed 10 years ago by stakats
- Owner changed from simon to stakats
- Status changed from new to assigned
comment:3 Changed 10 years ago by stakats
I now have a working translator for single books that pulls COinS metadata from WorldCat. Two problems remain:
1) some Google Books do not include the "Find in a Library" link to WorldCat. Not sure why this is the case, but see for example, Dena Goodman's Marie Antoinette: Writings in the Body of a Queen. In cases like this, we can either fall back to page scraping or simply fail gracefully.
2) importing items from the search results page requires a double jump, first to the individual book page to generate the WorldCat link and then a second request to pull the WorldCat pages. I am stumped here about how to work around our apparent limitation to have only one asynchronous operation in a translator, since the only way to resume from Zotero.wait() is by calling Zotero.done(), which stops the translator completely.
comment:4 Changed 10 years ago by stakats
- Resolution set to fixed
- Status changed from assigned to closed
comment:5 Changed 10 years ago by stakats
OCLC WorldCat links were not a reliable source of metadata. I have contacted Google Books about exposing their metadata more transparently. For now we will continue to page scrape as before.
Looks like there is a standard link on each Google Books page, "Find this book in a library," which goes to the WorldCat/OCLC canonical page for that ISBN, which gives you the bibliographical info.