Automatically add talks to researchseminars.org #37
Conversation
|
|
||
| # If a paper link is available, add it. Not all talks have preprints | ||
| if( talk.get('preprint', '') != '' ): | ||
| talk_payload["paper_link"] = "https://arxiv.org/abs/"+talk.get('preprint') |
There was a problem hiding this comment.
We had an example of somebody using a DOI and a different repository, so this may produce nonsense.
I suggest using something like this.
There was a problem hiding this comment.
Thanks, I thought those would have been filtered before getting to this point. I was wrong.
I assume you mean that we
- use the arxiv link if talk['preprint'] is an arxiv identifier,
- use talk['preprint'] as long as it is a valid URL,
- otherwise add this as a talk comment?
The researchseminars API will fail if this entry is not a proper URL, and we can't guarantee that for talk['preprint'], right?
There was a problem hiding this comment.
We could regex-match a DOI, however I propose that we specifically regex match an arxiv preprint, and ignore anything else.
|
One last remark about the preprint format. The rest looks good 👍 |
|
There may be a complication with LRC because we frequently edit those at a later stage. However right now we'd also create the events as TBA, so I propose to keep track of updating data in research seminars as a follow-up action. |
When adding a talk to researchseminars, it gets assigned a number (just the total talk number as a counter), that we would have to keep track of for editing it. We would have to add it to the yaml most likely, or otherwise search the talk list for a match (based on title,author,preprint?). |
We can also do extra queries (first get all talks, then edit). But let's defer the implementation of that step. |
No description provided.