Better fuzzy search for IDs #5
Description
Mainly this is probably for DOIs that can be input/searched in multiple forms:
- 10.1234/5678
- doi:10.1234/5678
- http://dx.doi.org/10.1234/5678
- https://doi.org/10.1234/5678
As well as probably other forms not considered above. Therefore, there probably need to be a generic approach to how you determine a search includes an ID and how to use the bits you identify as a URL to look for a match in the search index. From the other side maybe the parts of the ID need to be better indexed However, if you index in parts they you need extra configuration to say expand find all bits of the ID. Then what happens if you find part of one ID and another part of a second. It look like a matching results but it really isn't.