How would I grep "words" with same prefix within a line of text.

[sr320]

For example I have

100s of lines such as

Emma Timmins-Schiffman, Samuel J. White, Rhonda Elliott Thompson, Brent Vadopalas, Benoit Eudeline, Brook L. Nunn & Steven B. Roberts. (**2021**) [Coupled microbiome analyses highlights relative functional roles of bacteria in a bivalve hatchery](https://doi.org/10.1186/s40793-021-00376-z) Environmental Microbiome 16, 7 doi:10.1186/s40793-021-00376-z <span data-badge-popover="right" data-badge-type="2" data-doi="10.1186/s40793-021-00376-z" data-hide-no-mentions="true" class="altmetric-embed"></span>

Cristian Gallardo-Escárate, Valentina Valenzuela-Muñoz, Gustavo Nuñez-Acuña, Diego Valenzuela-Miranda, Ana Teresa Gonçalves, Hugo Escobar-Sepulveda, Ivan Liachko, Bradley Nelson, Steven Roberts & Wesley Warren. (**2021**) [Chromosome-scale genome assembly of the sea louse Caligus rogercresseyi by SMRT sequencing and Hi-C analysis](https://doi.org/10.1038/s41597-021-00842-w) Sci Data 8, 60 doi:10.1038/s41597-021-00842-w <span data-badge-popover="right" data-badge-type="2" data-doi="10.1038/s41597-021-00842-w" data-hide-no-mentions="true" class="altmetric-embed"></span>

and I want to just scrape dois.

Noting all have the prefix doi:

[sr320]

I am thinking ...

[kubu4]

You want to just pull out the doi:10.1186/s40793-021-00376-z, right?

[kubu4]

I found a crazily complex regex solution:

/\b(10[.][0-9]{4,}(?:[.][0-9]+)*/(?:(?!["&\'<>])\S)+)\b/g

That works (I tested it out with your examples above, using regxr.

It will pull 10.1186/s40793-021-00376-z and ignore the doi:.

Then, in theory, you should be able to pop that into awk to parse it out using some designated fields (e.g. : and spaces), but I'm having trouble getting awk to run the regex...

[sr320]

Nice Thanks! Related I found bbedit has a way to live play around with expressions- https://d.pr/i/eRc4Mo