for some SciELO Brazil BHL publications, the pid appears in the doi as is . . .
http://www.scielo.br/scielo.php?script=sci_pdf&pid=S0031-10492002001300001
has doi:
https://doi.org/10.1590/S0031-10492002001300001
however, for:
http://www.scielo.br/scielo.php?script=sci_pdf&pid=S2236-89062016000300517
the doi appears to be:
https://doi.org/10.1590/2236-8906-65/2016
In other words, the current DOI inference does not apply to some SciELO Brazil publications, rendering the doi inference trick unreliable.
fyi @myrmoteras
for some SciELO Brazil BHL publications, the pid appears in the doi as is . . .
http://www.scielo.br/scielo.php?script=sci_pdf&pid=S0031-10492002001300001
has doi:
https://doi.org/10.1590/S0031-10492002001300001
however, for:
http://www.scielo.br/scielo.php?script=sci_pdf&pid=S2236-89062016000300517
the doi appears to be:
https://doi.org/10.1590/2236-8906-65/2016
In other words, the current DOI inference does not apply to some SciELO Brazil publications, rendering the doi inference trick unreliable.
fyi @myrmoteras