bibadd... mr, zbl and doi shoul always skip over @unpublished

[pauloney]

They should skip over unpublished material because there is not much material that it can chew on to get any good match or make any checks to see if the match found is good.

For example, try bibzbladd on the record:

@UNPUBLISHED{doucha2016metric,
    author = {Doucha, Michal},
    title = {{Metric topological groups: their metric approximation and metric ultraproducts}},
    arxiv = {1601.07449},
    note = {To appear in Groups, Geometry, and Dynamics},
}

it produces

    zbl = {1322.03034},

as a match, which is non-sense, as one can see:

@Article{zbMATH06342368,
    Author = {Michal {Doucha}},
    Title = {{Scott rank of Polish metric spaces.}},
    FJournal = {{Annals of Pure and Applied Logic}},
    Journal = {{Ann. Pure Appl. Logic}},
    ISSN = {0168-0072},
    Volume = {165},
    Number = {12},
    Pages = {1919--1929},
    Year = {2014},
    Publisher = {Elsevier (North-Holland), Amsterdam},
    Language = {English},
    DOI = {10.1016/j.apal.2014.08.002},
    MSC2010 = {03E15 03C75 03C07 54E35},
    Zbl = {1322.03034}
}

[pauloney]

Some matches are good, like for example in this one here where it finds that the article has been published:

@UNPUBLISHED{1703.06092,
    author = {Schneider, Jakob and Thom, Andreas and Nikolov, Nikolay},
    title = {Some remarks on finitarily approximated groups},
    year = {2017},
    arxiv = {1703.06092},
    note = {To appear in Journal de l'\'Ecole polytechnique -- Math\'ematiques},
    mrnumber = {3749196},
}

which is a much better record:

@article {MR3749196,
    AUTHOR = {Nikolov, Nikolay and Schneider, Jakob and Thom, Andreas},
     TITLE = {Some remarks on finitarily approximable groups},
   JOURNAL = {J. \'Ec. polytech. Math.},
  FJOURNAL = {Journal de l'\'Ecole polytechnique. Math\'ematiques},
    VOLUME = {5},
      YEAR = {2018},
     PAGES = {239--258},
      ISSN = {2429-7100},
   MRCLASS = {20E26 (03C20 20E18 20F69)},
  MRNUMBER = {3749196},
}

In this case the titles are the same and that should serve as a check.

[pauloney]

And here is another one for which bibmradd find a MR record for an upublished article:

@MISC{Bhargava-Shankar-4selmer,
author = {Bhargava, Manjul and Shankar, Arul},
title = {{The average number of elements in the 4-Selmer groups of elliptic curves is 7}},
year = {2013-12-27},
note = {Preprint},
arxiv = {1312.7333},
}

it finds

 mrnumber = {3272925},

which is obviously wrong - showing the need to skip over unpublished OR check to see if the recor obtained is correct.