This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author terry.reedy
Recipients apalala, ezio.melotti, mrabarnett, rhettinger, serhiy.storchaka, terry.reedy
Date 2020-01-04.06:19:57
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1578118797.61.0.533435098719.issue39165@roundup.psfhosted.org>
In-reply-to
Content
I currently agree with Serhiy and am not persuaded that either new function is needed.

1. findalliter: For 3.0, I might have been more aggressive than we were in turning list-returning functions into iterators, as done with map, etc.  But the collective decision, led by Guido, was that for some, such as str.split, the efficiency gain was not worth the disruption.  The thought was that returned lists were typically 'small';  they are definitely finite (whereas map(func, infinite_iterator) is itself an infinite iterator).  It is too late to reverse that decision, and I would be wary of adding iterator versions.  There is no particular justification for this one.

2. findfirst: This seems to be a near duplicate of search.  

I don't know why people use findall(...)[0] instead of search (which I learned first), but I don't see 'lack of symmetry and completeness' as a reason. 'python-ideas' is a mailing list, not a person or even a defined group of people.  The respondents on any thread tend to be a biased selection of the community.  Such discussions suggest but do not determine.
History
Date User Action Args
2020-01-04 06:19:57terry.reedysetrecipients: + terry.reedy, rhettinger, ezio.melotti, mrabarnett, serhiy.storchaka, apalala
2020-01-04 06:19:57terry.reedysetmessageid: <1578118797.61.0.533435098719.issue39165@roundup.psfhosted.org>
2020-01-04 06:19:57terry.reedylinkissue39165 messages
2020-01-04 06:19:57terry.reedycreate