[seqfan] Re: Forcing of exact searches?

[Brendan McKay]

Hi Neil,  I'm surprised you mention only one annoying thing about
relying on grep.   Another is that phrases split across lines in the
internal format won't be found.  So any search for a phrase with
two or more words might fail.

There are messy ways to do multiline searches with standard unix
utilities, but the simplest way is to install "pcregrep".

Cheers, Brendan.

On 9/8/19 7:10 am, Neil Sloane wrote:
> The OEIS used to be on punched cards, and Russ has always made it
> possible for me (and a couple of other people) to download a copy of the
> current OEIS
> in punched-card format.   This is a flat ascii file. The file is called
> cat25 (that name has not changed for 25 years, although the file has grown).
>
> The current version (it is now a few weeks out of date) is:
>   $ wc cat25
>   5936314 53306288 408589671 cat25
>
> That means it contains 5936314 (virtual) punched cards.  It is what you see
> if the click on "internal format" for a sequence.
>
> It is nice to have this file, because I can grep in it for any string I
> want to see.
> I can do an exact search for any string.
> The one annoying thing is that names with accents (like Erd****s) have
> been distorted in the file, so it is hard to grep for them. (That is why I
> always try to
> give both the nonascii name (Erd****s) and the ascii name (Erdos) in
> sequence entries.)
>
> Of course, if you don't have a unix system, you are out of luck!
>
>
>
>
> On Fri, Aug 9, 2019 at 12:15 AM Peter Munn <techsubs at pearceneptune.co.uk>
> wrote:
>
>> I don't know how easy it would be to provide, but I offer this other
>> example of when a "raw" search option would be useful. There are 70 to 80
>> sequences that consist only of "1"s and "-1"s and I know no way to make
>> the search distinguish the two values, making it more difficult to find
>> such a sequence, and even more awkward to be sure such a sequence is
>> absent.
>>
>> Best Regards,
>> Peter
>>
>> On Thu, August 8, 2019 9:28 pm, Giovanni Resta wrote:
>>> I did encounter the same problem many times and I tried a few variations
>>> but it seems there is not an "exact" search.
>>>
>>> For example, I was looking  for powerful (A001694) numbers but since
>>> "powerful" is translated to "power" I got thousands of results: I had to
>>> enter a search like 4, 8, 9, 16, 25, 27, 32, 36, 49, 64, 72 to find the
>>> actual sequence.
>>>
>>> A simple search of a fixed string would be very useful.
>>>
>>> Il 07/08/2019 09:57, Hugo Pfoertner ha scritto:
>>>> SeqFans,
>>>>
>>>> is there a method to force "exact" searches for words?
>>>> A search
>>>>
>> https://oeis.org/search?q=%22factorization%22+keyword%3Ahard&sort=&language=&go=Search
>>>> produces 922 hits, because the occurrence of "factor" also produces a -
>>>> in
>>>> my case unwanted - match.
>>>> I can not see that the search hints https://oeis.org/hints.html deal
>>>> with
>>>> this problem.
>>>>
>>>> Hugo
>>>>
>>>> --
>>>> Seqfan Mailing list - http://list.seqfan.eu/
>>> --
>>> Seqfan Mailing list - http://list.seqfan.eu/
>>>
>>
>>
>> --
>> Seqfan Mailing list - http://list.seqfan.eu/
>>
> --
> Seqfan Mailing list - http://list.seqfan.eu/