bug#17687: enhancement request: contiguous lines

[al1chevee]

Dear All,

Hope this mail finds you well and that its content raises some interest if it 
can help the grep users community.

Given that I use very often grep on large files, I realized how useful the 
option '-max-line' can be. Nevertheless, I found that the following enhancement 
could be of interest for faster processing and hence develop the use of grep:

enhancement concept: in addition to the current --max-count=num option which 
takes a numerical input, one could imagine another possibility (like 
--max-count="contiguous") which would only retrieve contiguous matching lines 
since the first match.

advantages: the current --max-count option improves the speed of the search 
with an early stop (after n matches) but it assumes the user knows exactly the 
number of lines to look into. With the "contiguous" option, the search provides 
all contiguous searches and then stops.

remark: this enhancement does not aim to compete with a "database" approach 
especially for time-serie management, but still many large files can have for 
example timestamps and using this option would accelerate the searches in 
sorted files thanks to the early stop.

Again, if my enhancement request is redundant with a former one already 
proposed or does not fit in the grep approach, please forgive me.

Have a nice day and thanks again for the great work.

Alain