Surprisingly, there is evidence that nicotine really does improve performance on information processing tasks. For example. the paper Selective effects of nicotine on attentional processes (Psychopharmacology 146(2), 1999) begins as follows:
in reply to Re: coding under the influence
in thread coding under the influence
There is strong evidence that nicotine facilitates some
types of information processing (Edwards et al. 1985;
Warburton 1990; Sherwood 1993; Heishman et al. 1994),and that this effect does not result from a reversal effect
of a withdrawal-induced deficit (Wesnes and Warburton
1984; Le Houezec et al. 1994; Warburton and Arnall
1994; Foulds et al. 1996). However, the specific nature
of the nicotine-induced improvements on information
processing has proved more difficult to specify.
The paper goes on to describe a study that, in the authors' view, supports the hypothesis that nicotine achieves these benefits by increasing the intensity of attention.
Of course I wouldn't recommend smoking, because of the well-known risks, but it's not impossible that it would help your coding in the short-to-medium term.
Disclosure: I am a nicotine addict. I don't smoke though; I use snus (which appears to be the safest tobacco product around, but is of course not without risk).