Re: [bug] DEC[+-]something in infocmp -i

[Alain Bench]

Hi Thomas,

 On Monday, March 20, 2006 at 15:22:17 -0500, Thomas E. Dickey wrote:

> On Mon, 20 Mar 2006, Alain Bench wrote:
>> only the 1st digit of the number counts.
> thanks (will fix)

    The same problem touches other sequences, like:

 · "\E[10m" (select default charset) is analysed as "{SGR:+BOLD}".
 · "\E[0m" (sgr0 with explicit 0) is analysed as "{ISO DEC G0}m".


    While at init/reset strings, some more small glitches:

 -2) "tput init" never sends is3. Instead it sends a second time is2.

 -3) "tput reset" runs iprog. This makes sense, but is not documented in
man/terminfo.tail nor man/tput.1.

 -4) man/tput.1 should say exactly what happens in which order during
init and reset, or point to "man 5 terminfo".

 -5) man/terminfo.tail clearly explains the init case, but the part
about reset is much less clear: Is iprog run? Are mgc, smgl, smgr, tbc,
and hts also used? It says "A pair of sequences" talking about 4. And
the fallback to "i*" strings is not described.


Bye!    Alain.
-- 
How to Report Bugs Effectively
<URL:http://www.chiark.greenend.org.uk/~sgtatham/bugs.html>