John Marshall <[EMAIL PROTECTED]> wrote in message
news:61587@palm-dev-forum...
> I would appreciate it if someone could explain to me why they think this
is
> nightmarishly difficult. (I understand that some people find adding the
> annotations *annoying*, but that is a different thing from *difficult*.)
> AFAIK the only nightmare is the random crashes due to missing
declarations,
> and they're entirely detectable at compile time.
It took me half a night to move to segmented code (my app got to around
10,000 lines before it needed segmenting). The three biggest problems I had
were not realizing there was documentation for multigen (more on that in a
moment), figuring out the right way to call multigen (I was using the
excellent Foster book as my guide: it was just my luck that the one major
blemish I've found so far happened to be in the code sample illustrating how
to call multigen) and tracking down the missing declarations. Neither of the
latter was difficult: it just took a couple of late-night hours to figure
out. The multigen invocation was easy to fix and only took a couple of
minutes. The crashes due to missing declarations took a couple of hours to
track down: using `-Wmissing-declarations' could have saved all of that
time.
Also, I didn't realize that there was documentation for multigen until just
this morning when I saw your other post. I downloaded the PRC-Tools over a
56K dial-up connection back in March, and somehow I missed the
documentation. The only PRC-Tools documentation I've ever really needed
outside of the Foster book was the pilrc documentation, and that was
included in the download. If it hadn't been 2 a.m. when I first encountered
the segmenting problems (with a demo due the next afternoon), maybe I would
have been alert enough to go online and look for the multigen documentation.
:)
------> kell
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/tech/support/forums/
