[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: very obscure problem, help needed
From: |
Neil Jerram |
Subject: |
Re: very obscure problem, help needed |
Date: |
Sun, 8 Oct 2000 23:46:54 +0100 |
David Pirotte writes:
I just hope that some folks will take few minutes to see if they woud
have encounterred the same problem in the past and help me.
No solution, I'm afraid, just a question for clarification and a
suggestion for how to proceed further.
===> I get several "Content-Type: text/html" written as the beginning
of my html pages, just as if the guile script was called several
times:
exactly 1 "Content-Type: text/html"
per call to date-in-future
Does the rest of the script generate any HTML output? How exactly do
the multiple Content-Types look - do you mean
--------start----------
Content-Type: text/html
Content-Type: text/html
--------stop----------
or
--------start----------
Content-Type: text/html
Content-Type: text/html
--------stop----------
or
--------start----------
Content-Type: text/html
<HTML>
...
</HTML>
Content-Type: text/html
<HTML>
...
</HTML>
--------stop----------
NOW: if I momentaraly define date-in-future to always reply the same integer,
just to try to understand what is going on, everything is fine!
[...]
its too complicated to pass the all date engine software,
but I checked that the function 'open-day?' itself refering
to a 'binary-search' function are not involved (forcing them
to temporarily always return #t)
In a weird case like this, I think this is a good approach. You know
that `(define (date-in-future . args) 98)' does not produce multiple
Content-Types, but that the real code for date-in-future, but with
`(define (open-day? . args) #t)', does. So I would suggest
progressively simplifying date-in-future, from the real code to a
fixed integer, until the problem disappears.
One other idea - are you by any chance doing anything with strings
that you got back from Postgres? I found that it is dangerous to
treat such strings as shareable. Specifically, I found that
(use-modules (database postgres))
(use-modules (ice-9 string-fun))
...
(sans-surrounding-whitespace (pg-getvalue dataset row col))
would cause mysterious errors later in the program.
But if I use a private copy of sans-surrounding-whitespace, in which
`make-shared-substring' is replaced by `substring', the program
behaves correctly.
Good luck!
Neil