httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Christian Gross" <cgr...@eusoft.com>
Subject Re: Strcasecmp problems
Date Fri, 02 Aug 1996 15:07:53 GMT
----------
> From: Ben Laurie <ben@gonzo.ben.algroup.co.uk>
> To: new-httpd@hyperreal.com
> Subject: Re: Strcasecmp problems
> Date: vendredi, 2. ao{t 1996 15:34
> 
> Christian Gross wrote:
> > 
> > ----------
> > > From: Ben Laurie <ben@gonzo.ben.algroup.co.uk>
> > > To: new-httpd@hyperreal.com
> > > Subject: Re: Strcasecmp problems
> > > Date: vendredi, 2. ao{t 1996 14:31
> > > 
> > > Christian Gross wrote:
> > > > 
> > > > Hi,
> > > > 
> > > > I have several problems...
> > > > 
> > > > The problems that I am having relate to my port to a threaded
model
> > and
> > > > Windows NT.
> > > 
> > > When do we get to see the source for this?
> > It depends on what kind of source you want to see...  Currently Apache
for
> > NT is a bit of a mess.  That means if I was to release the source
today,
> > it would not compile, because it uses parts that are specific to my
> > solution.  Those parts I will not release.
> 
> Are these parts essential to making it work, or are they added extras?
> 
No, there are added extras, but there is a set of calls that are part of a
lower level debug framework that I have developed.  You do not need this
if you compile using a release option, but this bit of defines needs to
weeded out.

> > However, if you are willing to
> > accept this fact I would be happy to release the code this weekend. 
So
> > people tell me you want the source this weekend????   As well, this
source
> > is very specific to NT and the first thing any of you will say is
"what a
> > goof ball"  The reason for my delay is that I need to keep my paying
> > client happy first.
> > 
> > I wanted to spend some time cleaning up the code, but again if all you
> > deem it unncessary I would be more than happy to release the code and
then
> > assist in the cleaning up procedure.
> 
> If we are going to support it as an "official" option for Apache then
the code
> has to compile under both Unix and NT. My interest is actually in a
Windows95
> version - can you see a problem with that?
> 
Currently Apache does not work properly under Win95.  Again this part of
the problem.  Yes I would like to see it as an official release, but I
would like it to be threaded....  For me threading is "the way" to go. 
Keeps things cleaner and simpler.  I have access to a NT, 95 and Solaris
2.5 station so I would be willing to help support those platforms.

> Its difficult to say whether we need cleaned up code or not without
seeing it,
> so I suppose I'd like to get a look at a version...
> 
Ok, I will set it up on my web server
ftp://ftp.eusoft.com

then log on as apache using apache as password and it should be there
soon.  Please be aware there is a lot to clean and we will need to
communicate.  
> Which version of Apache did you base it on?
> 
I based it on Apache 1.1 last beta 4.
> Perhaps a summary of what needed changing would be informative.
> 
To be honest, I do not remember anymore, because I simply hacked it and
then went on from there.  For me I needed to quickly solve a problem and
later will come the 
> > 
> > > > 
> > > > I do not know if point 1 and 2 are related, but maybe someone can
help
> > me
> > > > on this one.
> > > > 
> > > > 3) I have a GPF very consistently that breaks my entire program,
in
> > the
> > > > routine strcasecmp.  Could someone explain to me what it does
exactly
> > and
> > > > how it is used.  The values of p at time of fault is NULL and q
seems
> > > > correct.  Did I stumble on something.  Note that this bug happens
only
> > > > sometimes, but very consistently.
> > > 
> > > strcasecmp is a synonym for stricmp ... it should _never_ be passed
a
> > NULL
> > > pointer. We need to know where it comes from...
> > > 
> > The null pointer comes from the table routines.  It thinks that there
is
> > an element, but in fact there is none.  I saw it happen from
init_mime.
> 
> Sounds like a bug!
> 
Don't know, lets look at my code and then take it from there.
> > 
> > BTW something I want to know is some of you seem to be able to get
Apache
> > 1.2 code and some of you submit code.  Please tell me how this
works...  I
> > use a version control system myself, but its SourceSafe or PVCS.
> 
> Anyone can get 1.2 code from ftp://www.apache.org/httpd/from-cvs. Only
those
> with commit privilege can submit code. The version control system is
CVS. If
> anyone else wants patches applied they have to persuade one of the core
members
> to do it - a somewhat unreliable process at the moment but hopefully (if
we
> ever get a bug-reporting system in place) becoming more streamlined...
> 
Okay I am puzzled how do I get CVS to work?

> The snapshots are updated every 6 hours.
> 
Ok

The source should be available in about 1 hour.  Check back then. 
Remember its RAW....


-- 
Christian Gross
euSOFT
Phone 41.1.492.7827
Fax 41.1.492.7757

Mime
View raw message