Re: Puzzled at store_client->seen_offset(copy_offset).

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Fri, 26 Apr 2002 20:25:56 +0200

The old code uses this when processing HTTP headers. If the full header isn't
yet available, the caller (the one calling storeClientCopy) indicates that
the data it got wasn't sufficient and that more is data needed before he can
process the HTTP header. When more data is available, the old plus the new
data will be sent to the caller via the callback registered in
storeClientCopy.

The new code works differently. Now the caller has to buffer the data he
could not process himself. storeClientCopy will only return the new data.

Regards
Henrik

maer727@sohu.com wrote:
> Thanks, Henrik pal!
>
> What means the following in you reply?
>
>
> They differ in the case where the data available is not sufficient to
> process it. In such case seen_offset is greater than copy_offset.
>
> I search it in clientHandleIMSReply, client_cide.c and find sometimes
> the value are different.
>
> But the case you mentioned puzzled me, I do not understand your case.
> Can you give me a more detailed descrption on your case?
>
> Bset regards,
> George Ma
>
> ----- 原文 -----
> From: Henrik Nordstrom
> To: maer727@sohu.com
> Cc: squid-dev@squid-cache.org
> Subject: Re: =?gb2312?q?=BB=D8=B8=B4=3ARe=3A=20Puzzled=20at?=
> store_client->seen_offset(copy_offset). Sent: Wed Apr 24 14:25:44 CST 2002
>
> > Valid enoug reason to look at the 2.4 code..
> >
> > Both are from the storeClientCopy() call.
> >
> > seen_offset == I (the one calling storeClientCopy) have already seen up
> > to this point in the object. Don't bother me until there is more data
> > available.
> >
> > copy_offset == I want data starting at this offset in the object.
> >
> > They differ in the case where the data available is not sufficient to
> > process it. In such case seen_offset is greater than copy_offset.
> >
> > Regards
> > Henrik
> >
> > maer727@sohu.com wrote:
> > > Thanks, Henrik pal!
> > >
> > > I think it is no use to understand seen_offset but
> > > what is the meaning and function of copy_offset? :-)
> > >
> > > Btw: my work now in my lab is to develop a system which is based
> > > on squid-2.4, so if it does not matter very much, can you give me
> > > a simple explanation about the meaning and function of the variable?
> > > :-) I am reading the source of a content delivery system which is based
> > > on squid-2.4. Our aim is to adapt it to freeware. Now it is not a
> > > freeware and just used in the lab. You know, reading earlier version of
> > > software is a boring thing but I have to do it. :-(
> > >
> > > I have skip as much as possible in earlier versions of Squid but I
> > > think I can not skip seen_offset, without it I can not understand the
> > > memory format of the store_client.
> > >
> > > I am a newbie of Squid and our system. I will thank again for your help
> > > these days.
> > >
> > > Best regards,
> > > George Ma
> > > ----- 原文 -----
> > > From: Henrik Nordstrom
> > > To: maer727@sohu.com ;squid-dev@squid-cache.org
> > > Subject: Re: Puzzled at store_client->seen_offset(copy_offset).
> > > Sent: Wed Apr 24 00:05:42 CST 2002
> > >
> > > > seen_offset no longer exists in Squid-HEAD. I see no reason to
> > > > explain this odd and error prone design which we have already decided
> > > > to abandon.
> > > >
> > > > Regards
> > > > Henrik
> > > >
> > > > maer727@sohu.com wrote:
> > > > > Hi, pals!
> > > > >
> > > > > What is the meaning and function of store_client->seen_offset?
> > > > > I search store_client->seen_offset is source files. I find
> > > > > only store.c and stat.c use the variable. But I failed to find
> > > > > the meaning and function of the variable.
> > > > >
> > > > > Can you give me a simple explanation about the meaning and
> > > > > fucntion of store_client->seen_offset and
> > > > > store_client->copy_offset? What is the difference between
> > > > > store_client->seen_offset and store_ client->copy_offset?
> > > > >
> > > > > Best regards,
> > > > > George Ma
Received on Fri Apr 26 2002 - 12:26:03 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:15:21 MST