Re: web updates that work with binary patches?
From: Jim McKay (no email)
Date: 02/27/05
- Next message: Herbert Sitz: "Re: web updates that work with binary patches?"
- Previous message: Ivan Pastine: "Re: looking for a text diff component/code"
- In reply to: Herbert Sitz: "Re: web updates that work with binary patches?"
- Next in thread: Herbert Sitz: "Re: web updates that work with binary patches?"
- Reply: Herbert Sitz: "Re: web updates that work with binary patches?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 27 Feb 2005 09:32:27 -0800
Herbert Sitz wrote:
>> I think most of the web update tools would support this. Perhaps
>>some would
>> only replace the program and then restart it, but obviously this
>>is more limited.
>>
>
>That's exactly distinction I'd like to find out between web updater
>programs. Which support running the binary patch on the client to
>create the new .exe (and/or modify files ancillary to .exe.). If a
>program supports only downloading of new file to replace old file, is
>it possible to intercept things on client end and link to code of my
>own that will run the binary patch. I'd prefer solution where the
>patch updating on client is applied automatically by the web updater
>component, but being able to control in code in client would be
>second-best solution.
...
>> I can see that one would be able to check what version is n the
>>client and what version the patch patches from and to, but what is
>>the normal logic
>for
>> this . . . does the server host patches for every combination of
>>from and to, i.e. 1.0 to 1.1, 1.0 to 1.2, and 1.1 to 1.2 and
>>download and apply the corrrect one? Or do you successively patch
>>1.0 to 1.1 to 1.2 if a client with 1.0 'discovers' that 1.2 is
>>available?
>>
>
>That's another thing I'd like to find out more about. I assume some
>of the web update products support this multiple version scenarios
>with regular files (not binary patches). Do they also support with
>binary patches?
>
>> Also, FTR, I don't see why it would be hard to write something
>>like this
>for
>> oneself using something like RO or kbmmw if custom flexibility is
>required,
>> but since these web update components are so cheap it is
>>presumably not worth the bother and often better to use something
>>tried and tested.
>>
>
>I was thinking the same myself. (As you know I use kbmMW.) But most
>of the web updater programs are pretty cheap (< $100 ), so if any of
>them offer special features that would take me a while to write from
>scratch then I'd probably just go with the out-of-box solution.
hmmm... I wonder if I'm missing something about this.
What advantage is there w/a patcher as opposed to using
a PlugIn system? W/kbmMW, I use plugin as a service, and
it works very well. All these issues Lauchlan brings up
are handled cleanly w/PlugIn.
This is one nice thing RO has out of the box w/Hydra.
Still, not so difficult building similar capability w/kbmMW.
I ended up settling on DragonSoft's tool, BTW.
-- Regards: Jim McKay "My theory of evolution is that Darwin was adopted." -- Steven Wright Posted with XanaNews: Ver: 1.17.2.7
- Next message: Herbert Sitz: "Re: web updates that work with binary patches?"
- Previous message: Ivan Pastine: "Re: looking for a text diff component/code"
- In reply to: Herbert Sitz: "Re: web updates that work with binary patches?"
- Next in thread: Herbert Sitz: "Re: web updates that work with binary patches?"
- Reply: Herbert Sitz: "Re: web updates that work with binary patches?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
