[Scons-users] sconsign db access -- different results from Dir() and File() apis

[Jay Damask]

Dirk Bächle <tshortik <at> gmx.de> writes:


> 

> 

>     Hi Jay,

>       On 03.06.2013 21:50, Jay Damask wrote:

> 

>     

>     ...

>     

>     by calling "sconsign()", you're lazy initializing the stored

>     .sconsign.dblite DB and then access the info that's stored in it for

>     the file entry "epa.proto".

>       

>               

>     ...

>

>     Here, you are creating a new File Node and initialize its ninfo

>     structure. But it doesn't get filled with info, this would happen

>     later during the actual build process in:

>       next_task

>         make_ready

>           make_ready_current

>             visited

>     . I hope this makes things clearer to you. Can you write a little

>     more about what you are actually trying to accomplish? Reaching into

>     the SCons DB and hacking around with ninfos/binfo looks a bit like

>     the "wrong path" to do things. ;)

>     Best regards,

>     Dirk

> 

> 

> _______________________________________________

> Scons-users mailing list

> Scons-users <at> scons.org

> http://four.pairlist.net/mailman/listinfo/scons-users

> 



Hi Dirk, 

Don't know if this reply will get posted.. still figuring out this system. 

Thank you for your reply. I understand the difference, although it wasn't 
clear to be just in reading the code. So, thanks, saves me a lot of time. 

Since I wrote I see that once I have a Node.FS object I can call .get_csig() 
on it to get the current on-disk md5sum, so that solves that pb. 

Both my questions related to protoc, a Google compiler that is called to 
build .pb.h and .pb.cc files from .proto files. 

My questions may help others so what I'm going to do is post another, fresh 
question on this topic. 

I would still like to know how to access the sconsign db from the api, 
clearly scons is. 

Thank you so much,
-Jay