[Scons-users] Implicit cache error after running a successful build

Matias Iturburu matias.iturburu at tallertechnologies.com
Thu Oct 31 16:21:25 EDT 2013


2013/10/31 Bill Deegan <bill at baddogconsulting.com>


> Matias,

>

> Does this happen if you are not running SCons in interactive mode?

> Can you see what depends on the .H file via : scons --tree=prune?

>



Interesting. Never used the --tree option.

I get this:

+-biopp:astyle
+-projects/biopp
| +-build/biopp/src/bio_molecular/bio_molecular.cpp
| +-build/biopp/src/bio_molecular/triplets.cpp
| +-build/biopp/src/rna_structure/rna_structure.cpp
| +-build/biopp/biopp/biopp.h
| +-projects/biopp/SConscript
| +-projects/biopp/bio_molecular.H
| +-projects/biopp/bio_molecular.h
| +-projects/biopp/bio_molecular.hh
| +-projects/biopp/bio_molecular.hpp
| +-projects/biopp/bio_molecular.hxx
| +-projects/biopp/biopp
| | +-projects/biopp/biopp/bio_alphabets_test.H
..... Long list follows

I grepped for *hxx* and it's only present in the qt4 tool, disabling it
does in fact makes the error code go away and prints a cleaned tree.
I would like to keep using the tool and to know what would be triggering
this behavior.
Any Idea?





> -Bill

>

>

> On Thu, Oct 31, 2013 at 11:39 AM, Matias Iturburu <

> matias.iturburu at tallertechnologies.com> wrote:

>

>>

>>

>>

>> 2013/10/31 William Deegan <bill at baddogconsulting.com>

>>

>>> Matias,

>>>

>>> My bet is you have an emitter which is manually adding ".H" somewhere?

>>>

>>

>> Aside from Dirk's qt4 tool there are no emitters in my environment, and

>> AFAIK the tool's emitter does not do that.

>>

>>

>>

>>>

>>> -Bill

>>>

>>> On Oct 31, 2013, at 6:22 AM, Matias Iturburu <

>>> matias.iturburu at tallertechnologies.com> wrote:

>>>

>>>

>>>

>>>

>>> 2013/10/31 Dirk Bächle <tshortik at gmx.de>

>>>

>>>> Hi Matías,

>>>>

>>>>

>>>>

>>> Such emphasis :) . Hi Dirk.

>>>

>>>

>>>> On 30.10.2013 18:48, Matias Iturburu wrote:

>>>>

>>>> Hello list.

>>>> I have a Builder to run Astyle (http://astyle.sourceforge.net/) to my

>>>> builds.

>>>> Thing is recently I've started getting this message when I try running

>>>> it:

>>>>

>>>> [...]

>>>> ipdb> n

>>>> scons: *** [biopp:astyle] Implicit dependency

>>>> `projects/biopp/bio_molecular.H' not found, needed by target `biopp:astyle'.

>>>>

>>>> Which is relevant as the offending file (*

>>>> projects/biopp/bio_molecular.H*) exists in my source but with the

>>>> appropriate lowercase *h *and after following the traceback I don't

>>>> see where the path gets mangled.

>>>>

>>>> So my question is: What is this "implicit dependency" thing, how it

>>>> works and how do I get it to tell me what is happening with my files?

>>>>

>>>>

>>>> "implicit dependency" means that SCons has scanned the source files for

>>>> your "astyle" target(s) and detected that one of them includes the header

>>>> file "bio_molecular.H". If you want to get rid of this error, you should

>>>> search though your sources for the offending line and correct it to

>>>> "bio_molecular.h" (with lower capital "h").

>>>>

>>>>

>>> That was the first thing that came to mind. But there isn't any

>>> occurrence of bio_molecular.H with uppercase H in my source files:

>>>

>>> biopp/$ grep -R bio_molecular.H *

>>> biopp/$ grep -R "bio_molecular\.H" *

>>> biopp/$ grep -R "bio_molecular\.h" *

>>> biopp/bio_molecular/exceptions.h: bio_molecular.h: Molecular Biology

>>> ++ Header file.

>>> biopp/bio_molecular/bio_molecular.h: bio_molecular.h: Molecular

>>> Biology ++ Header file.

>>> biopp/bio_molecular/definitions.h: bio_molecular.h: Molecular Biology

>>> ++ Header file.

>>> biopp/biopp.h: #include "biopp/bio_molecular/bio_molecular.h"

>>> tools/generate_graph.cpp:#include "biopp/bio_molecular/bio_molecular.h"

>>>

>>> I'm not saying that the issue is with scons proper, I'm sure I'm messing

>>> something somewhere. But I need to understand where this process (the

>>> implicit dependency search) is taking place so I can trace where the file

>>> name is getting mangled.

>>> It puzzles me that the astyle subprocess is being run and the "error" is

>>> rising afterwards.

>>>

>>>

>>>

>>>> Best regards,

>>>>

>>>> Dirk

>>>>

>>>>

>>>> _______________________________________________

>>>> Scons-users mailing list

>>>> Scons-users at scons.org

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

>>>>

>>>>

>>>

>>>

>>> --

>>> <http://www.tallertechnologies.com/>

>>> Matías Iturburu

>>> Software Engineer

>>>

>>> *San Lorenzo 47, 3rd Floor, Office 5

>>> Córdoba, Argentina

>>>

>>> *

>>> *Phone: +54 351 4217888 / +54 351 4218211

>>> Skype: matias.iturburu*

>>> _______________________________________________

>>> Scons-users mailing list

>>> Scons-users at scons.org

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

>>>

>>>

>>>

>>> _______________________________________________

>>> Scons-users mailing list

>>> Scons-users at scons.org

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

>>>

>>>

>>

>>

>> --

>> <http://www.tallertechnologies.com>

>> Matías Iturburu

>> Software Engineer

>>

>> *San Lorenzo 47, 3rd Floor, Office 5

>> Córdoba, Argentina

>>

>> *

>> *Phone: +54 351 4217888 / +54 351 4218211

>> Skype: matias.iturburu*

>>

>> _______________________________________________

>> Scons-users mailing list

>> Scons-users at scons.org

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

>>

>>

>

> _______________________________________________

> Scons-users mailing list

> Scons-users at scons.org

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

>

>



--
<http://www.tallertechnologies.com>
Matías Iturburu
Software Engineer

*San Lorenzo 47, 3rd Floor, Office 5
Córdoba, Argentina

*
*Phone: +54 351 4217888 / +54 351 4218211
Skype: matias.iturburu*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://four.pairlist.net/pipermail/scons-users/attachments/20131031/6ae11c53/attachment.html


More information about the Scons-users mailing list