kicad-developers team mailing list archive

[Dick Hollenbeck <dick@xxxxxxxxxxx>]

On 05/10/2012 09:14 AM, Lorenzo Marcantonio wrote:
> On Thu, May 10, 2012 at 08:10:33AM -0500, Dick Hollenbeck wrote:
>> Library search paths are on their death bed.  You should not have to search for a library
>> after you know where it is.
>>
>> New designs do away with this concept, and so if anyone steps up to fix the old paradigm,
>> it is only temporary.
> Explain the new design then... 

you mean please explain the new design?

> how do you tell where a lib can be found
> without some kind of search path ? At least *one* directory where to
> search should be supplied in one way or the other (I agree that a whole
> list of paths is not too useful)
>
> Keep in mind the case of exchanging files with other people... (OrCAD
> SDT actually saved whole pathnames in the project... these where *royal*
> PITAs)
>


(lib_table
  (lib (logical meparts)(type dir)(full_uri /tmp/eeschema-lib)(options useVersioning))
  (lib (logical old-project)(type schematic)(full_uri /tmp/old-schematic.sch)(options ""))
  (lib (logical www)(type http)(full_uri http://kicad.org/libs)(options ""))
)


Run the script in /new called

./make-dir-lib-source-test-data.sh


Then single step through LIB_TABLE class by running

new/build$ kdbg ./test_sch_lib_table

Read the doxygen docs generated by building target doxygen-docs in new.

Miguel, can you put these on the new website some place, the entire output of new's make
doxygen docs?


This way Lorenzo can have the explanation he demands in full color.


Thanks,

Dick