Hi Karen.
Thanks for your review. Comments inline.
Karen Tung wrote:
> Hi Jack,
>
>
> - You wrote the "find()" function in python to replace the "find"
> command for
> fixing 5802. That seemed to be kinda complicated. Can you explain why
> we need all the complicated logic, instead of just use os.walk()?
I generated a find command which can be used for more than just DC,
being that it is functionality which is applicable all over the place.
It can deliver files only , directories only, or all pathnames.
Can it be simplified for DC? Yes. DC needs only "all pathnames." If I
strip out the other functionality, we'll save only 19 lines though.
Since these 19 lines buy us extra functionality which may be useful in
the future, why not keep them in? This function is a general-purpose
utility.
What sparked this conversation was when you showed me in person (being
office-mates has its advantages :) ) a case where my find didn't work
like the find(1) command. My find didn't print links to directories
like find(1) does. I traced this back to os.walk itself, which lists
these pathnames in the returned subdirs list but then doesn't iterate on
them. I now list them in the "all pathnames" case and now both my
finds and the find command return the same files.
>
> - In bootroot_archive.py, the find() function or os.walk() will return
> the list of files/directories as a list.
> You wrote those list to temporary files, and then, just open the file
> and read
> in the list, when you are processing the list of files to fiocompress
> as well
> as when you process the list of files to overwrite. I think we can
> skip the
> temporary files and just manipulate the lists directly.
You are correctamundo! Fixed.
>
> - The changes in DC-manifest.rng, DC_defs.py, and ai_sparc_image.xml
> look fine to me.
Copyrights updated and webrev respun.
- Against slim_source:
http://cr.opensolaris.org/~schwartz/081220.1/webrev/
- Incremental since last webrev:
http://cr.opensolaris.org/~schwartz/081220.1/webrev.incr.1.2/
I'll figure out and correct the (duplicate) list of bugs under each file
before pushing.
I'll do a sanity retest and wait for your blessing before pushing.
Thanks,
Jack
>
> Thanks,
>
> --Karen
>
> Jack Schwartz wrote:
>> Hi everyone.
>>
>> Please review fixes for adding to the DC manifest a way of not
>> fiocompressing bootroot files, and an implementation for find using
>> python.
>>
>> 5821 Sparc AI microroot is not fully writable
>> 5802 system calls to find should be investigated to see if an os.walk
>> would be better
>>
>> http://cr.opensolaris.org/~schwartz/081220.1/webrev/
>>
>> I've built and booted an X86 iso to regression test. For SPARC
>> testing, I've only booted the sparc bootroot and verified that it
>> boots. I need someone on the AI team to verify the image more
>> thoroughly.
>>
>> Internally, image is available at
>> /home/schwartz/gates/ss_fiocompress/automated_installer_image_sparc.iso
>>
>> Thanks,
>> Jack
>> _______________________________________________
>> caiman-discuss mailing list
>> caiman-discuss at opensolaris.org
>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
>>
>
