TODO 3.1 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394
  1. TODO
  2. ====
  3. Various
  4. -------
  5. * Improve logging in ``dak control-suite``. ``--add`` is silent, not nice. same
  6. for --remove
  7. * Throw away all DD uploaded .debs. (Depend on "Lintian based automated
  8. rejects")
  9. - Need a way to define a build-architecture for arch_all debs. Some of
  10. them can only be built on certain architectures.
  11. A control file header build-architecture: YXY should do it.
  12. - It's a suite option, not active for all at once.
  13. - Should have all buildd machines under dsa control
  14. * Check TODO.old and move still-valid/useful entries over here.
  15. * Need to improve test suite coverage.
  16. * database table "binaries" contains a column 'type TEXT NOT
  17. NULL'. This should be made a FK on override_type, as it only contains
  18. deb/udeb strings.
  19. - sql query to do the db work for it:
  20. ALTER TABLE binaries ADD COLUMN new_type INT4 REFERENCES override_type(id);
  21. UPDATE BINARIES SET new_type = 7 WHERE type = 'deb';
  22. UPDATE BINARIES SET new_type = 8 WHERE type = 'udeb';
  23. ALTER TABLE binaries DROP COLUMN type;
  24. ALTER TABLE binaries RENAME COLUMN new_type TO type;
  25. - needs updateX.py written and then the rest of the code changed to deal
  26. with it.
  27. * use pythonX.Y-tarfile to check orig.tar.gz timestamps too.
  28. * security global mail overrides should special case buildd stuff so
  29. that buildds get ACCEPTED mails. Or maybe send them at
  30. new-security-install time. That way upload-security doesn't grow
  31. boundlessly.
  32. * debianqueued sucks. Reimplement in a sane way.
  33. NEW processing
  34. --------------
  35. * 'dak process-new' allows you to edit the section and change the
  36. component, but really shouldn't allow the component change.
  37. * 'dak process-new' should have a "Previous package" option.
  38. * 'dak process-new' should have an option to skip (un-)reviewed packages
  39. * 'dak process-new' should have an option to skip older versions of an upload
  40. * 'dak process-new' doesn't do the right thing with -2 and -1 uploads,
  41. as you can end up with the .orig.tar.gz not in the pool or belonging
  42. to a wrong suite.
  43. * 'dak process-new' doesn't trap signals from 'dak examine-package' properly
  44. * 'dak queue-report' should footnote the actual notes, and also * the
  45. versions with notes so we can see new versions since being noted...
  46. * <neuro> the orig was in NEW, the changes that caused it to be NEW
  47. were pulled out in -2, and we end up with no orig in the archive :(
  48. Override handling
  49. -----------------
  50. * 'dak check-overrides' should remove the src-only override when a
  51. binary+source override exists
  52. * override checks sucks; it needs to track changes made by the
  53. maintainer and pass them onto ftpmaster instead of warning the maintainer.
  54. * Fix component handling in overrides
  55. Cruft
  56. -----
  57. * 'dak cruft-report' could do with overrides
  58. * cruft-report could spot "half-dropped" binaries. Like if a package
  59. used to build A and B, but B is no longer built for half the
  60. architectures.
  61. * cruft-report's NVIU check doesn't catch cases where source package
  62. changed name, should check binaries too. [debian-devel@l.d.o,
  63. 2004-02-03]
  64. * 'dak cruft-report' doesn't look at debian-installer but should.