TODO.old 15 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381
  1. TODO
  2. ====
  3. [NB: I use this as a thought record/scribble, not everything on here
  4. makes sense and/or is actually ever going to get done, so IIWY I
  5. wouldn't use it as gospel for the future of dak or as a TODO
  6. list for random hacking.]
  7. ================================================================================
  8. Others
  9. ------
  10. o drop map-unreleased
  11. o check email only portions of addresses match too, iff the names
  12. don't, helps with the "James Troup <james@nocrew.org>"
  13. vs. "<james@nocrew.org>" case.
  14. o ensure .dsc section/prio match .changes section/prio
  15. o 'dak clean-suites' performance is kind of crap when asked to
  16. remove a lot of files (e.g. 2k or so).
  17. o we don't handle the case where an identical orig.tar.gz is
  18. mentioned in the .changes, but not in unchecked; but should we
  19. care?
  20. o 'dak ls' could do better sanity checking for -g/-G (e.g. not more
  21. than one suite, etc.)
  22. o need to decide on whether we're tying for most errors at once.. if
  23. so (probably) then make sure code doesn't assume variables exist and
  24. either way do something about checking error code of check_dsc and
  25. later functions so we skip later checks if they're bailing.
  26. o [?, wishlist, distant future] RFC2047-ing should be extended to
  27. all headers of mails sent out.
  28. o reject sparc64 binaries in a non '*64*' package.
  29. o queue.py(source_exists): a) we take arguments as parameters that
  30. we could figure out for ourselves (we're part of the Upload class
  31. after all), b) we have this 3rd argument which defaults to "any"
  32. but could in fact be dropped since no one uses it like that.
  33. o 'dak process-unchecked': doesn't handle bin-only NMUs of stuff
  34. still in NEW, BYHAND or ACCEPTED (but not the pool) - not a big
  35. deal, upload can be retried once the source is in the archive, but
  36. still.
  37. o 'dak security-install' should upload sourceful packages first,
  38. otherwise with big packages (e.g. X) and esp. when source is !i386,
  39. half the arches can be uploaded without source, get copied into
  40. queue/unaccepted and promptly rejected.
  41. o cnf[Rm::logfile] is misnamed...
  42. <aj> i'd be kinda inclined to go with insisting the .changes file take
  43. the form ---- BEGIN PGP MESSAGE --- <non -- BEGIN/END lines> --
  44. BEGIN PGP SIG -- END PGP MESSAGE -- with no lines before or after,
  45. and rejecting .changes that didn't match that
  46. o 'dak control-suite' should have a diff mode that accepts diff output!
  47. o 'dak process-unchecked' crashes if run as a user in -n mode when
  48. orig.tar.gz is in queue/new...
  49. <elmo_home> [<random>maybe I should reject debian packages with a non-Debian origin or bugs field</>]
  50. <Kamion> [<random>agreed; dunno what origin does but non-Debian bugs fields would be bad]
  51. o 'dak clean-suites' should make use of select..except select, temporary tables
  52. etc. rather than looping and calling SQL every time so we can do
  53. suite removal sanely (see potato-removal document)
  54. o 'dak rm' will happily include packages in the Cc list that aren't
  55. being removed...
  56. o 'dak rm' doesn't remove udebs when removing the source they build from
  57. o check_dsc_against_db's "delete an entry from files while you're
  58. not looking" habit is Evil and Bad.
  59. o 'dak rm' needs to, when not sending bug close mails, promote Cc: to
  60. To: and send the mail anyways.
  61. o the lockfile (Archive_Maintenance_In_Progress) should probably be in a conf file
  62. o 'dak ls' should cross-check the b.source field and if it's not
  63. null and s.name linked from it != the source given in
  64. -S/--source-and-binary ignore.
  65. o 'dak process-unchecked' should do q-d stuff for faster AA [ryan]
  66. o split the morgue into source and binary so binaries can be purged first!
  67. o per-architecture priorities for things like different arch'es
  68. gcc's, silly BSD libftw, palo, etc.
  69. o use postgres 7.2's built-in stat features to figure out how indices are used etc.
  70. o 'dak init-archive' shouldn't be using location, it should run down suites instead
  71. o by default hamstring dak's mail sending so that it won't send
  72. anything until someone edits a script; it's been used far too
  73. much to send spam atm :(
  74. o $ftpdir/indices isn't created by 'dak init-dir' because it's not in dak.conf
  75. o sanity check depends/recommends/suggests too? in fact for any
  76. empty field?
  77. [minor] 'dak process-accepted's copychanges, copydotdak handling
  78. sucks, the per-suite thing is static for all packages, so work out
  79. in advance dummy.
  80. [dak ls] # filenames ?
  81. [dak ls] # maintainer, component, install date (source only?), fingerprint?
  82. o UrgencyLog stuff should minimize it's bombing out(?)
  83. o Log stuff should open the log file
  84. o 'dak import-users-from-passwd' should sync debadmin and ftpmaster (?)
  85. o <drow> Can't read file.:
  86. /srv/security.debian.org/queue/accepted/accepted/apache-perl_1.3.9-14.1-1.21.20000309-1_sparc.dak.
  87. You assume that the filenames are relative to accepted/, might want
  88. to doc or fix that.
  89. o SecurityQueueBuild doesn't handle the case of foo_3.3woody1 with a
  90. new .orig.tar.gz followed by a foo_3.3potato1 with the same
  91. .orig.tar.gz; 'dak process-unchecked' sees it and copes, but the AA
  92. code doesn't and can't really easily know so the potato AA dir is
  93. left with no .orig.tar.gz copy. doh.
  94. o orig.tar.gz in accepted not handled properly (?)
  95. o 'dak security-install' doesn't include .orig.tar.gz but it should
  96. o permissions (paranoia, group write, etc.) configurability and overhaul
  97. o database.get_location_id should handle the lack of archive_id properly
  98. o the whole versioncmp thing should be documented
  99. o aj's bin nmu changes
  100. o 'dak process-new':
  101. * priority >> optional
  102. * arch != {any,all}
  103. * build-depends wrong (via 'dak compare-suites')
  104. * suid
  105. * conflicts
  106. * notification/stats to admin daily
  107. o trap 'dak examine-package' exiting
  108. o distinguish binary only versus others (neuro)
  109. o cache changes parsed from ordering (careful tho: would be caching
  110. .changes from world writable incoming, not holding)
  111. o dak doesn't recognise binonlyNMUs correctly in terms of telling
  112. who their source is; source-must-exist does, but the info is not
  113. propagated down.
  114. o maintainers file needs overrides
  115. [ change override.maintainer to override.maintainer-from +
  116. override.maintainer-to and have them reference the maintainers
  117. table. Then fix 'dak make-maintainers' to use them and write some scripting
  118. to handle the Santiago situation. ]
  119. o Validate Depends (et al.) [it should match \(\s*(<<|<|<=|=|>=|>|>>)\s*<VERSIONREGEXP>\)]
  120. o Clean up DONE; archive to tar file every 2 weeks, update tar tvzf INDEX file.
  121. o testing-updates suite: if binary-only and version << version in
  122. unstable and source-ver ~= source-ver in testing; then map
  123. unstable -> testing-updates ?
  124. o hooks or configurability for debian specific checks (e.g. check_urgency, auto-building support)
  125. o morgue needs auto-cleaning (?)
  126. o dak stats: two modes, all included, separate
  127. o dak stats: add non-US
  128. o dak stats: add ability to control components, architectures, archives, suites
  129. o dak stats: add key to expand header
  130. ================================================================================
  131. queue/approved
  132. --------------
  133. o What to do with multi-suite uploads? Presumably hold in unapproved
  134. and warn? Or what? Can't accept just for unstable or reject just
  135. from stable.
  136. o Whenever we check for anything in accepted we also need to check in
  137. unapproved.
  138. o non-sourceful uploads should go straight through if they have
  139. source in accepted or the archive.
  140. o security uploads on auric should be pre-approved.
  141. ================================================================================
  142. Less Urgent
  143. -----------
  144. o change utils.copy to try rename() first
  145. o [hard, long term] unchecked -> accepted should go into the db, not
  146. a suite, but similar. this would allow dak to get even faster,
  147. make 'dak ls' more useful, decomplexify specialacceptedautobuild
  148. and generally be more sane. may even be helpful to have e.g. new
  149. in the DB, so that we avoid corner cases like the .orig.tar.gz
  150. disappearing 'cos the package has been entirely removed but was
  151. still on stayofexecution when it entered new.
  152. o Logging [mostly done] (todo: 'dak clean-suites' (hard), .. ?)
  153. o 'dak process-unchecked': the tar extractor class doesn't need to be redone for each package
  154. o reverse of source-must-exist; i.e. binary-for-source-must-not-exist
  155. o REJECT reminders in 'dak clean-queues'.
  156. o 'dak examine-package' should check for conflicts and warn about them visavis priority [rmurray]
  157. o store a list of removed/files versions; also compare against them.
  158. [but be careful about scalability]
  159. o dak examine-package: print_copyright should be a lot more intelligent
  160. @ handle copyright.gz
  161. @ handle copyright.ja and copyright
  162. @ handle (detect at least) symlinks to another package's doc directory
  163. @ handle and/or fall back on source files (?)
  164. o To incorporate from utils:
  165. @ unreject
  166. o auto-purge out-of-date stuff from non-free/contrib so that testing and stuff works
  167. o doogie's binary -> source index
  168. o jt's web stuff, matt's changelog stuff (overlap)
  169. o [Hard] Need to merge non-non-US and non-US DBs.
  170. o Do a checkpc(1)-a-like which sanitizes a config files.
  171. o fix parse_changes()/build_file_list() to sanity check filenames
  172. o safety check and/or rename debs so they match what they should be
  173. o Improve 'dak compare-suites'.
  174. o Need to optimize all the queries by using EXAMINE and building some INDEXs.
  175. [postgresql 7.2 will help here]
  176. o Need to enclose all the setting SQL stuff in transactions (mostly done).
  177. o Need to finish 'dak init-db' (a way to sync dak.conf and the DB)
  178. o Need the ability to rebuild all other tables from dists _or_ pools (in the event of disaster) (?)
  179. o Make the --help and --version options do stuff for all scripts
  180. o 'dak make-maintainers' can't handle whitespace-only lines (for the moment, this is feature)
  181. o generic way of saying isabinary and isadsc. (?)
  182. o s/distribution/suite/g
  183. o --help doesn't work without /etc/dak/dak.conf (or similar) at
  184. least existing.
  185. o rename 'dak compare-suites' (clashes with existing 'dak compare-suites')...
  186. * Harder:
  187. o interrupting of stracing 'dak process-unchecked' causes exceptions errors from apt_inst calls
  188. o dependency checking (esp. stable) (partially done)
  189. o need to do proper rfc822 escaping of from lines (as opposed to s/\.//g)
  190. o Revisit linking of binary->source in install() in dak.
  191. o binary-all should be done on a per-source, per-architecture package
  192. basis to avoid, e.g. the perl-modules problem.
  193. o a source-missing-diff check: if the version has a - in it, and it
  194. is sourceful, it needs orig and diff, e.g. if someone uploads
  195. esound_0.2.22-6, and it is sourceful, and there is no diff ->
  196. REJECT (version has a dash, therefore not debian native.)
  197. o check linking of .tar.gz's to .dsc's.. see proftpd 1.2.1 as an example
  198. o archive needs md5sum'ed regularly, but takes too long to do all
  199. in one go; make progressive or weekly.
  200. o test sig checking stuff in test/ (stupid thing is not modularized due to global abuse)
  201. o when encountering suspicious things (e.g. file tainting) do something more drastic
  202. * Easy:
  203. o suite mapping and component mapping are parsed per changes file,
  204. they should probably be stored in a dictionary created at startup.
  205. o don't stat/md5sum files you have entries for in the DB, moron
  206. boy (Dak.check_source_blah_blah)
  207. o promote changes["changes"] to mandatory in dak.py(dump_vars)
  208. after a month or so (or all .dak files contain in the queue
  209. contain it).
  210. o 'dak rm' should behave better with -a and without -b; see
  211. gcc-defaults removal for an example.
  212. o Reject on misconfigured kernel-package uploads
  213. o utils.extract_component_from_section: main/utils -> main/utils, main rather than utils, main
  214. o Fix 'dak process-unchecked' to warn if run when not in incoming or p-u
  215. o dak should validate multi-suite uploads; only possible valid one
  216. is "stable unstable"
  217. o cron.daily* should change umask (aj sucks)
  218. o byhand stuff should send notification
  219. o 'dak rm' should update the stable changelog [joey]
  220. o update tagdb.dia
  221. * Bizzarre/uncertain:
  222. o drop rather dubious currval stuff (?)
  223. o rationalize os.path.join() usage
  224. o logging: hostname + pid ?
  225. o ANAIS should be done in dak (?)
  226. o Add an 'add' ability to 'dak rm' (? separate prog maybe)
  227. o Handle the case of 1:1.1 which would overwrite 1.1 (?)
  228. o maybe drop -r/--regex in 'dak ls', make it the default and
  229. implement -e/--exact (a la joey's "elmo")
  230. o dsc files are not checked for existence/perms (only an issue if
  231. they're in the .dsc, but not the .changes.. possible?)
  232. * Cleanups & misc:
  233. o db_access' get_files needs to use exceptions not this None, > 0, < 0 return val BS (?)
  234. o The untouchable flag doesn't stop new packages being added to ``untouchable'' suites
  235. ================================================================================
  236. Packaging
  237. ---------
  238. o Fix stuff to look in sensible places for libs and config file in debian package (?)
  239. ================================================================================
  240. --help manpage
  241. -----------------------------------------------------------------------------
  242. check-archive X
  243. check-overrides X X
  244. clean-queues X
  245. clean-suites X X
  246. compare-suites X
  247. control-overrides X X
  248. control-suite X X
  249. cruft-report X
  250. decode-dot-dak X
  251. examine-package X
  252. generate-releases X
  253. import-archive X
  254. import-users-from-passwd X X
  255. init-db X
  256. init-dirs X
  257. ls X X
  258. make-maintainers X X
  259. make-overrides X
  260. make-suite-file-list X
  261. process-accepted X X
  262. process-new X X
  263. process-unchecked X
  264. queue-report X
  265. rm X X
  266. security-install X
  267. stats X
  268. symlink-dists X
  269. ================================================================================
  270. Random useful-at-some-point SQL
  271. -------------------------------
  272. UPDATE files SET last_used = '1980-01-01'
  273. FROM binaries WHERE binaries.architecture = <x>
  274. AND binaries.file = files.id;
  275. DELETE FROM bin_associations
  276. WHERE EXISTS (SELECT id FROM binaries
  277. WHERE architecture = <x>
  278. AND id = bin_associations.bin);
  279. ================================================================================