4 [NB: I use this as a thought record/scribble, not everything on here
5 makes sense and/or is actually ever going to get done, so IIWY I
6 wouldn't use it as gospel for the future of dak or as a TODO
7 list for random hacking.]
9 ================================================================================
14 o 'dak check-overrides' should remove the src-only override when a
15 binary+source override exists
17 o reject on > or < in a version constraint
19 o 'dak reject-proposed-updates' should only start an editor once to
20 capture a message; it will usually be the same message for all
21 files on the same command line.
23 23:07 < aba> elmo: and, how about enhancing 'dak cruft-report' to spot half-dropped
24 binaries on one arch (i.e. package used to build A and B, but B is
25 no longer built on some archs)?
31 o check email only portions of addresses match too, iff the names
32 don't, helps with the "James Troup <james@nocrew.org>"
33 vs. "<james@nocrew.org>" case.
35 o ensure .dsc section/prio match .changes section/prio
37 o 'dak clean-suites' performance is kind of crap when asked to
38 remove a lot of files (e.g. 2k or so).
40 o we don't handle the case where an identical orig.tar.gz is
41 mentioned in the .changes, but not in unchecked; but should we
44 o 'dak ls' could do better sanity checking for -g/-G (e.g. not more
47 o use python2.2-tarfile (once it's in stable?) to check orig.tar.gz
50 o need to decide on whether we're tying for most errors at once.. if
51 so (probably) then make sure code doesn't assume variables exist and
52 either way do something about checking error code of check_dsc and
53 later functions so we skip later checks if they're bailing.
55 o the .dak stuff is fundamentally braindamaged, it's not versioned
56 so there's no way to change the format, yay me. need to fix.
57 probably by putting a version var as the first thing and checking
58 that.. auto-upgrade at least from original format would be good.
59 might also be a good idea to put everything in one big dict after
62 o [?, wishlist, distant future] RFC2047-ing should be extended to
63 all headers of mails sent out.
65 o reject sparc64 binaries in a non '*64*' package.
67 o queue.py(source_exists): a) we take arguments as parameters that
68 we could figure out for ourselves (we're part of the Upload class
69 after all), b) we have this 3rd argument which defaults to "any"
70 but could in fact be dropped since no one uses it like that.
72 o 'dak process-unchecked': doesn't handle bin-only NMUs of stuff
73 still in NEW, BYHAND or ACCEPTED (but not the pool) - not a big
74 deal, upload can be retried once the source is in the archive, but
77 o security global mail overrides should special case buildd stuff so
78 that buildds get ACCEPTED mails (or maybe 'dak security-install' (?)), that way
79 upload-security doesn't grow boundlessly.
81 o 'dak security-install' should upload sourceful packages first,
82 otherwise with big packages (e.g. X) and esp. when source is !i386,
83 half the arches can be uploaded without source, get copied into
84 queue/unaccepted and promptly rejected.
86 o 'dak cruft-report's NVIU check doesn't catch cases where source
87 package changed name, should check binaries
88 too. [debian-devel@l.d.o, 2004-02-03]
90 o cnf[Rm::logfile] is misnamed...
92 <aj> i'd be kinda inclined to go with insisting the .changes file take
93 the form ---- BEGIN PGP MESSAGE --- <non -- BEGIN/END lines> --
94 BEGIN PGP SIG -- END PGP MESSAGE -- with no lines before or after,
95 and rejecting .changes that didn't match that
97 o 'dak cruft-report' should check for source packages not building any binaries
99 o 'dak control-suite' should have a diff mode that accepts diff output!
101 o 'dak clean-proposed-updates' doesn't deal with 'dak rm'-d
102 packages, partial replacements etc. and more.
104 o 'dak reject-proposed-updates' blindly deletes with no check that
105 the delete failed which it might well given we only look for
106 package/version, not package/version _in p-u_. duh.
108 o 'dak rm' should remove obsolete changes when removing from p-u, or
109 at least warn. or 'dak reject-proposed-updates' should handle it.
111 o need a testsuite _badly_
113 o 'dak process-unchecked' crashes if run as a user in -n mode when
114 orig.tar.gz is in queue/new...
116 <elmo_home> [<random>maybe I should reject debian packages with a non-Debian origin or bugs field</>]
117 <Kamion> [<random>agreed; dunno what origin does but non-Debian bugs fields would be bad]
119 o 'dak clean-suites' should make use of select..except select, temporary tables
120 etc. rather than looping and calling SQL every time so we can do
121 suite removal sanely (see potato-removal document)
123 o 'dak rm' will happily include packages in the Cc list that aren't
126 o 'dak rm' doesn't remove udebs when removing the source they build from
128 o check_dsc_against_db's "delete an entry from files while you're
129 not looking" habit is Evil and Bad.
131 o 'dak process-new' allows you to edit the section and change the
132 component, but really shouldn't.
134 o 'dak rm' needs to, when not sending bug close mails, promote Cc: to
135 To: and send the mail anyways.
137 o the lockfile (Archive_Maintenance_In_Progress) should probably be in a conf file
139 o 'dak ls' should cross-check the b.source field and if it's not
140 null and s.name linked from it != the source given in
141 -S/--source-and-binary ignore.
143 o 'dak reject-proposed-updates' sucks; it should a) only spam d-i
144 for sourceful rejections, b) sort stuff so it rejects sourceful
145 stuff first. the non-sourceful should probably get a form mail, c)
146 automate the non-sourceful stuff (see b).
148 o 'dak process-unchecked' should do q-d stuff for faster AA [ryan]
150 o split the morgue into source and binary so binaries can be purged first!
152 o per-architecture priorities for things like different arch'es
153 gcc's, silly BSD libftw, palo, etc.
155 o use postgres 7.2's built-in stat features to figure out how indices are used etc.
157 o 'dak init-archive' shouldn't be using location, it should run down suites instead
159 o 'dak clean-proposed-updates' needs to know about udebs
161 o by default hamstring dak's mail sending so that it won't send
162 anything until someone edits a script; it's been used far too
163 much to send spam atm :(
165 o $ftpdir/indices isn't created by 'dak init-dir' because it's not in dak.conf
167 o sanity check depends/recommends/suggests too? in fact for any
170 [minor] 'dak process-accepted's copychanges, copydotdak handling
171 sucks, the per-suite thing is static for all packages, so work out
174 [dak ls] # filenames ?
175 [dak ls] # maintainer, component, install date (source only?), fingerprint?
177 o UrgencyLog stuff should minimize it's bombing out(?)
178 o Log stuff should open the log file
180 o 'dak queue-report' should footnote the actual notes, and also *
181 the versions with notes so we can see new versions since being
184 o 'dak queue-report' should have alternative sorting options, including reverse
185 and without or without differentiaion.
187 o 'dak import-users-from-passwd' should sync debadmin and ftpmaster (?)
189 o <drow> Can't read file.:
190 /org/security.debian.org/queue/accepted/accepted/apache-perl_1.3.9-14.1-1.21.20000309-1_sparc.dak.
191 You assume that the filenames are relative to accepted/, might want
194 o <neuro> the orig was in NEW, the changes that caused it to be NEW
195 were pulled out in -2, and we end up with no orig in the archive
198 o SecurityQueueBuild doesn't handle the case of foo_3.3woody1 with a
199 new .orig.tar.gz followed by a foo_3.3potato1 with the same
200 .orig.tar.gz; 'dak process-unchecked' sees it and copes, but the AA
201 code doesn't and can't really easily know so the potato AA dir is
202 left with no .orig.tar.gz copy. doh.
204 o orig.tar.gz in accepted not handled properly (?)
206 o 'dak security-install' doesn't include .orig.tar.gz but it should
208 o permissions (paranoia, group write, etc.) configurability and overhaul
210 o remember duplicate copyrights in 'dak process-new' and skip them, per package
212 o <M>ove option for 'dak process-new' byhand proecessing
214 o 'dak cruft-report' could do with overrides
216 o database.get_location_id should handle the lack of archive_id properly
218 o the whole versioncmp thing should be documented
220 o 'dak process-new' doesn't do the right thing with -2 and -1 uploads, as you can
221 end up with the .orig.tar.gz not in the pool
223 o 'dak process-new' exits if you check twice (aj)
225 o 'dak process-new' doesn't trap signals from 'dak examine-package' properly
227 o queued and/or perl on sparc stable sucks - reimplement it.
229 o aj's bin nmu changes
232 * priority >> optional
234 * build-depends wrong (via 'dak compare-suites')
237 * notification/stats to admin daily
238 o trap 'dak examine-package' exiting
239 o distinguish binary only versus others (neuro)
241 o cache changes parsed from ordering (careful tho: would be caching
242 .changes from world writable incoming, not holding)
244 o dak doesn't recognise binonlyNMUs correctly in terms of telling
245 who their source is; source-must-exist does, but the info is not
248 o Fix BTS vs. dak sync issues by queueing(via BSMTP) BTS mail so
249 that it can be released on deman (e.g. ETRN to exim).
251 o maintainers file needs overrides
253 [ change override.maintainer to override.maintainer-from +
254 override.maintainer-to and have them reference the maintainers
255 table. Then fix 'dak make-maintainers' to use them and write some scripting
256 to handle the Santiago situation. ]
258 o Validate Depends (et al.) [it should match \(\s*(<<|<|<=|=|>=|>|>>)\s*<VERSIONREGEXP>\)]
260 o Clean up DONE; archive to tar file every 2 weeks, update tar tvzf INDEX file.
262 o testing-updates suite: if binary-only and version << version in
263 unstable and source-ver ~= source-ver in testing; then map
264 unstable -> testing-updates ?
266 o hooks or configurability for debian specific checks (e.g. check_urgency, auto-building support)
268 o morgue needs auto-cleaning (?)
270 o dak stats: two modes, all included, seperate
271 o dak stats: add non-US
272 o dak stats: add ability to control components, architectures, archives, suites
273 o dak stats: add key to expand header
275 ================================================================================
280 o What to do with multi-suite uploads? Presumably hold in unapproved
281 and warn? Or what? Can't accept just for unstable or reject just
284 o Whenever we check for anything in accepted we also need to check in
287 o non-sourceful uploads should go straight through if they have
288 source in accepted or the archive.
290 o security uploads on auric should be pre-approved.
292 ================================================================================
297 o change utils.copy to try rename() first
299 o [hard, long term] unchecked -> accepted should go into the db, not
300 a suite, but similar. this would allow dak to get even faster,
301 make 'dak ls' more useful, decomplexify specialacceptedautobuild
302 and generally be more sane. may even be helpful to have e.g. new
303 in the DB, so that we avoid corner cases like the .orig.tar.gz
304 disappearing 'cos the package has been entirely removed but was
305 still on stayofexecution when it entered new.
307 o Logging [mostly done] (todo: 'dak clean-suites' (hard), .. ?)
309 o 'dak process-unchecked': the tar extractor class doesn't need to be redone for each package
311 o reverse of source-must-exist; i.e. binary-for-source-must-not-exist
312 o REJECT reminders in 'dak clean-queues'.
313 o 'dak examine-package' should check for conflicts and warn about them visavis priority [rmurray]
314 o store a list of removed/files versions; also compare against them.
315 [but be careful about scalability]
317 o dak examine-package: print_copyright should be a lot more intelligent
318 @ handle copyright.gz
319 @ handle copyright.ja and copyright
320 @ handle (detect at least) symlinks to another package's doc directory
321 @ handle and/or fall back on source files (?)
323 o To incorporate from utils:
326 o auto-purge out-of-date stuff from non-free/contrib so that testing and stuff works
327 o doogie's binary -> source index
328 o jt's web stuff, matt's changelog stuff (overlap)
330 o [Hard] Need to merge non-non-US and non-US DBs.
332 o experimental needs to auto clean (relative to unstable) [partial:
333 'dak cruft-report' warns about this]
335 o Do a checkpc(1)-a-like which sanitizes a config files.
336 o fix parse_changes()/build_file_list() to sanity check filenames
337 o saftey check and/or rename debs so they match what they should be
339 o Improve 'dak compare-suites'.
340 o Need to optimize all the queries by using EXAMINE and building some INDEXs.
341 [postgresql 7.2 will help here]
342 o Need to enclose all the setting SQL stuff in transactions (mostly done).
343 o Need to finish 'dak init-db' (a way to sync dak.conf and the DB)
344 o Need the ability to rebuild all other tables from dists _or_ pools (in the event of disaster) (?)
345 o Make the --help and --version options do stuff for all scripts
347 o 'dak make-maintainers' can't handle whitespace-only lines (for the moment, this is feature)
349 o generic way of saying isabinary and isadsc. (?)
351 o s/distribution/suite/g
354 @ weekly postins to d-c (?)
355 @ backup of report (?)
356 @ backup of changes.tgz (?)
358 o --help doesn't work without /etc/dak/dak.conf (or similar) at
361 o rename 'dak compare-suites' (clashes with existing 'dak compare-suites')...
365 o interrupting of stracing 'dak process-unchecked' causes exceptions errors from apt_inst calls
366 o dependency checking (esp. stable) (partially done)
367 o override checks sucks; it needs to track changes made by the
368 maintainer and pass them onto ftpmaster instead of warning the
370 o need to do proper rfc822 escaping of from lines (as opposed to s/\.//g)
371 o Revisit linking of binary->source in install() in dak.
372 o Fix component handling in overrides (aj)
373 o Fix lack of entires in source overrides (aj)
374 o direport misreports things as section 'devel' (? we don't use direport)
375 o vrfy check of every Maintainer+Changed-By address; valid for 3 months.
376 o binary-all should be done on a per-source, per-architecture package
377 basis to avoid, e.g. the perl-modules problem.
378 o a source-missing-diff check: if the version has a - in it, and it
379 is sourceful, it needs orig and diff, e.g. if someone uploads
380 esound_0.2.22-6, and it is sourceful, and there is no diff ->
381 REJECT (version has a dash, therefore not debian native.)
382 o check linking of .tar.gz's to .dsc's.. see proftpd 1.2.1 as an example
383 o archive needs md5sum'ed regularly, but takes too long to do all
384 in one go; make progressive or weekly.
385 o something needs to clear out .changes files from p-u when
386 removing stuff superseded by newer versions. [but for now we have
387 'dak clean-proposed-updates']
388 o test sig checking stuff in test/ (stupid thing is not modularized due to global abuse)
389 o when encountering suspicous things (e.g. file tainting) do something more drastic
393 o suite mapping and component mapping are parsed per changes file,
394 they should probably be stored in a dictionary created at startup.
395 o don't stat/md5sum files you have entries for in the DB, moron
396 boy (Dak.check_source_blah_blah)
397 o promote changes["changes"] to mandatory in dak.py(dump_vars)
398 after a month or so (or all .dak files contain in the queue
400 o 'dak rm' should behave better with -a and without -b; see
401 gcc-defaults removal for an example.
402 o Reject on misconfigured kernel-package uploads
403 o utils.extract_component_from_section: main/utils -> main/utils, main rather than utils, main
404 o Fix 'dak process-unchecked' to warn if run when not in incoming or p-u
405 o dak should validate multi-suite uploads; only possible valid one
407 o cron.daily* should change umask (aj sucks)
408 o 'dak cruft-report' doesn't look at debian-installer but should.
409 o 'dak cruft-report' needs to check for binary-less source packages.
410 o 'dak cruft-report' could accept a suite argument (?)
411 o byhand stuff should send notification
412 o 'dak poolize' should udpate db; move files, not the other way around [neuro]
413 o 'dak rm' should update the stable changelog [joey]
418 o drop rather dubious currval stuff (?)
419 o rationalize os.path.join() usage
420 o 'dak cruft-report' also doesn't seem to warn about missing binary packages (??)
421 o logging: hostname + pid ?
422 o ANAIS should be done in dak (?)
423 o Add an 'add' ability to 'dak rm' (? separate prog maybe)
424 o Replicate old dinstall report stuff (? needed ?)
425 o Handle the case of 1:1.1 which would overwrite 1.1 (?)
426 o maybe drop -r/--regex in 'dak ls', make it the default and
427 implement -e/--exact (a la joey's "elmo")
428 o dsc files are not checked for existence/perms (only an issue if
429 they're in the .dsc, but not the .changes.. possible?)
433 o db_access' get_files needs to use exceptions not this None, > 0, < 0 return val BS (?)
434 o The untouchable flag doesn't stop new packages being added to ``untouchable'' suites
436 ================================================================================
441 o Fix stuff to look in sensible places for libs and config file in debian package (?)
443 ================================================================================
446 -----------------------------------------------------------------------------
449 check-proposed-updates X
450 clean-proposed-updates X
454 control-overrides X X
461 import-users-from-passwd X X
467 make-suite-file-list X
479 ================================================================================
481 Random useful-at-some-point SQL
482 -------------------------------
484 UPDATE files SET last_used = '1980-01-01'
485 FROM binaries WHERE binaries.architecture = <x>
486 AND binaries.file = files.id;
488 DELETE FROM bin_associations
489 WHERE EXISTS (SELECT id FROM binaries
490 WHERE architecture = <x>
491 AND id = bin_associations.bin);
493 ================================================================================