sup

A curses threads-with-tags style email client

sup-website.git

git clone https://supmua.dev/git/sup-website/

community/pipermail-archives/sup-devel/2013-01.txt (18113B) - raw

      1 From dmishd@gmail.com  Tue Jan  1 18:33:09 2013
      2 From: dmishd@gmail.com (Hamish D)
      3 Date: Tue, 1 Jan 2013 18:33:09 +0000
      4 Subject: [sup-devel] after second heliotrope import,
      5 	new messages not appearing
      6 In-Reply-To: <CAMiZLn1Uv-5Csf7O2kxijbwanVsT78W4puy_ypN0vMr=3mA0Yg@mail.gmail.com>
      7 References: <CAOxvSbextWLEObcKrU-93SzFqnhZ-up7cV6MwUboFgN2TM7GNw@mail.gmail.com>
      8 	<CAMiZLn1Uv-5Csf7O2kxijbwanVsT78W4puy_ypN0vMr=3mA0Yg@mail.gmail.com>
      9 Message-ID: <CAOxvSbcBPGCsL_ZDqMgi6zk2uodnLqrdh89SkzMFg=F8aYdCHA@mail.gmail.com>
     10 
     11 > Your problem may have multiple origins, and I have no idea where you could
     12 > start :
     13 >
     14 > - Did you use the correct directories for messages and index for _both_ the
     15 > import and the server ? (sorry if this sounds stupid)
     16 
     17 Always good to check the basics, but yes, I do use the same directory.
     18 I have scripts for the import and the server and both use the -d
     19 argument to ensure they use the same directory.
     20 
     21 > - Did the index size in heliotrope change ? (as given by a GET to
     22 > /size.json)
     23 
     24 No - it remained the same. (And is about half of the count that sup reports).
     25 
     26 > - Did the messages file size change ?
     27 
     28 It did on one import, but not the most recent.
     29 
     30 > - Are the messages absent in inbox only or from the whole turnsole ?
     31 
     32 All searches that I do stop in May.
     33 
     34 > - As a last resort, could you send the mbox file so that I can try this at
     35 > home, if your mails aren't too personal ?
     36 
     37 It's my work email, with a total of over 100,000 messages. So I'd
     38 prefer not to send it over.
     39 
     40 Any logs that might be worth looking at? While doing the import, I do
     41 get various lines along the line of:
     42 
     43     ; scanned 71, indexed 71, skipped 0 bad and 0 seen messages in
     44 22.1s = 3.2 m/s
     45 
     46 I do use the state argument (-t) to save reimporting old mail. I have
     47 also tried doing a reindex (as I import from multiple mboxes) but to
     48 no avail.
     49 
     50 I could just try another import of the whole lot into a clean directory I guess.
     51 
     52 I'm also happy to type stuff into heliotrope-console if that will help
     53 diagnose stuff.
     54 
     55 Thanks for your suggestions so far
     56 Hamish
     57 
     58 
     59 > On Thu, Dec 27, 2012 at 5:19 PM, Hamish D <dmishd at gmail.com> wrote:
     60 >>
     61 >> Hello
     62 >>
     63 >> I've just been trying to carry on from where I left off with migrating
     64 >> my work email into heliotrope (which I last worked on in May).
     65 >>
     66 >> So I've rsynced the mbox files across to where I will run heliotrope
     67 >> and tried running heliotrope-import to put all the new messages in the
     68 >> index. It appeared to work properly, reporting that messages were
     69 >> being added to the index and the index files themselves have had their
     70 >> timestamps updated. But when I start turnsole, I can't find any email
     71 >> after May.
     72 >>
     73 >> Any ideas on where to start looking to work out what is going on?
     74 >>
     75 >> Hamish
     76 
     77 From matthieu.rakotojaona@gmail.com  Wed Jan  2 20:47:06 2013
     78 From: matthieu.rakotojaona@gmail.com (Matthieu Rakotojaona)
     79 Date: Wed, 2 Jan 2013 21:47:06 +0100
     80 Subject: [sup-devel] after second heliotrope import,
     81 	new messages not appearing
     82 In-Reply-To: <CAOxvSbcBPGCsL_ZDqMgi6zk2uodnLqrdh89SkzMFg=F8aYdCHA@mail.gmail.com>
     83 References: <CAOxvSbextWLEObcKrU-93SzFqnhZ-up7cV6MwUboFgN2TM7GNw@mail.gmail.com>
     84 	<CAMiZLn1Uv-5Csf7O2kxijbwanVsT78W4puy_ypN0vMr=3mA0Yg@mail.gmail.com>
     85 	<CAOxvSbcBPGCsL_ZDqMgi6zk2uodnLqrdh89SkzMFg=F8aYdCHA@mail.gmail.com>
     86 Message-ID: <CAMiZLn2CNtF+9YULy9mXJKi0HVNXh9TJPuypjN4mfKC058HiVQ@mail.gmail.com>
     87 
     88 So, it appears heliotrope tells you your mails are indexed, yet they don't
     89 appear. All I see is `MetaIndex#add_message` returning in the test that
     90 checks if the mail is already in the store. Problem is, the exact same
     91 check is already done before that moment, and if the mail really was
     92 already present, it should be marked as "seen" and not "indexed".
     93 
     94 Another possibility could be `MetaIndex#gen_new_docid` which, for an
     95 unknown reason, returns a wrong value, so it would always overwrite an old
     96 message.
     97 
     98 I'm not really good at debugging efficiently, but I would re-launch the
     99 import after having placed a few "puts" in this `add_message` method, see
    100 if the existence test returns true or not. If not, print the docid, and see
    101 what the message with that docid is. If the message with that docid is the
    102 one you just imported, then there is a problem with how we generate them
    103 (because it was already given to another message).
    104 
    105 
    106 On Tue, Jan 1, 2013 at 7:33 PM, Hamish D <dmishd at gmail.com> wrote:
    107 
    108 > > Your problem may have multiple origins, and I have no idea where you
    109 > could
    110 > > start :
    111 > >
    112 > > - Did you use the correct directories for messages and index for _both_
    113 > the
    114 > > import and the server ? (sorry if this sounds stupid)
    115 >
    116 > Always good to check the basics, but yes, I do use the same directory.
    117 > I have scripts for the import and the server and both use the -d
    118 > argument to ensure they use the same directory.
    119 >
    120 > > - Did the index size in heliotrope change ? (as given by a GET to
    121 > > /size.json)
    122 >
    123 > No - it remained the same. (And is about half of the count that sup
    124 > reports).
    125 >
    126 > > - Did the messages file size change ?
    127 >
    128 > It did on one import, but not the most recent.
    129 >
    130 > > - Are the messages absent in inbox only or from the whole turnsole ?
    131 >
    132 > All searches that I do stop in May.
    133 >
    134 > > - As a last resort, could you send the mbox file so that I can try this
    135 > at
    136 > > home, if your mails aren't too personal ?
    137 >
    138 > It's my work email, with a total of over 100,000 messages. So I'd
    139 > prefer not to send it over.
    140 >
    141 > Any logs that might be worth looking at? While doing the import, I do
    142 > get various lines along the line of:
    143 >
    144 >     ; scanned 71, indexed 71, skipped 0 bad and 0 seen messages in
    145 > 22.1s = 3.2 m/s
    146 >
    147 > I do use the state argument (-t) to save reimporting old mail. I have
    148 > also tried doing a reindex (as I import from multiple mboxes) but to
    149 > no avail.
    150 >
    151 > I could just try another import of the whole lot into a clean directory I
    152 > guess.
    153 >
    154 > I'm also happy to type stuff into heliotrope-console if that will help
    155 > diagnose stuff.
    156 >
    157 > Thanks for your suggestions so far
    158 > Hamish
    159 >
    160 >
    161 > > On Thu, Dec 27, 2012 at 5:19 PM, Hamish D <dmishd at gmail.com> wrote:
    162 > >>
    163 > >> Hello
    164 > >>
    165 > >> I've just been trying to carry on from where I left off with migrating
    166 > >> my work email into heliotrope (which I last worked on in May).
    167 > >>
    168 > >> So I've rsynced the mbox files across to where I will run heliotrope
    169 > >> and tried running heliotrope-import to put all the new messages in the
    170 > >> index. It appeared to work properly, reporting that messages were
    171 > >> being added to the index and the index files themselves have had their
    172 > >> timestamps updated. But when I start turnsole, I can't find any email
    173 > >> after May.
    174 > >>
    175 > >> Any ideas on where to start looking to work out what is going on?
    176 > >>
    177 > >> Hamish
    178 > _______________________________________________
    179 > Sup-devel mailing list
    180 > Sup-devel at rubyforge.org
    181 > http://rubyforge.org/mailman/listinfo/sup-devel
    182 >
    183 
    184 
    185 
    186 -- 
    187 Matthieu RAKOTOJAONA
    188 -------------- next part --------------
    189 An HTML attachment was scrubbed...
    190 URL: <http://rubyforge.org/pipermail/sup-devel/attachments/20130102/ca1aa823/attachment-0001.html>
    191 
    192 From matthieu.rakotojaona@gmail.com  Thu Jan 17 20:15:01 2013
    193 From: matthieu.rakotojaona@gmail.com (Matthieu Rakotojaona)
    194 Date: Thu, 17 Jan 2013 21:15:01 +0100
    195 Subject: [sup-devel] Github organization for sup and heliotrope
    196 Message-ID: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    197 
    198 Hello,
    199 
    200 I've set up a github organization at https://github.com/sup-heliotrope
    201 for managing sup, heliotrope, turnsole and any needed bit in the
    202 future. I've already added Hamish Downer (known as foobacca on github)
    203 as an administrator; he has the rights to do whatever is needed on the
    204 organization. Feel free to give me your github account names and I
    205 will add you to the administrators team. I believe this is doable for
    206 the moment because we are only a few people interested in
    207 sup/heliotrope. If (when !) the popularity grows, we might want to
    208 change the governance model.
    209 
    210 I've been quite hesitant to use github because of the closed model (I
    211 like running my own thing if possible) and wanted to run my own
    212 redmine. However, github offers many advantages (uptime, speed, quick
    213 visibility) and the data isn't that closed : wiki are just plain text
    214 in git that can be exported, and there are more than complete APIs [0]
    215 for issues. However, I'd like to keep the wiki in Hamish's ikiwiki
    216 because I like it much better (but this is just personal preference)
    217 
    218 I took the liberty of creating a repo for sup, which is just a
    219 straight copy of git://gitorious.org/sup/mainline.git. I must admit I
    220 don't follow sup's development that much, so feel free to do what you
    221 want there. I've also set up a repo for sup with master being
    222 William's master branch. I'm quite attracted to the git-flow [1]
    223 model; I propose we follow it, and have therefore created a develop
    224 branch which is just a branch at the same level as master. Please,
    225 feel free to fork develop and code/test over it. I already have a
    226 bunch of propositions to do and will soon put them in the repo (I just
    227 realized my own repo was way too dirty to put it up now)
    228 
    229 William, I know you are alive and well thanks to your tweetstream; If
    230 you read this, I hope you won't take this organization as a treason of
    231 any sort. It's just that the lack of activity on your repo is too
    232 saddening for the development of heliotrope.
    233 
    234 I'm sure we can do it.
    235 
    236 See you soon,
    237 
    238 [0] http://developer.github.com/v3/issues/
    239 [1] http://nvie.com/posts/a-successful-git-branching-model/
    240 
    241 --
    242 Matthieu RAKOTOJAONA
    243 
    244 From eg@gaute.vetsj.com  Thu Jan 17 20:52:01 2013
    245 From: eg@gaute.vetsj.com (Gaute Hope)
    246 Date: Thu, 17 Jan 2013 21:52:01 +0100
    247 Subject: [sup-devel] [sup-talk] Github organization for sup and
    248 	heliotrope
    249 In-Reply-To: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    250 References: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    251 Message-ID: <50F86471.3000900@gaute.vetsj.com>
    252 
    253 Hi,
    254 
    255 I think this is a step in the right direction - given that there has 
    256 been silence on wmorgans part so long.
    257 
    258 Kepping things in one place does work alot better, even if it means 
    259 sacrificing a few features - I think that keeping the wiki along with 
    260 the rest of the organization is the best.  Also github is not linked or 
    261 dependant on any one admin / developer, so I think for this kind of 
    262 project it is ideal - expanding to a proper web site is no biggie with 
    263 github either, making future possible migration even easier.
    264 
    265 I'd like to be part of the organization, but I am probably not going to 
    266 be active enough to be an administrator, github name: gauteh.
    267 
    268 Cheers, Gaute
    269 
    270 On to. 17. jan. 2013 kl. 21.15 +0100, Matthieu Rakotojaona wrote:
    271 > Hello,
    272 >
    273 > I've set up a github organization at https://github.com/sup-heliotrope
    274 > for managing sup, heliotrope, turnsole and any needed bit in the
    275 > future. I've already added Hamish Downer (known as foobacca on github)
    276 > as an administrator; he has the rights to do whatever is needed on the
    277 > organization. Feel free to give me your github account names and I
    278 > will add you to the administrators team. I believe this is doable for
    279 > the moment because we are only a few people interested in
    280 > sup/heliotrope. If (when !) the popularity grows, we might want to
    281 > change the governance model.
    282 >
    283 > I've been quite hesitant to use github because of the closed model (I
    284 > like running my own thing if possible) and wanted to run my own
    285 > redmine. However, github offers many advantages (uptime, speed, quick
    286 > visibility) and the data isn't that closed : wiki are just plain text
    287 > in git that can be exported, and there are more than complete APIs [0]
    288 > for issues. However, I'd like to keep the wiki in Hamish's ikiwiki
    289 > because I like it much better (but this is just personal preference)
    290 >
    291 > I took the liberty of creating a repo for sup, which is just a
    292 > straight copy of git://gitorious.org/sup/mainline.git. I must admit I
    293 > don't follow sup's development that much, so feel free to do what you
    294 > want there. I've also set up a repo for sup with master being
    295 > William's master branch. I'm quite attracted to the git-flow [1]
    296 > model; I propose we follow it, and have therefore created a develop
    297 > branch which is just a branch at the same level as master. Please,
    298 > feel free to fork develop and code/test over it. I already have a
    299 > bunch of propositions to do and will soon put them in the repo (I just
    300 > realized my own repo was way too dirty to put it up now)
    301 >
    302 > William, I know you are alive and well thanks to your tweetstream; If
    303 > you read this, I hope you won't take this organization as a treason of
    304 > any sort. It's just that the lack of activity on your repo is too
    305 > saddening for the development of heliotrope.
    306 >
    307 > I'm sure we can do it.
    308 >
    309 > See you soon,
    310 >
    311 > [0] http://developer.github.com/v3/issues/
    312 > [1] http://nvie.com/posts/a-successful-git-branching-model/
    313 >
    314 > --
    315 > Matthieu RAKOTOJAONA
    316 > _______________________________________________
    317 > sup-talk mailing list
    318 > sup-talk at rubyforge.org
    319 > http://rubyforge.org/mailman/listinfo/sup-talk
    320 
    321 From dmishd@gmail.com  Thu Jan 17 21:09:18 2013
    322 From: dmishd@gmail.com (Hamish D)
    323 Date: Thu, 17 Jan 2013 21:09:18 +0000
    324 Subject: [sup-devel] [sup-talk] Github organization for sup and
    325 	heliotrope
    326 In-Reply-To: <50F86471.3000900@gaute.vetsj.com>
    327 References: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    328 	<50F86471.3000900@gaute.vetsj.com>
    329 Message-ID: <CAOxvSbctHrgaVXcpLVzasxKo8B+qLP6PCNLJBb2=YDFc5HObWA@mail.gmail.com>
    330 
    331 > I'd like to be part of the organization, but I am probably not going to
    332 > be active enough to be an administrator, github name: gauteh.
    333 
    334 OK, I've set up a team called committers (with commit access to all
    335 repos) and added you. Also happy to make you an owner if you want.
    336 
    337 I'm don't have strong views on where the wiki should be, but it would
    338 be nice if people can just go and edit it with minimal fuss. It looks
    339 like you'd have to be given access to edit the github wiki, so a bit
    340 of a higher barrier to entry which I would prefer to not put in
    341 people's way.  But if we change our minds later (or if everyone
    342 prefers it) then I have the pages as markdown, so transferring it
    343 across to the github wiki would be very little work.
    344 
    345 Right, should probably go and dig up various patches I've copied off
    346 the email list that I wasn't sure about and turn them into pull
    347 requests to open up discussion on them.
    348 
    349 Hamish
    350 
    351 From dmishd@gmail.com  Thu Jan 17 23:01:00 2013
    352 From: dmishd@gmail.com (Hamish D)
    353 Date: Thu, 17 Jan 2013 23:01:00 +0000
    354 Subject: [sup-devel] [sup-talk] Github organization for sup and
    355 	heliotrope
    356 In-Reply-To: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    357 References: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    358 Message-ID: <CAOxvSbdYRKwsuuZVuNkuhT-C-JV4nxGr1WQ-0-o59+zbgy04AA@mail.gmail.com>
    359 
    360 > I'm quite attracted to the git-flow [1]
    361 > model; I propose we follow it, and have therefore created a develop
    362 > branch which is just a branch at the same level as master. Please,
    363 > feel free to fork develop and code/test over it. I already have a
    364 > bunch of propositions to do and will soon put them in the repo (I just
    365 > realized my own repo was way too dirty to put it up now)
    366 
    367 I've pushed the next branch from the old sup repo to the develop
    368 branch of the new sup repo. Those are patches I'd previously reviewed
    369 and accepted.
    370 
    371 What is currently on master is what I was thinking about asking
    372 William to release as sup 0.13 but I never quite made that happen.
    373 
    374 There are various other branches in the old git repo aswell. Do we
    375 want to put them all on github for merging in our own sweet time? Or
    376 just review them locally and only push the good ones?
    377 
    378 Hamish
    379 
    380 From dmishd@gmail.com  Thu Jan 17 23:43:01 2013
    381 From: dmishd@gmail.com (Hamish D)
    382 Date: Thu, 17 Jan 2013 23:43:01 +0000
    383 Subject: [sup-devel] [sup-talk] Github organization for sup and
    384 	heliotrope
    385 In-Reply-To: <CAOxvSbctHrgaVXcpLVzasxKo8B+qLP6PCNLJBb2=YDFc5HObWA@mail.gmail.com>
    386 References: <CAMiZLn0QPzDPTxAuR9BcdWv2KPU4Zw-8wCQSSnLrYkq-tGSvow@mail.gmail.com>
    387 	<50F86471.3000900@gaute.vetsj.com>
    388 	<CAOxvSbctHrgaVXcpLVzasxKo8B+qLP6PCNLJBb2=YDFc5HObWA@mail.gmail.com>
    389 Message-ID: <CAOxvSbd=VBvhC6YMN+xEJDj3Jk2rfVEv5HkdUv1_ufrrA96CeA@mail.gmail.com>
    390 
    391 > Right, should probably go and dig up various patches I've copied off
    392 > the email list that I wasn't sure about and turn them into pull
    393 > requests to open up discussion on them.
    394 
    395 I've started doing that with 4 well formatted patches I wasn't
    396 entirely sure about (3 from Gaute) and put my comments on those pull
    397 requests. If I get good answers we can merge them.
    398 
    399 Got a few more diffs I'll have to get to another night.
    400 
    401 Hamish
    402 
    403 From matthieu.rakotojaona@gmail.com  Sat Jan 19 15:01:22 2013
    404 From: matthieu.rakotojaona@gmail.com (Matthieu Rakotojaona)
    405 Date: Sat, 19 Jan 2013 16:01:22 +0100
    406 Subject: [sup-devel] Simplification of heliotrope usage
    407 Message-ID: <CAMiZLn3fX5CbLPBJORzC6YQzaKjrqtR-i4iobWM8KgD9+VrC4g@mail.gmail.com>
    408 
    409 Hello everyone,
    410 
    411 I've simplified the usability of heliotrope through this pull-request :
    412 
    413 https://github.com/sup-heliotrope/heliotrope/pull/1
    414 
    415 Basically, to install heliotrope you need to :
    416 
    417 1. Install ruby
    418 2. git clone the branch
    419 (https://github.com/sup-heliotrope/heliotrope/tree/usability)
    420 3. Install the bundle gem and run it :
    421     $ gem install bundle
    422     $ bundle install
    423 4. run heliotrope :
    424     $ ruby /path/to/bin/heliotrope-server
    425 
    426 This should simplify the management of gems versions (the versions are
    427 checked-in, thus using `bundle install` will install the exact same
    428 version as what I have), the management of paths (everything
    429 heliotrope-related is in $XDG_DATA_HOME/heliotrope, so you don't need
    430 to specify a folder). The other command are also modified, so that you
    431 can simply run
    432 
    433     $ ruby /path/to/bin/heliotrope-import -l <maildir directory>
    434 
    435 to import all your mails from a maildir. The README is updated, so you
    436 can read what's up.
    437 
    438 I'd really like to push it as fast as possible into the develop branch
    439 because it helps a lot. Feel free to use/test it.
    440 
    441 --
    442 Matthieu RAKOTOJAONA
    443