README_maintainer 10 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270
  1. # -*- mode:org -*-
  2. #+TITLE: Maintainer tasks
  3. #+STARTUP: noindent
  4. This document describes the tasks the Org-mode maintainer has to do
  5. and how they are performed.
  6. * Git workflow
  7. The git repository has two branches:
  8. - master :: for current development.
  9. - maint :: for bug fixes against latest major or minor release.
  10. Bug fixes always go on =maint= then are merged on =master=.
  11. New features always go on =master=.
  12. * Releasing
  13. ** Major release
  14. The release number for main releases look like this: =7.13=
  15. Main releases are made whenever Org is in a state where the feature
  16. set is consistent and we feel that the features that are implemented
  17. is something we want to support in the future.
  18. A major release turns the current state of the master branch into a
  19. release.
  20. When doing a /major release/, make sure all changes from the maint
  21. branch are merged into the the master branch, then merge the master
  22. branch back into maint to synchronize the two.
  23. ** Minor release
  24. The release number for minor releases look like this: =7.13.01=
  25. Minor releases are small amends to main releases. Usually they fix
  26. critical bugs discovered in a main release. Minor bugs are usually
  27. not fixed -- they will be adressed in the next main release.
  28. Only the fix to the bug is bundled into a release, without the main
  29. development work going on in the master branch. Since the bug fix
  30. will also be needed in the master branch, usually the fix is made in
  31. maint then merged in master.
  32. ** Tagging the release
  33. When doing a major and a minor release, after all necessary merging
  34. is done, tag the _maint_ branch for the release with:
  35. git tag -a "Adding release tag" release_7.9.1
  36. and push tags with
  37. git push --tags
  38. ** Uploading the release files from the orgmode.org server
  39. Log on the orgmode.org server as the emacs user and cd to
  40. ~/git/org-mode
  41. From there do
  42. make release
  43. make upload
  44. to create the .tar.gz and .zip files, the documentation, and to
  45. upload everything at the right place.
  46. * Working with patchwork
  47. John Wiegley is running a patchwork server that looks at the
  48. emacs-orgmode mailing list and extracts patches. The maintainer and
  49. his helpers should work through such patches, give feedback on them
  50. and apply the ones which are good and done. A task for the maintainer
  51. is to every now and then try to get old stuff out of that list, by
  52. asking some helpers to investigate the patch, by rejecting or
  53. accepting it.
  54. I have found that the best workflow for this is using the pw script by
  55. Nate Case, with the modifications for Org-mode made by John Wiegley
  56. and Carsten Dominik. The correct version of this script that should
  57. be used with Org mode is distributed in the =mk/= directory of the Org
  58. mode distribution. Here is the basic workflow for this.
  59. ** Access to the patchwork server
  60. If you want to work on patchwork patches, you need write access at the
  61. patchwork server. You need to contact John Wiegley to get this
  62. access.
  63. There is a web interface to look at the patches and to change the
  64. status of patches. This interface is self-explanatory. There is also
  65. a command line script which can be very convenient to use.
  66. ** Testing patches
  67. To start testing a patch, first assign it to yourself
  68. : pw update -s "Under Review" -d DELEGATE-NAME NNN
  69. where =NNN= is a patch number and =DELEGATE-NAME= is your user name on
  70. the patchwork server.
  71. The get the patch into a branch:
  72. : pw branch NNN
  73. This will create a local topic branch in your git repository with the
  74. name =t/patchNNN=. You will also be switched to the branch so that
  75. you can immediately start testing it. Quite often small amends need
  76. to be made, or documentation has to be added. Also, many contributors
  77. do not yet provide the proper ChangeLog-like entries in the commit
  78. message for the patch. As a maintainer, you have two options here.
  79. Either ask the contributor to make the changes and resubmit the patch,
  80. or fix it yourself. In principle, asking to contributor to change the
  81. patch until it is complete is the best route, because it will educate
  82. the contributor and minimize the work for the maintainer. However,
  83. sometimes it can be less hassle to fix things directly and commit the
  84. changes to the same branch =t/patchNNN=.
  85. If you ask the contributor to make the changes, the patch should be
  86. marked on the patchwork server as "changes requested".
  87. : pw update -s "Changes Requested" -m "What to change" NNN
  88. This will send an email to the contributor and the mailing list with a
  89. request for changes. The =-m= message should not be more than one
  90. sentence and describe the requested changes. If you need to explain
  91. in more detail, write a separate email to the contributor.
  92. When a new version of the patch arrives, you mark the old one as
  93. superseded
  94. : pw update -s "Superseded" NNN
  95. and start working at the new one.
  96. ** Merging a final patch
  97. Once the patch has been iterated and is final (including the
  98. ChangeLog-like entries in the commit message), it should be merged.
  99. The assumption here is that the final version of the patch is given by
  100. the HEAD state in the branch =t/patchNNN=. To merge, do this:
  101. : pw merge -m "maintainer comment" NNN
  102. This will merge the patch into master, switch back to master and send
  103. an email to both contributor and mailing list stating that this change
  104. has been accepted, along with the comment given in the =-m= message.
  105. At some point you might then want to remove the topic branch
  106. : git branch -d t/patchNNN
  107. * Synchonization with Emacs
  108. This is still a significant headache. Some hand work is needed here.
  109. Emacs uses bzr. A useful introduction to bzr for Emacs developers can
  110. be found [[http://www.emacswiki.org/emacs/BzrForEmacsDevs][here]]. While I see all the advantages this would have, I
  111. cannot bring myself to switch away from git for my day-to-day work,
  112. because I know git so well, and because git seems to me as being much
  113. more powerful, conceptionally simple (once you have [[http://newartisans.com/2008/04/git-from-the-bottom-up/][bent your head
  114. around it]]), and so much faster.
  115. So the way I have been doing things with Emacs is this:
  116. 1. I do not update the version in Emacs too often. Just once every
  117. few months - this is frequently enough for the Emacs release cycle.
  118. Care must be taken to get in a *new and stable* version shortly
  119. before Emacs goes into feature freeze and pretest, because that
  120. version is going to be in the wild for a long time.
  121. 2. I watch the Emacs diffs for changes made by the maintainers of
  122. Emacs in the org-mode files in Emacs. Any changes that come up
  123. there, I merge into the development version of Org-mode.
  124. Occasionally I do not do this, if I do not agree with a change.
  125. The changes go into Org /without/ a ChangeLog-like entry in the
  126. commit message. The reason for this is that we will later generate
  127. a ChangeLog file from our commit messages, and I do not want double
  128. ChangeLog entries in the Emacs ChangeLog file.
  129. 3. When I have made a release (usually I wait for the minor releases
  130. to stabilize), I *copy* org files into the Emacs repository. Yes,
  131. I do not merge, I copy. This has been the source of some problems
  132. in the past - Emacs developers are not happy when I accidentally
  133. overwrite changes they made. But I have not had the patience to
  134. work out a better mechanism, and I really dislike the idea that the
  135. version in Emacs starts diverging from my own.
  136. Careful: Copy /org.texi/ and /orgcard.tex/ into the right places,
  137. and also copy the lisp files with *two exceptions*: Do *not* copy
  138. /org-colview-xemacs.el/ and /org-loaddefs.el/. The former does not
  139. belong in Emacs. And the latter would actually be harmful because
  140. Emacs generates its own autoloads.
  141. 4. Generate the ChangeLog entries
  142. For this, I do in the org-mode git repository
  143. : mk/make_emacs_changelog release_7.02.05..release_7.03.02
  144. This will spit out ChangeLog entries (for the given commit range)
  145. that need to go into the ChangeLog files in Emacs. Org-mode
  146. contributes to 3 different ChangeLog files in Emacs:
  147. : lisp/org/ChangeLog (for lisp changes)
  148. : doc/misc/ChangeLog (for org.texi changes)
  149. : etc/ChangeLog (for refcard changes)
  150. When you run the =make_emacs_changelog= program, you will be
  151. prompted for a date in ISO format YYYY-MM-DD, this date will be
  152. used in the ChangeLog entries - Emacs developers want these dates
  153. to be the time when the change has been installed into Emacs, not
  154. the time when we made the change in our own repository. So all the
  155. ChangeLog entries will get the same date. You will also be
  156. prompted for the kind of ChangeLog you want to make, possible
  157. answers are =lisp=, =texi=, and =card=. The program will then
  158. select the correct entries for the specified ChangeLog file. If
  159. you don't like being prompted, you can give the date and type as
  160. second and third command line arguments to =make_emacs_changelog=,
  161. for example
  162. : mk/make_emacs_changelog release_7.02.05..release_7.03.02 2010-12-11 lisp
  163. These entries need to be added to the ChangeLog files in Emacs.
  164. You should, in the ChangeLog file, select the inserted region of
  165. new entries and do =M-x fill-region=, so that the entries are
  166. formatted correctly. I then do look through the entries quickly to
  167. make sure they are formatted properly, that the email addresses
  168. look right etc.
  169. 5. Commit the changes into the bzr repository and you are done. Emacs
  170. developers often look throught the commit and make minor changes -
  171. these need to be merged back into our own repo.
  172. * Updating the list of hooks/commands/options on Worg
  173. Load the =mk/eldo.el= file then =M-x eldo-make-doc RET=.
  174. This will produce an org file with the documentation.
  175. Import this file into =worg/doc.org=, leaving the header untouched
  176. (except for the release number).
  177. Then commit and push the change on the =worg.git= repository.
  178. * Copyright assignments
  179. The maintainer needs to keep track of copyright assignments. Even
  180. better, find a volunteer to do this.
  181. The list of all contributors from who we have the papers is kept on
  182. Worg at http://orgmode.org/worg/org-contribute.php, so that
  183. committers can check if a patch can go into the core.
  184. The assignment process does not allways go smoothly, and it has
  185. happened several times that it gets stuck or forgotten at the FSF.
  186. The contact at the FSF for this is: copyright-clerk@fsf.org
  187. Emails from the paper submitter have been ignored in the past, but
  188. an email from me (Carsten) as the maintainer of Org mode has usually
  189. fixed such cases within a few days.