File: TODO

package info (click to toggle)
sql-ledger 3.2.6-1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm, bullseye, buster, forky, sid, trixie
  • size: 33,080 kB
  • sloc: perl: 73,172; sql: 36,530; sh: 34; makefile: 21
file content (81 lines) | stat: -rw-r--r-- 4,055 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
	Debian packaging TODO for sql-ledger

  - Reference also any possible bugs for Debian or Ubuntu:
https://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=sql-ledger
https://bugs.launchpad.net/ubuntu/+source/sql-ledger

===============================================================================
-------------------------------------------------------------------------------
  - The following Lintian result comes up on the current upstream;  however,
    upstream does not provide anything for testing:
I: sql-ledger source: testsuite-autopkgtest-missing

-------------------------------------------------------------------------------
  - For the 'SQL-Ledger / Add Dataset' page in admin.pl, the "User" field 
  defaults to "sql-ledger" but IIRC Pg (SQL?) doesn't allow user/role names
  with dashes in them. (It may have at one time.)
  https://www.postgresql.org/docs/current/static/sql-syntax-lexical.html
  - Find and patch that to default to "sqlledger" instead of "sql-ledger"?
  - Update README Debian as necessary to reflect that? The role name is
  mentioned as an alternative for using 'postgres' for the DB admin role.

-------------------------------------------------------------------------------
  debian/rules:
override_dh_auto_build
 - Replace the 'sed' command line generating a default configuration file with
   a patch that does so, specific to the Debian package.
 - Use ucf to install the patched & possibly renamed file?
 - Update the README.Debian file as necessary about it.

-------------------------------------------------------------------------------

- The existing latex-cmdline-option patch did not apply as is to the new
  upstream. That was resolved by removing the patch, at least for now, and
  adding a note about it to debian/NEWS.

-------------------------------------------------------------------------------
    - This was found for the v3.0.8-1 version of the package.
  - From: https://packages.qa.debian.org/s/sql-ledger.html

Package sql-ledger shares 9.0 MB of similar files with itself[1], please 
investigate whether it is possible to reduce the duplication[2].'

[1] https://dedup.debian.net/compare/sql-ledger/sql-ledger
[2] https://wiki.debian.org/dedup.debian.net
-------------------------------------------------------------------------------
  - Review, as it is unclear when the page was last updated & therefore how
  much might already have been referenced in the Debian packaging:
http://www.sql-ledger.com/cgi-bin/nav.pl?page=misc/packaging.html&title=Packaging

-------------------------------------------------------------------------------
  -Original TODO (& therefore, old) from package:

Known bugs :
- postinst doesn't handle well apache2. It tries to create /etc/apache/httpd.conf 
  => Make use of "wwwconfig-common" to be nicer about apache configuration ?

1/ Improve README.Debian to provide reliable secure instructions.

2/ Automatize as much as possible:
   - add plpgsql support in "template1" db automatically
   - create a "sql-ledger" user in Postgres (password should be requested
     by debconf)
   - configure PostgreSQL to use password authentication for "sql-ledger"
     user by default :
     Currenly this means adding a line in pg_hba.conf:
     local   all         sql-ledger                          md5
     This line should be before the lines:
     # "local" is for Unix domain socket connections only
     local   all         all                               ident sameuser

3/ Upstream docs speak of locking SQL-ledger during "manual upgrades". Maybe
   we should do that automatically: lock in preinst, unlock in postinst.
   If the upgrade involves a new upstream version, we could decide to not
   unlock to make sure that the administrator upgrades the database by
   logging itself and then unlocking the DB manually. Otherwise we might
   have "active sessions" working with the new code but with the old
   database.

4/ Starting from now (v2.6.4), anything described in various UPGRADE-*
   files should be done automatically if at all possible.