-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathSTARLINK_BRANCHES
96 lines (71 loc) · 3.8 KB
/
STARLINK_BRANCHES
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
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
Automake branches
=================
automake tree created by the import of automake-1.8.2, on 2004 January 15.
It is best to edit this file only on the HEAD, as there is no point in
giving details about a branch in a version of this file which is
visible only on that branch.
FSF
---
Branch tag common to AUTOMAKE AND AUTOCONF (and m4 and libtool, as it
happens).
Branch FSF is the `vendor' branch, which new releases of automake and
autoconf are imported onto, and immediately tagged (at present, there
are tags autoconf-2-59, automake-1-8-2 and automake-1-9-6 here --
others should be added after the same pattern).
dev-nxg-20040116-add-fpp-support
--------------------------------
Branch tag applied to AUTOMAKE AND AUTOCONF. This is a branch off the
FSF branch.
The patches for adding FPP support, originally announced by Martin
Wilck on
<http://sources.redhat.com/ml/autoconf-patches/2000-07/msg00287.html>,
have been added on this branch, and subsequently developed. They have
been merged onto the main development branch (HEAD) multiple times --
see below.
This branch is semi-permanent: any patches to the FSF distributions
which we intend to submit back to the autoconf/automake mainlines
should go on this branch. At present, this means support for
preprocessable Fortran and a number of extra Fortran-specific macros
in autoconf, plus the fairly minimal additional support necessary for
automake.
At various times, we have identified snapshots of this branch. These
should be tagged on this branch, for both autoconf and automake, with
tags of the form dev-nxg-20040116-add-fpp-support-fpp<n>. This has
been done inconsistently in the past, with tags of the form
autoconf-2-59-fpp4 -- don't do that again, because it's less clear
that these are snapshots of the dev-nxg-20040116-add-fpp-support branch.
Changes on this branch should be merged from time to time onto the
HEAD -- the Starlink-specific mainline. Do that merge with a command
like the following (in a temporary directory -- NB, these commands
untested (FIXME)):
% cvs -d xxx co -jdev-nxg-20040116-add-fpp-support-MERGED \
-jdev-nxg-20040116-add-fpp-support thirdparty/fsf/autoconf
% cvs -d xxx rtag -F -rdev-nxg-20040116-add-fpp-support \
dev-nxg-20040116-add-fpp-support-MERGED thirdparty/fsf/autoconf
Then cd to thirdparty/fsf/autoconf, resolve any conflicts, and commit
the result. Before doing the commit, you can verify that you will
commit to the HEAD by doing 'cvs status -v configure.ac' (for
example), and verifying that the `Sticky Tag' is `(none)'. Do the
analogous operation for automake at the same time. The ...-MERGED tag
logs that you have done this, makes a command like this possible the
next time, and makes it clear which edits should and should not be
expected to be visible on the mainline.
DEV-nxg-20040720-special-install
--------------------------------
I'm afraid I don't really know the status of this branch! I presume it was
for the changes to STAR_SPECIAL_INSTALL_COMMAND, but I appear to have
done them on the trunk instead. How ill-disciplined of me.
However
http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0407&L=stardev&P=R15945&I=-1
gives a clue, and suggests that I may well have merged this at some
point in the past. At any rate, the special-install stuff currently
works, so if there's anything left unmerged on this branch it's
probably best abandoned.
DEV-nxg-20040730-fix-ac-canonical
---------------------------------
Automake handling of AC_CANONICAL_* is wrong, I think. Fix on this
branch, and the patch offered to [email protected]:
http://lists.gnu.org/archive/html/automake-patches/2004-07/msg00023.html
Merged onto trunk 2004 July 30. Though the code itself didn't make it
into the automake tree, the test case did, and the bug was fixed, and
that fix hand-applied to automake.in. Branch now abandoned.