r837753 | bkorb | 2000-10-20 18:41:50 +0000 (Fri, 20 Oct 2000)
levels of "verbose"
r837782 | bkorb | 2000-10-21 23:44:02 +0000 (Sat, 21 Oct 2000)
augment global opts with long names
r837783 | bkorb | 2000-10-21 23:49:55 +0000 (Sat, 21 Oct 2000)
There are enough commands that it is worth sorting
r837849 | bkorb | 2000-10-31 04:07:54 +0000 (Tue, 31 Oct 2000)
a couple minor nits
r837858 | bkorb | 2000-11-01 01:48:32 +0000 (Wed, 01 Nov 2000)
use a format string to construct ``svn status'' output
r837859 | bkorb | 2000-11-01 01:54:06 +0000 (Wed, 01 Nov 2000)
revise version to revision
r837971 | bkorb | 2000-11-12 17:34:36 +0000 (Sun, 12 Nov 2000)
provide: itemize mark
r837972 | bkorb | 2000-11-12 17:58:09 +0000 (Sun, 12 Nov 2000)
The plural form of "status" is "status"
r837973 | bkorb | 2000-11-12 18:06:15 +0000 (Sun, 12 Nov 2000)
The plural form of "status" is "status" :-)
r837974 | bkorb | 2000-11-12 18:08:55 +0000 (Sun, 12 Nov 2000)
It was too simple to do right :-(
r837979 | bkorb | 2000-11-12 20:25:48 +0000 (Sun, 12 Nov 2000)
bail out on test failures
r837980 | bkorb | 2000-11-12 20:38:12 +0000 (Sun, 12 Nov 2000)
Mark the oopses so I can tell where I am
r837988 | bkorb | 2000-11-13 01:40:58 +0000 (Mon, 13 Nov 2000)
tabularize the command dispatch
r837990 | bkorb | 2000-11-13 01:54:25 +0000 (Mon, 13 Nov 2000)
stumbling over eachother addition
r837991 | bkorb | 2000-11-13 02:06:44 +0000 (Mon, 13 Nov 2000)
stumbling over eachother addition
r837992 | bkorb | 2000-11-13 02:09:35 +0000 (Mon, 13 Nov 2000)
fix inconsistency between command name and procedure name
r838011 | bkorb | 2000-11-14 03:46:36 +0000 (Tue, 14 Nov 2000)
(svn_cl__cmd_proc_t): Clean up the commentary on the callout procedure type.
(svn_cl__cmd_desc_t): Remove the anticipatory fork(2) marker
(svn_cl__te_command): describe the SVN Client command enumeration type
(svn_cl__cmd_proc_t list): enumerate the known command procedures.
r838017 | bkorb | 2000-11-14 04:42:01 +0000 (Tue, 14 Nov 2000)
Clean up some commentary.
The svn client main procedure has now been rewritten to use a command
dispatch/descriptor table. That table plus the `help' command is the
only command specific information that should be in the main.c file.
The option processing is creaky, still there and on intimate terms with
the various commands. It will go away RSN and be replaced with some
sort of table-driven version. Each command will have its own option
table and that table will be referenced from the command descriptor table.
r838018 | bkorb | 2000-11-14 04:49:50 +0000 (Tue, 14 Nov 2000)
`add' did not belong in the xml_file required list
r838033 | bkorb | 2000-11-16 05:19:10 +0000 (Thu, 16 Nov 2000)
There is no `-d' option, so it wound up being treated as
the default argument: a target. That would have been invalid,
except that the argument to the -d option overwrote the old
(invalid) target of `-d' with the real target. Net result:
the test worked by accident. :-)
r838034 | bkorb | 2000-11-16 05:28:50 +0000 (Thu, 16 Nov 2000)
In preparation for ultimately doing some sort of table
thingey for options, I want to lay some ground work now. To
do this, I have made two new struct types. One contains the
data necessary to determine which options are active for the
current command. I do this the old way: with the command
code enumeration and code in the option processing routine.
That must get fixed. The second struct contains all the
data currently used in the client options, plus the active
options descriptor. Each command table entry has a pointer
to a struct of this latter type. The current option
processing code is called from main() instead of the command
procedures. The command procedures receive only command
arguments (the options stripped off). With this done,
changing the underlying option processing will no longer
perterb much of any other code.
r838035 | bkorb | 2000-11-16 13:59:24 +0000 (Thu, 16 Nov 2000)
I ought to update all the files I modified :-(