Add support for a tags file to the build system.

[cvsps-yd/commitid.git] / cvsps.1

.TH "cvsps" 1
.SH NAME
CVSps \- create patchset information from CVS
.SH SYNOPSIS
.B cvsps
[\-h] [\-x] [\-u] [\-z <fuzz>] [\-g] [\-s <patchset>] [\-a <author>] [\-f <file>] [\-d <date1> [\-d <date2>]] [\-l <text>] [\-b <branch>] [\-r <tag> [\-r <tag>]] [\-p <directory>] [\-v] [\-t] [\-\-norc] [\-\-summary\-first] [\-\-test\-log <filename>] [\-\-bkcvs] [\-\-no\-rlog] [\-\-diff\-opts <option string>] [\-\-cvs\-direct] [\-\-debuglvl <bitmask>] [\-Z <compression>] [\-\-root <cvsroot>] [\-q] [\-A] [<repository>] 
.SH DESCRIPTION
CVSps is a program for generating 'patchset' information from a CVS
repository.  A patchset in this case is defined as a set of changes made
to a collection of files, and all committed at the same time (using a
single 'cvs commit' command).  This information is valuable to seeing the
big picture of the evolution of a cvs project.  While cvs tracks revision
information, it is often difficult to see what changes were committed
'atomically' to the repository.
.SH OPTIONS
.TP
.B \-h
display usage summary
.TP
.B \-x
ignore (and rebuild) ~/.cvsps/cvsps.cache file
.TP
.B \-u
update ~/.cvsps/cvsps.cache file
.TP
.B \-z <fuzz>
set the timestamp fuzz factor for identifying patch sets
.TP
.B \-g
generate diffs of the selected patch sets
.TP
.B \-s <patchset>[\-[<patchset>]][,<patchset>...]
generate a diff for a given patchsets and patchset ranges
.TP
.B \-a <author>
restrict output to patchsets created by author
.TP
.B \-f <file>
restrict output to patchsets involving file
.TP
.B \-d <date1> \-d <date2>
if just one date specified, show
revisions newer than date1.  If two dates specified,
show revisions between two dates.
.TP
.B \-l <regex>
restrict output to patchsets matching regex in log message
.TP
.B \-b <branch>
restrict output to patchsets affecting history of branch.
If you want to restrict to the main branch, use a branch of 'HEAD'.
.TP
.B \-r <tag1> \-r <tag2>
if just one tag specified, show
revisions since tag1. If two tags specified, show
revisions between the two tags.
.TP
.B \-p <dir>
output individual patchsets as files in <dir> as <dir>/<patchset>.patch
.TP
.B \-v
show very verbose parsing messages
.TP
.B \-t
show some brief memory usage statistics
.TP
.B \-\-norc
when invoking cvs, ignore the .cvsrc file
.TP
.B \-\-summary\-first
when multiple patchset diffs are being generated, put the patchset
summary for all patchsets at the beginning of the output.
.TP
.B \-\-test\-log <captured cvs log file>
for testing changes, you can capture cvs log output, then test against
this captured file instead of hammering some poor CVS server
.TP
.B \-\-bkcvs
(see note below) for use in parsing the BK\->CVS tree log formats only.  This enables
some hacks which are not generally applicable.
.TP
.B \-\-no\-rlog
disable the use of rlog internally.  Note: rlog is
required for stable PatchSet numbering.  Use with care.
.TP
.B \-\-diff\-opts <option string>
send a custom set of options to diff, for example to increase
the number of context lines, or change the diff format.
.TP
.B \-\-cvs\-direct (\-\-no\-cvs\-direct)
enable (disable) built\-in cvs client code. This enables the 'pipelining' of multiple
requests over a single client, reducing the overhead of handshaking and
authentication to one per PatchSet instead of one per file.
.TP
.B \-\-debuglvl <bitmask>
enable various debug output channels.
.TP
.B \-Z <compression>
A value 1\-9 which specifies amount of compression.  A value of 0 disables compression.
.TP
.B \-\-root <cvsroot>
Override the setting of CVSROOT (overrides working dir. and environment).  For --cvs-direct only.
.TP
.B \-q
Be quiet about warnings.
.B \-A
Show ancestor branch when a new branch is found.
.TP
.B \<repository>
Operate on the specified repository (overrides working dir.)
.SH "NOTE ON TAG HANDLING"
Tags are fundamentally 'file at a time' in cvs, but like everything else,
it would be nice to imagine that they are 'repository at a time.'  The
approach cvsps takes is that a tag is assigned to a patchset.  The meaning
of this is that after this patchset, every revision of every file is after
the tag (and conversely, before this patchset, at least one file is still
before the tag).  However, there are two kinds of inconsistent (or 'funky')
tags that can be created, even when following best practices for cvs.  
.PP
The first
is what is called a FUNKY tag.  A funky tag is one where there are patchsets
which are chronologically (and thus by patchset id) earlier than the tag, but
are tagwise after.  These tags will be marked as '**FUNKY**' in the Tag: section
of the cvsps output.  When a funky tag is specified as one of the '\-r' arguments,
there are some number of patchsets which need to be considered out of sequence.  
In this case, the patchsets themselves will be labeled FUNKY and will be processed
correctly.
.PP
The second is called an INVALID tag.  An invalid tag is a tag where there are
patchsets which are chronologically (and thus by patchset id) earlier than the tag,
but which have members which are tagwise both before, and after the tag, in the
same patchset.  If an INVALID tag is specified as one of the '\-r' arguments,
cvsps will flag each member of the affected patchsets as before or after the tag
and the patchset summary will indicate which members are which, and diffs will 
be generated accordingly.
.SH "NOTE ON CVS VERSIONS"
Among the different cvs subcommands used by cvsps is the 'rlog' command.  The
rlog command is used to get revision history of a module, and it disregards
the current working directory.  The important difference between 'rlog' and 'log'
(from cvsps perspective) is the 'rlog' will include log data for files not in
the current working directory.  The impact of this is mainly when there are 
directories which at one time had files, but are now empty, and have been pruned
from the working directory with the '\-P' option.  If 'rlog' is not used, these
files logs will not be parsed, and the PatchSet numbering will be unstable.
.PP
The main problem with 'rlog' is that, until cvs version 1.11.1, 'rlog' was an
alias for the 'log' command.  This means, for old versions of cvs, 'rlog' has
different semantics and usage.  cvsps will attempt to work around this problem
by detecting capable versions of cvs.  If an old version is detected, 'log' will
be used instead of 'rlog', and YMMV.
.SH "NOTE ON GENERATED DIFFS"
Another important note is that cvsps will attempt, whenever possible, to use the
r\-commands (rlog, rdiff  and co) instead of the local commands (log, diff, and update).
This is to allow cvsps to function without a completely checked out tree.  Because
these r\-commands are used, the generated diffs will include the module directory in 
them, and it is recommended to apply them in the working directory with the \-p1 option
to the patch command.  However, if the \-\-diff\-opts option is specified (to change, for 
example, the lines of context), then rdiff cannot be used, because it doesn't support
arbitrary options.  In this case, the patches will be generated without the module
directory in the path, and \-p0 will be required when applying the patch.  When 
diffs are generated in cvs\-direct mode (see below), however, they will always
be \-p1 style patches.
.SH "NOTE ON BKCVS"
The \-\-bkcvs option is a special operating mode that should only be used when parsing
the log files from the BK \-> CVS exported linux kernel trees.  cvsps uses special
semantics for recreating the BK ChangeSet metadata that has been embedded in the log
files for those trees.  The \-\-bkcvs option should only be specified when the cache
file is being created or updated (i.e. initial run of cvsps, or when \-u and \-x options
are used).
.SH "NOTE ON CVS\-DIRECT"
As of version 2.0b6 cvsps has a partial implementation of the cvs client code built 
in.  This reduces the RTT and/or handshaking overhead from one per patchset member
to one per patchset.  This dramatically increases the speed of generating diffs
over a slow link, and improves the consistency of operation.  Currently the \-\-cvs\-direct
option turns on the use of this code, but it very well may be default by the time
2.0 comes out.  The built\-in cvs code attempts to be compatible with cvs, but may
have problems, which should be reported.  It honors the CVS_RSH and CVS_SERVER 
environment variables, but does not parse the ~/.cvsrc file.
.SH "NOTE ON CVSPS RC FILE"
CVSps parses an rc file at startup.  This file should be located in ~/.cvsps/cvspsrc.
The file should contain arguments, in the exact syntax as the command line, one per line.
If an argument takes a parameter, the parameter should be on the same line as the argument.
.SH "NOTE ON DATE FORMATS"
All dates are reported in localtime.  This can be overridden (as usual) using the TZ
environment variable.  Dates as arguments must be in the format 'yyyy/mm/dd hh:mm:ss'; for example,
.IP "" 4
$ cvsps -d '2004/05/01 00:00:00' -d '2004/07/07 12:00:00'
.SH "SEE ALSO"
.BR cvs ( 1 ),
.BR ci ( 1 ),
.BR co ( 1 ),
.BR cvs ( 5 ),
.BR cvsbug ( 8 ),
.BR diff ( 1 ),
.BR grep ( 1 ),
.BR patch ( 1 ),
.BR rcs ( 1 ),
.BR rcsdiff ( 1 ),
.BR rcsmerge ( 1 ),
.BR rlog ( 1 ).
.SH "REPORTING BUGS"
Report bugs to "David Mansfield <cvsps@dm.cobite.com>"
.SH BUGS
No known bugs.