r.buffer.html:<B>r.cats input=</B>roads.buf
r.li.patchdensity.html:r.cats landcover.30m
r.report.html:<DD>Report for cats fp ranges (fp maps only)</DD>
r.statistics.html:r.cats elevstats
r.statistics.html:<A HREF="r.cats">r.cats</A>
r.stats.html:<DD>Report for cats fp ranges (fp maps only)</DD>
v.net.iso.html:# specify range of center cats (easier to catch all):
v.to.rast.html:r.cats myfields
Markus
On Feb 19, 2008 8:25 PM, Michael Barton <michael.barton@asu.edu> wrote:
I just noticed that we are missing the help file for r.cats.
Michael
______________________________
Michael Barton, Professor
Professor of Anthropology
Director of Graduate Studies
School of Human Diversity & Social Change
Center for Social Dynamics & Complexity
> The file was renamed to r.category... r.cats as module is there
> for backwards compatibility
....
> Additionally, several manual pages refer to the old name:
> $ grep r.cats *
Ivan wrote:
Please note that this searches for ``r'', any character,
``cats'', not for ``r'', period, ``cats''. Hence, the list
below has three false positives.
The proper command to search for the latter (assuming POSIX
`grep' [1]) is:
$ grep -F r.cats *
Right. FWIW my method was 'grep -rI r\.cats *', but whatever route it
all gets to the same destination I guess...
I should have updated all of those when I renamed r.cats, oops. Done
now, all help page and GUI menu entries are updated in SVN/trunk, and
Markus has backported to the 6.3.0 branch.
r.category expands on the old r.cats as it can manage raster category
labels, whilst r.cats could only report them. There are interesting
things to play with like dynamic label names. It's very old library
functionality that has been somewhat hidden from the user for years.
>>> The proper command to search for the latter (assuming POSIX `grep'
>>> [1]) is:
>>> $ grep -F r.cats *
>> Right. FWIW my method was 'grep -rI r\.cats *'
> If you're doing that from the shell, it needs two backslashes:
> grep -rI r\\.cats *
> or quotes:
> grep -rI 'r\.cats' *
> so that grep actually "sees" a backslash.
> Or just use "grep -F" or "fgrep" for searching for fixed strings.
And since one cannot reliably and portably quote a string into a
regular expression in Shell, one has to use:
$ grep -F -- "$string" ...
to search for a user-supplied string in a Shell script.
(The `--' ``option'' prevents `"$string"' from being processed
as an option even if it begins with `-'; consider, e. g.: `$
grep -l *' vs. `$ grep -- -l *'.)
>> If you're doing that from the shell, it needs two backslashes:
>> grep -rI r\\.cats *
>> or quotes:
>> grep -rI 'r\.cats' *
>> so that grep actually "sees" a backslash.
>> Or just use "grep -F" or "fgrep" for searching for fixed strings.
> true. another annoying flase positive thing since the move to SVN is
> that duplicate '/.svn/' entries pop up and must be filtered out with
> 'grep -v'.
GNU grep has `--include', so I often use something like the
following:
$ grep -rF --include='*.[ch]' G_foo lib/ raster/
$ grep -rF --include=\*.html g.bar ./
Unfortunately, there's no way to skip directories matching a
wildcard (e. g.: --exclude=.svn), at least as of GNU grep 2.5.1.
The `-l' option could be useful as well, e. g.:
$ less $(grep -rlF --include='*.[ch]' G_foo raster/)