[Oberon] Re: Text search is THE main task.
jmdrake_98 at yahoo.com
Wed Jun 16 22:24:47 CEST 2004
--- easlab at absamail.co.za wrote:
> John Drake wrote:-
> > Anyway I've added a command RX.GrepAll that will
> > search all files meeting a specific pattern.
> > For instance:
> > RX.GrepAll "*.Mod" ("Write" | "String")
> Are you making the code available [for beta
Yes. In the process now of deciding where to host
it. Thinking seriously about SourceForge. Perhaps
an "Oberon Sys 3 enhancements" project? I thought
about pasting the new code (about 139 lines) to the
Wiki, but I wasn't sure if that was appropriate.
> Yes please !
> Google blows my mind away: I need the press/print
> report[s] of
> municipal corruption exposed on TV last night [here
> is deep-dark
> africa]. By ANDing a few key-words I've got
> Obviously the print occured before the TV - google
> won't update
> from distant locations within hours ?
> What attracted me to [the existing] "Find" is that
> uses "Find.Domain <file list>", so you can
> 'refine/restrict' the
> file search-space by existing tools. Eg. by
> [which I suspect is not included in the
> distributions, but which
> I use daily, and should be included].
Interesting. So this was part of the NO
distribution? If not, where did you find it?
> Running RX.Grep <file of list-of-files> \i
> <extentions/types to omit>
> trims out the fileTypes which are not to be
> Eg. in a BIG SYS: partition, you can skip all:
> *.Obj, *.Bak,
> *.Scn.Fnt, ...etc. [which can't be moved out to
> their own
> partition] by building a file list, [for repeated
> use], by repeated
> applications of RX.Grep.
> BTW. v2.3.2 has a facility & panel to include a date
> range for
> the files to be searched [it's potentially
> problematic, since the
> date format required is not obvious]. I'm wondering
> why this
> was omitted from later versions ?
I'm not sure. But that does sound like a nice
> BTW is 'WebNavigator' usefull for local files ?
> It sounds like being intended for inet searching ?
WebNavigator is an HTML 4.0 web browser. (Think
"Netscape Navigator".) It was a student project.
I recall the author asking for help with it some
time back, but I don't know if anyone ever
responded. Anyway the parser works pretty good
(I've used it in a project at work to parse HTML
output from a statistics program), and it renders
some pages nicely, but it needs a lot of work.
In some ways the HTML 2.0 parser that comes with
Oberon is more sophisticated since it can handle
forms. (WebNavigator can't handle Google for
instance, but the built in HTML 2.0 parser can).
Incidently this means that WebNavigator currently
can't handle editing the Wiki either.
Unfortunately the HTML 2.0 browser doesn't seem
to be able to handle the Wiki either. You can
edit a page but your edits don't get saved.
I wonder if this has something to do with Oberon's
weak (practically non existent) support for cookies?
Anyway Google does support XMLRPC. And since
we have an XMLRPC client it should be possible to
build a Google search client in Oberon.
John M. Drake
Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.
More information about the Oberon