Did you know ... Search Documentation:
Packs (add-ons) for SWI-Prolog

Package "upsh"

Title:Save a state for executing commands.
Rating:Not rated. Create the first rating!
Latest version:2.1
SHA1 sum:9cc904776b919c1033b3e9ac24a2c5a36bf1b49f
Author:Nicos Angelopoulos http://stoics.org.uk/~nicos
Maintainer:Nicos Angelopoulos http://stoics.org.uk/~nicos
Packager:Nicos Angelopoulos http://stoics.org.uk/~nicos
Home page:http://stoics.org.uk/~nicos/sware/upsh
Download URL:http://stoics.org.uk/~nicos/sware/packs/upsh/upsh-*.tgz

Reviews

No reviews. Create the first review!.

Details by download location

VersionSHA1#DownloadsURL
2.19cc904776b919c1033b3e9ac24a2c5a36bf1b49f2http://stoics.org.uk/~nicos/sware/packs/upsh/upsh-2.1.tgz

Upsh 2.0.4

Package.

To install within SWI-Prolog, do:

?- pack_install( upsh ). ?- use_module(library(upsh)). ?- upsh_make.

This will create executable state upsh in same directory as swipl. So you then run Prolog scripts from the command line:

> upsh say naku % /home/nicos/.rcpl compiled 0.00 sec, 8 clauses % /home/nicos/bin/cline_upsh/say.pl compiled 0.00 sec, 5 clauses naku

> upsh v Upsh (2:1:0)-swi(7:5:7)-(2017/6/3-0:34:36)

Overview.

This is upsh (Unix-Prolog SHell, or something like that- btw, I pronounce it, oupssss). The unix scripts in this directory will built an executable state (upsh*). This can be used for running Prolog programs from the unix command line.

There are two main ways in which the state can be used: (a) command_line invocation, by issuing something like, % upsh say hello world (b) script invocation, by % say_pl bye

In both cases, execution will start from <Script>/1,0 or main/1,0 one of which should be defined in source Program (this case `say.pl'). If more that one of these predicates are defined, the first one in the order given above will be executed.

See doc/upsh-manual.ps for more information.

Prologs.

Upsh was written on SICStus prolog. Currently is known to work with anything later than : SICStus (3.8.5) SWI (4.0.2) Yap (4.3.18)

Installation.

For the Prolog of you choice do % ./make_Prolog

If this works, it will create an upsh* executable in ~/bin/ Obviously if you want upsh-es for more than one Prolog engine, move this executable to (say) ~/bin/upsh_Prolog1 and repeat the above for Prolog2.

Things that you can change are : (a) the dir in which upsh* will go. Edit make_<Prolog> (b) the dir where command_line invocations of upsh* will search for sources. Edit the file_search_path( upsh, '~/bin/upsh_cline' ) line of file src/upsh.pl
After installation do % rehash (or the equivelant thing for your shell) to make current shell, aware of the new executable.

If you want to use the automatic finding of command_line sources, create the directory ~/bin/upsh_cline or the directory of (b) above. Any Prolog program left there will be accessible to upsh for command_line invocation (see examples). In addition you can create file Lookup.pl in the same directory. See scripts/Lookup.pl for an example file.

Upsh options.

The proper syntax for invoking upsh* expects a number of atoms that are taken to be upsh* options followed by a - seperator, followed by the Program to be executed and its Arguments. Thus, upsh {h,v}~ {<c>,s} {<m>,n} [b,e/1,f/1,d,p,o,r,w,a]- Program [Arguments]*

Curly brackets ({}) mean either-or. Less, and greater mean default value. [Argumentss]* means a number of arguments (possibly none). Things starting with a capital letter are place-holders, lower case starters are literals. Order is not important (apart of the placement of `-', that is).

Normally any progam arguments (Arg) of the form nice=tara, will be translated to nice(tara) (similarly isnogood goes to is(no(good))). These will be seen as terms in the call to the script. If you use '-' and option 'o' then the isnogood is passed as picked by current_prolog_flag( argv, AllArgs ), (i.e. as one atom). Every other Arg is passed as atom (including numbers, so make sure you convert them, before using them in your program).

Sloppy syntax, is also supported. This allows for the omission of `-'. It is the default behaviour (change with 'r' option) and it means upsh* will take anything that looks like an upsh option from the command line and treats anything that it didnt use as if it appearred after a hyphen, '-'. (In this case your Args will get translated for sure because that's what happens to upsh arguments).

The following options are thought to be recognised.

h : prints a help message (on error stream) and exits. v : prints the version id (on error stream) and exits. b : be verbose c : command line invocation (default). s : script invocation. e : errors file (default: File=std_out). This the only way to redirect the error stream on some Prologs, since upsh muffles the std_err to avoid the logo message of these Prologs. You can use /dev/null ok. Also, dev_null is recognised. f : name of functor for top goal (default: <filename>/1{.pl} then, main/1) d : working directory (default: Dir=\'.\'). m : compile Program (default). n : consult Program. Not very useful as is because upsh will throw you out of the Prolog system, in the event of something gone wrong, anyway. The original use was to be able to debug when execution failed/aborted unexpectedly. If you know what you are doing comment out the portray_message( informational, _ ) lines in upsh_Prolog.pl rebuild and you can then do some debugging from the Prolog prompt. p : surpress loading info (of Program). All those compiling and compiled messages. o : donot translate, Arg. thus `verynicetara' behind `-' will not be translated to term very(nice(tara)). r : be strict about the \`-\' . Options to the left Program and it Args to the right. Anything that is not an upsh option and appears to the left goes to the bin. Similarly Args that could be seen as upsh options and appear to the right of '-' will be passed to the Program. w : wait for input at end of execution. a : last argument holds all read variable lists (upsh_vs/1).

For scripts add the following, as the first two lines in the prolog file,

/bin/sh exec upsh s - $0 "$@"

Then, do : % chomod u+x <PlFile>

If you have problems with working in the wrong directory, try the following as a second line, instead :

exec upsh s d=pwd - $0 "$@

Examples.

Upsh, comes with two examples, say and tall which can be found in directory scripts/ . 'say' simply spews out its input arguments one at the time. tall, touches all its arguments (where `touch'; is the unix command). These come in two flavours, File.pl for running as an upsh* command_line argument, or used from within the interpreter, while File_pl is an executable script.

Some examples with these two programs, follow :

% upsh say hello world % upsh p - say hello tara % upsh n - say bye % upsh e=errors - say hello all % more errors % rm errors % ./say_pl hello % cp tall_pl ~/bin/tall (or similar dir in your $PATH) % rehash (or similar, for your shell) % tall tara sha tom (these 3 files will be `touched') % rm ~/bin/tall % rm tara sha tom % upsh d='..' - upsh/tall.pl panos petros dhmhtra giannis (these 4 files will be `touched' on ..) % rm ../panos ../petros ../dhmhtra ../giannis % upsh upsh/tall.pl d='..' panos petros dhmhtra giannis (illustrates sloppy syntax) % rm ../panos ../petros ../dhmhtra ../giannis % upsh say kalh=sebasth % upsh o - say tavladoros=andrea (illustrates translation) % mkdir ~/bin/upsh_cline (if you havent done so yet) % cp say.pl ~/bin/upsh_cline % cd ~ % upsh say bye-t-bye (~/bin/upsh_cline/say.pl is used) % rm ~/bin/upsh_cline/say.pl (if you so wish)

Distribuition.

This software is not distributated, per se. It simply exists. Whether you use it or not, is your problem.

Interesting suggestions, can be send to n.angelopoulos@nki.nl

Nicos, April 2004.

Last update: April 2012. Made into a pack May 2016.

Contents of pack "upsh"

Pack contains 69 files holding a total of 654K bytes.