The manual and release notes for upcoming version 10.2.0 of newLISP have been updated here:
http://www.newlisp.org/downloads/development/latest/
If anybody wants to make corrections (including better wording), there are not many days left before release.
(>> 0x800000000000000 1)
(>> 0x800000000000000
Expressions in exp-default are evaluated in the function's current environment.
(define
(define (foo (a 10) (b (div a 2)))
If no step size is specified, 1
If no step size is specified, 1 is assumed.
If payment is at the end of the period, int-type is 0 (zero); for payment at the end of each period, int-type is 1. If num-type is omitted, payment at the end of
If payment is at the end of the period, int-type is 0 (zero); for payment at the end of each period, int-type is 1. If num-type is omitted, payment at the end of
because
99808.03333
> (fv (div 0.07 12) 1 775.3 -100000 1)
99803.51075
> (fv (div 0.07 12) 1 775.3 -100000)
99808.03333
If
If
maybe.
Sorry, I ended the translation up to I, yet.
The last lines of the newLISP manual function definitions:
======
Use the ++ function for incrementing numbers in integer mode.
----------
Use the inc function to increment numbers.
======
Need to be changed to something like:
======
Use the ++ function for incrementing numbers in integer mode.
Use the dec function to decrement numbers.
----------
Use the -- function for decrementing numbers in integer mode.
Use the inc function to increment numbers.
======
(And if it's not too much trouble ;p)
You added
in:
Function Reference
1. Syntax of symbol variables and numbers
2. Data types and names in the reference
3. Functions in groups
* List processing, flow control, and integer arithmetic
Now the table entry is:
Besides looking messy ;p)
When deconstructing the single newLISP manual html file into individual html pages, the
Could you change the table to:
-- xytroxon
Many thanks to Johu and Xytroxon for the corrections. Updated versions of the documents can be found here:
http://www.newlisp.org/downloads/development/latest/
Thank you Lutz.
and,
mat
Using the second syntax, all cells in matrix-A are
Using the second syntax, all cells in matrix-A are
maybe.
I ended the translation up to M.
Sorry, I have not ended the translation up to N, yet.
This function has only usage on
This function has only usage on
An interface address can be defined as either
An interface address can be defined as either IP address or a name.
maybe and also,
Thanks for your help, Johu. Release will be next week, so everything coming in until Monday or Tuesday, will be in the distribution packages. But I always post revisions with corrections for the on-line manual between releases too.
newLISP manual spelling errors:
addtionally -> additionally
arithmetik -> arithmetic
beginnig -> beginning
charcaters -> characters
comforms -> conforms
comminicate -> communicate
directoy -> directory
Erros -> errors
errror -> errors
espressions -> expressions
exspired -> expired
fucntion -> function
imformation -> information
implementarion -> implementation
insteead -> instead
interacive -> interactive
libraryi -> library
specific matric operations -> specific matrix operations
mssage -> message
nsufficient -> insufficient
occurrance -> occurrence
Ofset -> Offset
othet -> other
previusly -> previously
recognizabe -> recognizable
routinesi -> routines
succesfully -> successfully
in thr sender -> in the sender
Whe the prefix -> When the prefix
woth -> with
-- xytroxon
Thanks Xytroxon, updated here:
http://www.newlisp.org/downloads/development/latest/newlisp_manual.html
If exp evaluates to nil, then true is returned; otherwise, nil is returned.
If exp evaluates to nil
Next time, I will post in a new topic.
Spelling errors in Code Patterns 10.1.12 02-23-2010
allthough -> although
chracter -> character
controlls -> controls
conventsions -> conventions
distibuted -> distributed
fucntion -> function
libararies -> libraries
mesaging ->messaging
messsages -> messages
mulitple -> multiple
persisten -> persistent
shapters -> chapters
sring -> string
sruct mystruct -> struct mystruct
transfomed -> transformed
-- xytroxon
Many thanks again to Johu and Xytroxon for the latest corrections. You can find files in the usual downloads/development/latest/ -location. Please try to post pending corrections until March 16th, 12:00 UTC.
See also the address, get-int, get-long get-char, get-string, and unpack functions.
See also the address, get-int, get-long
If >str-data is specified, there is no limitation on the length of tokens.
If
Instead of { and } (left and right curly brackets), quotes can be used to limit the pattern.
Instead of { and } (left and right curly brackets),
And,
Are the following the same ?
I think that
If payment is at the end of the period, int-type is 0 (zero) or int-type is omitted; for payment at the beginning of each period, int-type is 1.
If payment is at the end of the period, int-type is 0; else it is 1. If int-type is omitted, 0 is assumed.
When paying at the end of each period,
When paying at the end of each period,
If payment is at the end of each period, 0 (zero) is assumed for int-type; otherwise, 1 is assumed.
Quote
Lutz,
can you post the link to the odt file when the documentation becomes stable.
I want to compile it to a CHM file as you suggested.
Here I have CHM files belonging to the previous version 10.1.x ,
namely
codepatterns and
introduction-to-newlisp from Wikibooks
Would you prefer me to update them first?
Also I was working on a compiled help file containing the GTK+ Reference Manual for GTK+ 2.18.9 .
The good thing will be to use the reference as contextsensitive help in editors.
Do you and others like them separately or all in one file?
Regards,
Michael
Here is the latest version in OpenOffice ODT format:
http://www.newlisp.org/newlisp_manual-10201.odt
The PDF and HTML versions have been updated too.
Ps: I would keep different formats in different files.
Sorry to have missed out on the editing this time round! Thanks to everyone for making theh newLISP manual (even) better!
In some spare moments I've been thinking about ebooks and electronic docs. I've been using the newlisp manual as a real-world example. One thing I tried was this:
(xml-type-tags nil nil nil)
(set 'xf (xml-parse f 15))
Yes, it doesn't work. But that's just because the manual is in HTML Transitional. I ran the HTML source through BBEdit's version of Tidy - converting it to XHTML. It appears to make little difference to the 'user experience' in a browser - a lot of changes to BR tags and so on, but not much else.
But now that code above works, and produces SXML of the source:
(content "text/html; charset=utf-8")))
(meta ((name "author") (content "Lutz Mueller")))
(meta ((name "keywords") (content "newLISP Lisp SCHEME programming language manual reference Artificial Intelligence AI NUEVATEC")))
(meta ((name "description") (content "newLISP Users Manual and Reference")))
...
(a ((name "shell") (id "shell")))
(h2 (span ((class "function")) "!"))
(h4 "syntax: (! " (em "str-shell-command") ")")
(p "Executes the command in " (em "str-command") " by shelling out to the operating system and executing. This function returns a different value depending on the host operating system.")
(pre "n(! "vi") n(! "ls -ltr")n")
...
Yes it looks crazy (but better than HTML), but I can't help thinking that this format might be useful, because you can easily use newLISP on it. For example:
(dolist (r xr) (println (rest (xf (chop r)))))
gives a list of all syntax definitions, suitable for producing a quick reference guide...
A few questions arise: what are the problems with distributing the manual in XHTML rather than HTML, such that a trip through Tidy wouldn't be necessary? Also, how could you go about outputting the SXML structure to other formats, such as plain old HTML? It kind of looks to me like it could be executable, somehow, if there were definitions for the major tags... And another possibility - could this format be the basis for some kind of newLISP-oriented text editor?
Just a few thoughts... :)
peanut gallery here. all of that sounds very cool and useful to me. :-)
This is a great idea. If you can send me an XHTML translation of the current rev-2 manual from:
http://www.newlisp.org/downloads/newlisp_manual.html
and I can try if it works with my PDF conversion.
In the end I succeded to do the XHTML conversion myself, and its online now:
http://www.newlisp.org/downloads/newlisp_manual.html
it can be parsed using:
(xml-parse (read-file "newlisp_manual.html") 15)
and you can do this:
((h1 "Contents") (h1 "newLISP Users Manual") (h1 "newLISP Function Reference"))
>
The 'true' flag at the end of 'ref/ref-all' was introduced in 10.1.7 to return the content instead of the index-vector.
Cool - I was going to do it when I got back today but you beat me to it. I think Tidy is freely available, even though BBEdit costs money...
I can see how you'd scan the SXML for analysis:
(define (count-tags expr)
(cond
((list? (first expr))
(count-tags (first expr))
(dolist (s (rest expr)) (count-tags s)))
((symbol? (first expr))
(dolist (s (rest expr))
(count-tags s))
(set 'key (string (first expr)))
(if (Tags key)
(Tags key (inc (Tags key)))
(Tags key 1))
)
)
)
(count-tags xf)
(println (Tags))
(
("a" 2298)
("align" 47)
("alink" 1)
("b" 150)
("bgcolor" 1)
("blockquote" 21)
("body" 1)
("border" 20)
("br" 1079)
("cellpadding" 20)
("center" 42)
("class" 2031)
("color" 29)
...
But presumably it's possible to actually execute the SXML list directly, provided that there are suitable functions already define for each symbol/tag...? How would I do that?
I tried to work out how to make such an SXML list can be an expression that can be directly evaluated. The problem is with the double-parenthesized expressions. For example, this is OK:
(p "Quoted symbols can also be used to pass data by reference, but this method has disadvantages:")
because user-defined functions can be defined that output some representation of paragraph and underlined text. But I can't make these work:
because of the double parentheses?
You need to add the SXML attribute tag '@' via option 16
(xml-parse "<a href='#thelink'>a Link</a>" 31)
=> ((a (@ (href "#thelink")) "a Link"))
Now the '@' could be defined as processing the s-expressions follwing it. Basically it would work like a 'begin'.
Yes - that works, thanks! Although now I've got even further into the jungle:
First I had to redefine the built-in symbols that also appear in the SXML, but then:
(set-ref-all 'name xf '_name)
(set-ref-all '@ xf '_@)
(define (_@) (begin (args)))
(map
(fn (f)
(letex
((fn-name (sym (string f))))
(define (fn-name a)
(string a))))
(map first (Tags)))
(println (apply eval xf))
it gives a stack overflow error... Nested too much? I didn't think so. I think I'm close to a solution but ...
In newlisp_manual.html, some lines that have a unbalanced double quotes(") are found.
Please see diff file below:
+++ newlisp_manual-10.2.1-rev2.html Tue Mar 30 03:12:13 2010
@@ -2099,7 +2099,7 @@
; strings
-(set 's "NewLISP)
+(set 's "NewLISP")
(setf (s 0) "n") <span class='arw'>→</span> "n"
@@ -2590,7 +2590,7 @@
</pre>
<p>The <a href="#symbols">symbols</a> function is used to show all symbols
-belonging to a context"</p>
+belonging to a "context"</p>
<pre>
(symbols FOO) <span class='arw'>→</span> (FOO:func FOO:var FOO:x FOO:y FOO:z)
@@ -2896,7 +2896,7 @@
;; Variables used throughout this namespace
(define db:handle)
-(define db:host "http://localhost)
+(define db:host "http://localhost")
;; Constants
Thanks to Kosh and Sammo (in a different thread) for the latest manual corrections, online since yesterday:
http://www.newlisp.org/downloads/newlisp_manual.html
What could be the cause of this error?
called from user defined function br
called from user defined function h4
called from user defined function body
called from user defined function html
Is it that the list is too deeply nested, or that the list is too long?
sys-info reports this at the beginning:
(78309 268435456 501 8 0 2048 0 56523 10110 131)
and this just before the overflow:
(86380 268435456 501 8 0 2048 0 56523 10110 131)
but I can't see where a 'call stack' would be exceeded...?
Setting a value with the -s switch doesn't appear to change these figures:
Currently on a trip with internet on a phone only. Will look into it on Friday.
I assume you have the body tag defined as a 'begin' operation. The 'begin' function is limited to 4096 expressions, but you could use either 'lambda/fn' or 'dolist' to evaluate longer lists of expressions:
; or
(dolist (ex long-list-of-expressions)
(eval x))
Hi Lutz - thanks for the idea. But I think the problem may be elsewhere. Sorry about my code, it's not the best, but I'm just hacking to see if this is a valid path for a project.... Can't see where I've gone wrong. (And sorry about the wrong thread, too...:)
This code expects the manual to be in XHTML format...
(xml-type-tags nil nil nil)
(set 'xf (xml-parse f 31))
(set-ref-all 'div xf '_div)
(set-ref-all 'name xf '_name)
(set-ref-all '@ xf '_@)
(define (_@)
(dolist (ex (args))
(eval ex)))
(dolist (tag '(html a align alink b bgcolor blockquote
body border br cellpadding center class color
content em font h1 h2 h3 h4 head hr href
http-equiv i id li link media meta ol p
pre size span style summary table td text
th title tr tt type u ul valign vlink
width xmlns _div _name))
(letex
((f-name (sym (string tag))))
(define (f-name) (string "{" (join (args) "n") "}"))))
(println (eval xf))
ERR: call or result stack overflow : dolist<E5E3>
called from user defined function _@
called from user defined function span
called from user defined function pre
called from user defined function body
called from user defined function html
Put 'println' statements at the beginning of both 'dolist' loops (tag and _@) and see what is going on. This will show you the path of execution and the reason for the overflow message. Note that only callstack depth is reflected in 'sys-info' and only before the error occurs. The error itself will cause a stack cleanup. The result stack cannot be traced with 'sys-info'.
The other possibility would be:
to see what is happening.
Thanks, Lutz. Which is the call stack depth -
It says this just before the call or result stack overflow
(86755 268435456 449 10 0 2048 0 60948 10201 131)
I've gone through with the debugger - just can't work out what I should be looking for. The code is simple, but fails after about 21000 function calls.