timport -s <schema>  <datafiles>

ALTER USER user-name IDENTIFIED BY password

Changes user-name's password to password. You must be logged in as that user or _SYSTEM to perform this operation.

CREATE [UNIQUE] | {[METAMORPH] [INVERTED | COUNTER]} INDEX index-name

  ON table-name

  (column-name1 [DESC]

  [, column-name2 [DESC] ] ...)

CREATE TABLE table-name

  (column-name1 data-type 

  [, column-name2 data-type ] ...)

CREATE TABLE table-name AS 
  
  SELECT  {expression1 [, expression2]...}{...}

  FROM  table-name1 [alias1] [, table-name2 [alias2] ]...

  [WHERE     search-condition]

  [GROUP BY  column-name1 [, column-name2]...

  [ORDER BY  column-name3 [DESC] [, column-name4  [DESC] ] ...]

CREATE USER user-name IDENTIFIED BY password

Creates a new user with the specified user-name and password.

DELETE FROM table-name

  [WHERE search-condition]

DROP INDEX index-name

DROP TABLE table-name

DROP USER user-name

GRANT  {ALL}

       {privilege1 [,privilege2]...}

  ON   {table-name}

       {view-name}

  TO   {PUBLIC}

       {userid1 [,userid2]...}

See also: REVOKE, SYSUSERS system table, CHMOD(1), CHOWN(1)

INSERT INTO table-name[(col-name1 [, col-name2]...)]

  VALUES(value1 [, value2]...)

INSERT INTO table-name[(col-name1 [, col-name2]...)]

  SELECT-statement

UPDATE table-name

 SET column-name1 = expression1

  [, column-name2 = expression2]...

 [WHERE search-condition]

The first is that Texis WebScript will name the variable the same as the column name. If the default name of the column would be an illegal name, or would conflict with an existing variable, renaming the column lets you change the variable name at the same time.

The second situation is within the query itself. If you have a GROUP BY clause with either a HAVING clause or an ORDER BY clause, then any aggregate or calculated fields must be aliased, and the alias name used in the HAVING or ORDER BY clause. This is the only time the alias name can be used in the query.

<EXPORT $nquery>
.
.
<SQL "select Body, text2mm(Body) nquery from ...">
$Body
<A HREF=$url/morelike.html>Find Similar</A>
</SQL>

<A NAME=morelike>
<SQL "select id, Title from ... WHERE Body LIKEP $nquery">
<showrow>
</SQL>
</A>

SELECT cust.name, orders.text FROM cust,orders WHERE
       cust.name = 'John Doe' and orders.custid = cust.id;

• Integers
• Floating point numbers
• Character Strings

The range available in the numeric literals is machine dependent. You can typically count on at least the range in a long or double respectively.

Character strings are delimited by single quotes. To include a single quote in a string it should be doubled. For example:

'He''s one of us now.'

Function	Description
File Functions
fromfile(filename) fromfiletext(filename)	Returns the contents of a file
totext(text or filename)	Converts a PDF or wordprocessor file to text
toind(text)	Stores the text into a texis managed indirect and returns the filename
String Functions
abstract(text[, maxsize[, style[, query]]])	Abstract a block of text
text2mm(text[, words])	Generate a query from a block of text
length(text)	Returns the length of a block of text
lower(text)	Returns the lowercase form of a block of text
upper(text)	Returns the uppercase form of a block of text
initcap(text)	Returns the text with the first letter of every word in uppercase
sandr(search, replace, text)	Returns the text with search replaced by replace
Other Functions
exec(cmd[, args])	Allows execution of external commands
mminfo(query, text)	Find out hit offset information
convert(data, type)	Converts fields from one type to another
seq(increment[, init])	Returns a number increment than the last, or init
random(max[, seed])	Returns a random number less than max
math functions	acos, asin, atan, atan2, ceil, cos, cosh, exp, fabs, floor, fmod, log, log10, pow, sin, sinh, sqrt, tan, tanh
date functions	dayname, month, monthname, dayofmonth, dayofweek, dayofyear, quarter, week, year, hour, minute, second

The second, and more interesting use, is that it allows a table to be joined to itself. One example of this would be to generate a listing of employees and bosses.

SELECT boss.name Boss, emp.name Employee
    FROM employees emp, employees boss
    WHERE boss.id = emp.boss_id

SET property = value

Properties that affect Metamorph indices are kept with the index and do not need to be respecified. If you want to change the properties of an index you will need to drop it and re-create it. If searching must be maintained you can create a new index first with a different name, and then drop the old index.

The WHERE clause will affect which rows are selected from the table to be grouped. The HAVING clause will determine which groups are actually returned.

For example the query:

SELECT DEPT, AVG(SALARY) AVSAL FROM EMPLOYEES
       WHERE SALARY > 50000
       GROUP BY DEPT

SELECT DEPT, AVG(SALARY) AVSAL FROM EMPLOYEES
       GROUP BY DEPT
       HAVING AVSAL > 50000

<A NAME=name [PUBLIC|PRIVATE] [param[=value|$var] ...]>
  ... statements ...
</A>

Flag	Action	Default
PUBLIC	Function is user-visible: can be entry point to script.	If no functions are declared PUBLIC or PRIVATE
PRIVATE	Function is private: cannot be called outside of script.	If any function is declared PUBLIC or PRIVATE

<BREAK>

<CAPTURE [flags]>
  ...
</CAPTURE>

Flag	Action
BIN	Returned value is binary.
OUTPUT	Also print output normally.

Return value in $ret: output of statements within block.

<DB = path>

<EXEC [flags] program [arg ...]>
  ... input to program ...
</EXEC>

Flag	Action
NOBR	Don't split output into lines.
BIN	Output is binary. Implies NOBR.
BKGND	Execute in background; will not wait for program exit. Output ignored, $ret value is PID of program.

<EXPORT $var1 [$var2 ...] [flags]>

Vars saved to	Which vars	When saved
State table	Non-LOOP/SQL vars	First time $url printed when vars set
URL	LOOP/SQL vars	Every time $url printed inside a loop
URL query string	<EXPORT QUERY> vars	Every time $urlq printed

Flag	Action
URL	Always export these variables to URL
TABLE	Always export these variables to table
USEROK	User (form) variables may override state values
QUERY	Export to query string via $urlq

<IF condition1>
  ... statements if condition1 true ...
[<ELSEIF condition2>
  ... statements if condition2 true ...]
[<ELSE>
  ... statements if both are false ...]
</IF>

Or condition may be a SQL where clause.

<LOCAL var[=value|$var] ...>
  ...
[</LOCAL>]

Declares variables to be local. Scope ends at </LOCAL> tag, or end of enclosing block if not given.

<LOOP [SKIP=s] [MAX=m] [REV] $var1 [$var2 ...]>
  ... statements ...
</LOOP>

Iterates over named variables, one value per loop iteration.

Attribute	Value
SKIP	Initial number of values to skip
MAX	Max number of values to loop
REV	Whether to loop in reverse order

Variable	Value	When Set
$loop	Loop count (starts at SKIP, default 0)	Start of each iteration
$next	Next SKIP value	Start of each iteration, and at loop exit

<PUTMSG call|log|print|all|override on|off>

Directive: sets whether to log, print, and/or call <putmsg> function on errors.

<READLN [options] $file>
  ...
</READLN>

Option	Action
REV	Read file in reverse order, and start from end.
START=$expr	Start returning lines with first one to match REX expresion $expr.
END=$expr	Stop returning lines when one matches REX expression $expr.
MAXLINES=$lines	Read at most $lines from the file, regardless of START or END.
MAX=$n	Return at most $n lines, after START lines.
ROW	Don't loop $ret: replace previous value each line.
SKIP=$lines	Skip the first $lines returned lines (after START).

Return value in $ret: each line from file.

<RETURN [$ret]>

Exits current function, optionally setting $ret.

<SCHEDULE [options] WHEN=schedule [var=val] [...]>

Schedule script for periodic execution according to schedule, which is a vCalendar schedule or one of the following syntaxes:

minutes
[EVERY] [n] DAY[S]|DAILY timeofday [minutes]
[EVERY] [n] WEEK[S|LY] [ON] weekday timeofday [minutes]
[EVERY] [n] MONTH[S|LY] [ON] [THE] FIRST|SECOND|THIRD|FOURTH|n [TO LAST] weekday|DAY [OF THE MONTH] timeofday [minutes]

where:

minutes = [EVERY] [n] MIN[UTE[S]] [[REPEAT] n [TIME[S]]
timeofday = AT [h]h[:][[m]m] [A|P[M]]
weekday = SUN|MON|TUE|WED|THU|FRI|SAT

Option	Controls
START=date	Start schedule on or after Texis-parseable date (default now)
URLEND=/func.txt	Append /func.txt to URL when running script
COMMENT=text	Add comment to schedule listing for -LS command-line option

After all options, a list of variable assignments to pass to the script can be given.

<SQL [flags] "SQL command" [$cmd ...]>
  ... statements ...
</SQL>

Executes SQL command once, looping for each row returned. Returned fields become variables, one value added per iteration as in a <LOOP>.

Attribute	Value
SKIP=n	Initial number of rows to skip
MAX=n	Max number of rows to return
NOVARS	Flag: don't return variables from SQL statement
ROW	Flag: return vars in a single row, don't append into list
DB=db	Override <DB> for this statement
USER=user	Override <USER> for this statement
PASS=pass	Override <PASS> for this statement
NULL=$val	Override $null for this statement
NONULL	Flag: don't use $null for this statement

Variable	Value	When Set
$loop	Number of rows returned (does not include SKIP)	Start of each iteration
$next	SKIP value for next <SQL>	Start of each loop iteration and loop exit
$indexcount	Number of index rows (if index used)	Start of first iteration

Values of arguments after flags are concatenated to form SQL command. Embedded variable references in command become parameters.
If $null set, parameters matching it and their clauses are dropped from query.

<SQLCACHE = n>

Directive: number of Texis handles to cache for SQL statements.

<STACK=n>

Directive: sets Vortex call stack limit (-1 for no limit). Default is 250. One stack item used per function, block or loop call.

<SWITCH switchval>
  <CASE value1>
    ... statements if switchval = value1 ...
  [<CASE op value2>
    ... statements if switchval op value2 is true ...
  ...]
  [<DEFAULT>
    ... statements if no matches above ...
  ]
</SWITCH>

<TIMEOUT = n>
  ... HTML to print on timeout ...
</TIMEOUT>

<TIMPORT [flags] $schema [FROMFILE] $data|$file>
  ... statements ...
</TIMPORT>

Imports data from each element of buffer $data (or file $file if FROMFILE) using TIMPORT schema $schema. Returned fields become variables; loop executed once per row ala <SQL>.

Attribute	Value
SKIP=n	Initial number of rows to skip
MAX=n	Max number of rows to return
NOVARS	Flag: don't return any variables
ROW	Flag: return vars in a single row, don't append into list
FROMFILE	$file is filename list, not text buffer

Variable	Value	When Set
$loop	Number of rows returned (does not include SKIP)	Start of each iteration
$next	SKIP value for next <TIMPORT> (given same data)	Start of each iteration and loop exit

<TRACESQL on|off|N>

Directive: turns SQL statement tracing on or off.

<TRAP [CONNRESET=on|off|N] [[SIGNALS=]on|off]>

Directive: turns signal and connection-reset trapping on or off. For diagnostive purposes; not for general use.

<USER = user>
<PASS = pass>

<USES [DATE=date] module[=file] [module[=file] ...]>

List of modules to link in from library.

<VERB [NOESC]>
  ... raw text ...
</VERB>

<WHILE condition>
  ... statements ...
</WHILE>

Loops over statements while condition is true. Condition has same syntax as <IF>. $loop set every iteration, starting with 0 before first iteration. <BREAK> will also exit the loop.

<WRITE [APPEND] $file>
  ...
</WRITE>

Redirects output inside block to $file. Appends instead of truncates if APPEND given. If $file is "-" (single dash), output goes to stdout (ie. no-op). If $file is "--" (two dashes), output goes to top-level stdout (ie. "break through" outer <CAPTURE> etc. blocks).

Return value in $ret: nothing (no effect on $ret).

<abstract $text [$maxsz [$style [$query]]]>

<adminsql [arg ...]>

Execute an arbitrary SQL statement, the concatenation of args. Prints result rows in an HTML table. Return value in $ret: None

<apicp $name $value>

Query protection settings:

Name	Setting	Default
alpostproc	Allow post-processing	off
allinear	Allow linear searches	off
alequivs	Allow equivalences	off
alintersects	Allow @ intersections	off
alnot	Allow "NOT" logic (- operator)	on
alwithin	Allow /w within operator	off
alwild	Allow wildcards	on
qminprelen	Minimum character length of wildcard prefix	2
qminwordlen	Minimum character length of term	2
qmaxsets	Maximum allowed number of sets	100
qmaxsetwords	Maximum allowed number of words per set	100
qmaxwords	Maximum allowed number of words in overall query	100
denymode	Action to take on disallowed query	warning
texisdefaults	Set Texis defaults (allow linear, post-processing, etc.)	--
defaults	Restore Vortex defaults	--

Query processing settings:

Name	Setting	Default
exactphrase	Exact phrase processing	off
prefixproc	Prefix processing	off
suffixproc	Suffix processing	on
rebuild	Word rebuilding	on
incsd	Include start delimiter in hits	off
inced	Include end delimiter in hits	on
withinproc	Within (/w) processing	on
minwordlen	Minimum word length for prefix/suffix processing	255
intersects	Default number of intersections	255
see	Look up "see also" references during equivalence lookup	off
keepeqvs	Use equivalences for words/phrases	off
keepnoise	Keep noise words in query	off
sdexp	Default start delimiter expression	None
edexp	Default end delimiter expression	None
eqprefix	Equivalence file name	builtin
ueqprefix	User equivalence file name	/usr/local/morph3/eqvsusr; or USROBJECTS.tbl file in database for Webinator
suffix	Suffix list used during search	see manual
suffixeq	Suffix list used during equivalence lookup	see manual
prefix	Prefix list used during search	see manual
noise	Noise words	see manual

Return value in $ret: None

<cal [options] [$dates [$events [$var ...]]]>
[</cal>]

Prints a calendar of events with links. $dates are the dates, $events are the corresponding text to print. In the looping syntax, the programmer must print each day's <TD>, text and </TD>.

<caldate $varprefix>

<calendar $year $month $day $hour $min ["am"|pm" [$prefix [flags ...]]]>

<calendar $date [$prefix [flags ...]]>

<checkbox $name $values $selected $output>

Argument	Value
$name	Name of checkboxes
$values	Option values
$selected	Which $values should be SELECTED
$output	User-visible values

Return value in $ret: None

<clist $values>

Return value in $ret: $values in one parenthetical, comma-separated string

<code2geo $num>

<count $var>

Return value in $ret: Number of values of $var

<createdb $db $syspass $publicpass>

Creates the database $db with given _SYSTEM and PUBLIC passwords.
Return value in $ret: success or fail message

<decrypt $data $key>

<encrypt $data $key>

<exit [$code]>

Exits script, with optional exit code (default 0).

<fetch $theURL [$rawdoc]>

<fetch PARALLEL[=n] [URLS=]$urls [$loopvar ...]>
  ...
</fetch>

Return value in $ret: Raw (unformatted) document at URL $theURL, or contents of $rawdoc as if it were.

<flush>

Flushes any pending output. Returns nothing.

<fmt format [arg ...]>
<strfmt format [arg ...]>

Standard printf formats
Format code Prints argument as %d, %i Decimal integer %u Unsigned decimal integer %x, %X Hexadecimal integer %o Octal integer %e, %E Floating-point exponential %f Floating-point (no exponent) %g, %G Floating-point (%e or %f) %c Single character %s String %% % sign	Flag What Applies to formats - Left adjustment All + Sign positive numbers All numeric (space) Leading space for positive numbers All numeric 0 Pad with 0's instead of spaces All numeric # Alternate format %o, %x, %X, %e, %E m.n Width, precision All except %ms, %mU, %mH h Arg is short int Integer numeric l Arg is long int Integer numeric

Additional formats/flags

Format code Prints argument as %t Date/time (Texis date or counter field) %T UTC (Universal time) date/time %U URL encode string %H HTML encode string %V UTF-8 encode an ISO-8859 (ASCII) string %v UTF-16 encode a UTF-8 string %B base64 encode string; field width is chars per line to print %b Binary integer %r, %R Roman numeral %F Fractional output of float

Flag Subflag What Applies to m   Metamorph query and markup; next arg is query %s, %U, %H   b Bold hit highlighting   h HTML hit highlighting (default)   n No backoff of hits inside <>   p Paragraph formatting: double newlines replaced with <P>   P Paragraph formatting: next arg (REX expression) replaced with <P>   c Continue subhit count into next query a   Next arg is strftime() format %t, %T k   Put a "," every 3 places to left of decimal %d, %i, %u, %f, %g K   Put next arg (string) every 3 places to left of decimal &   Pad with &nbsp; instead of space %s, %d, %i, %u, %f, %g _ (underscore)   Pad with ASCII 160 byte (&nbsp; code) instead of space %s, %d, %i, %u, %f, %g _ (underscore)   Skip (don't print) leading BOM in UTF-16 output (%v); save (don't delete) leading BOM in UTF-16 input (%!v) %v, %!v < (less-than)   Force UTF-16LE (little-endian) input/output %v, %!v > (greater-than)   Force UTF-16BE (big-endian) input/output %v, %!v !   Decode instead of encode %H, %U, %V, %v, %B, %c h   HTML-escape out-of-range (>255 for %!V, >0x10FFFF for %v) characters instead of printing as ? %!V, %v h   HTML-unescape sequences in source data %V, %!v hh   Also HTML-escape low (7-bit) characters %V, %!V, %v, %!v l (el)   Only encode low (7-bit) characters, or decode to 7-bit (multi-char) strings %H q   Full encoding: encode forward-slash and at-sign too %U p   Paragraph markup for %s and %H codes; path escapement (%20 instead of +) for %U code %s and %H; %U w   Pass argument as "whopping" integer size (max available, usually 64-bit). Automatically set. %d, %i, %u, %x, %o

Escape sequences in format string:

Escape	Meaning
\n	Newline (ASCII 10)
\r	Carriage return (ASCII 13)
\t	Tab (ASCII 9)
\a	Bell (ASCII 7)
\b	Backspace (ASCII 8)
\e	Escape (ASCII 27)
\f	Form feed (ASCII 12)
\v	Vertical tab (ASCII 11)
\\	Backslash
\xhhh	Hexadecimal escape; hhh is 1 to 3 hex digits.
\0ooo	Octal escape; ooo is 0 to 3 octal digits.

Return value in $ret: None for <fmt>, printed string for <strfmt>

<geo2code $lat $lon [$radius]>

<getpid>

<getvar $var>

<hash $buf>

Returns a counter type hash value for text buffer $buf. Different buffers almost certainly produce different hash values.

<header NAME=$name [VALUE=$value]>
<header COOKIE=$cookie [VALUE=$value] [cookie-options]>

Prints HTTP header, if called before any output is generated. Prints generic header $name with $value, or Set-Cookie header if COOKIE used. More useful alternative to .bin URL syntax.

COOKIE Attribute	Value
MAX-AGE=seconds	Maximum age of cookie, in seconds
EXPIRES=date	Expiration time of cookie (alternate to MAX-AGE)
PATH=path	Root URL path cookie will be returned for
DOMAIN=.domain	Domain cookie will be returned for

<kill $pid [$sig]>

<loadavg>

Returns the system load averages in 3 values of $ret.

<loguser $REMOTE_ADDR [$limit]>

Return value in $ret: success or fail message (fail if over $limit (5) unique users in the past minute)

<lower $str>

Returns each value of $str in all lowercase.

<mm> ... $variables ... </mm>

<nsinfo $name [$which]>

<nslookup [options] $host|$ip>

<nslookup PARALLEL[=n] [options] [HOSTS|ADDRS=]$hosts|$ips [$loopvar ...]>
  ...
</nslookup>

Looks up $host or IP address $ip and returns IP address or host name.

Parallel version: n at a time (default all) of $hosts or $ips list are looked up. $loop, $next set each loop iteration, results returned fastest first.

Option	Action
BYADDR	Force lookup by address (IP to host)
BYNAME	Force lookup by name (host to IP)
MTERR	Return empty string on lookup failure

<options $values $selected $output>

Argument	Value
$values	Option values
$selected	Which $values should be SELECTED
$output	User-visible values

Return value in $ret: None

<pdfxml $query $Body [$color]>

Returns Metamorph hit information on PDF document $Body for a PDF browser plugin.

<procexists $pid>

<profiler INIT $table [$profile [$field]]>
<profiler GET $msg [$profile]>

Help manage a table of fixed message queries.

<pwencrypt $password $salt>

<radiobutton $name $values $selected $output>

Argument	Value
$name	Name of buttons
$values	Option values
$selected	Which $values should be SELECTED
$output	User-visible values

Return value in $ret: None

<rand $min $max>

Return value in $ret: Random integer N, $min <= N < $max

<randpick $var>

Return value in $ret: One randomly chosen value of $var

<read $files [$offsets [$lengths]]>

Return value in $ret: Contents of each file in $files (1 file per value). Reading starts at $offset into file, default 0, negative offsets are from end of file. Length read is $length, default rest of file, negative length is rest of file.

<readvars $varnames $urldata>

Reads variables from URL-encoded strings in $urldata. Only vars named in $varnames (if non-empty) are assigned.

<resetstats $REMOTE_ADDR>

Return value in $ret: None

<rex $expr $data>                      <split $expr $data>
  or                                     or
<rex [options] $expr $data>            <split [options] $expr $data>
  ...                                    ...
</rex>                                 </split>

Return value in $ret: Data from $data matching REX expressions in $expr for <rex>; non-match data for <split>. Non-looping version by default. If any options given, looping syntax instead, and $loop/$next/$ret.off set.

Option	What it does
ROW	Do not accumulate hits in $ret
SKIP=$n	Skip the first $n hits
MAX=$n	Return at most $n hits
SPLIT	Return non-match data instead (<split> behavior)
NONEMPTY	Return only non-empty hits

<rmcommon $data $template [$maxrm]>

Removes the prefix and suffix text, on a word boundary, from each $data value, that is shared with the corresponding $template value. Up to $maxrm characters are removed (default: maximum amount of common text).

<sandr $expr $replace $data>

Return value in $ret: Values of $data with each matching $expr REX expression replaced with corresponding $replace value

<sb> ... $variables ... </sb>

<send $values>

Prints $values as-is. Return value in $ret: None

<setvar [APPEND] $var $values>

<sleep $seconds [wake]>

If wake given, return value in $ret: Number of seconds remaining (eg. could be non-zero if signal received)

<slist $intersect $values>

Return value in $ret: $values as Metamorph set query for $intersect intersections

<sort $var [ICASE|NUM] [DESC|ASC] [$var2 ...]>

Flag	Action
ICASE	Ignore case, and sort alphabetically
NUM	Sort numerically
DESC	Sort descending
ASC	Sort ascending

Flags apply per-variable.
Return value in $ret if 1 variable, otherwise variables sorted in place.

<spew $files [$offsets [$lengths]]>

Prints contents of each of the $files as-is, starting from $offsets into files (default start of file, negative offset is from end of file), for $lengths bytes (default rest of file, negative length is rest of file). Return value in $ret: None

<sqlcp $name $value>

<srand $seed>

Seeds the pseudo-random number generator.

<stat [options] $path> [</stat>]

Option	Controls
ROW	Do not accumulate return variables in arrays
SKIP=n	Skip first n results
MAX=n	Return at most n results after initial SKIP
MAXDEPTH=n	Recursively traverse subdirectories too, to a maximum depth of n. Default is 0; -1 is no limit.
NAME=wildcard	Only return results whose file name matches shell-style wildcard expression
ALL	Return . and .. directory entries too
SAMEDEV	Do not cross filesystems from original $path when traversing directories
DEPTHFIRST	Returns directories' entries before the directory itself when traversing directories
SYMLINK	Return information about symlinks, not the files they point to (Unix)
FOLLOWSYM	Also traverse symlinks that point to directories
GLOB	Expand shell-style wildcards in initial $path first
SORT=method	Sort each directory's contents by method, which can be none, name (default), size, atime, mtime or ctime
ASC	Sort each directory's contents in ascending order (default)
DESC	Sort each directory's contents in descending order

Return Var	Type	Value
$ret	varchar	File path
$ret.err	varchar	Error from stat() or empty on success
$ret.depth	long	Number of dirs traversed from top-level $path
$ret.symlink	varchar	Raw target of symlink (if Unix and path is a symlink)
$ret.sympath	varchar	Corrected path of symlink
$ret.size	long or double	Size of file
$ret.owner	varchar	Owner of file
$ret.group	varchar	Group of file
$ret.isrd	long	1 if file is readable, 0 if not
$ret.iswr	long	1 if file is writable, 0 if not
$ret.isex	long	1 if file is executable, 0 if not
$ret.mode	varchar	10-character ls style permissions/mode string
$ret.atime	date	Last access time of file
$ret.mtime	date	Last modify time of file
$ret.ctime	date	Last status change time of file
$ret.nlinks	long	Number of hard links to file, if filesystem supports it
$ret.devtype	long	Device type file is on (if filesystem supports it)
$ret.dev	long	Device major/minor number (Unix) or 0 for A:, 1 for B: etc. under Windows
$ret.ino	long	Inode of the file (if supported)
$ret.blks	long	Number of blocks consumed by file (if supported)
$ret.blksize	long	Preffered block size for file transfers (if supported)

<strcmp $a $b>
<strcmpi $a $b>

Return value in $ret: an integer for each comparison of each value of $a (as a string) against each value of $b:

• < 0 if $a < $b
• 0 if $a = $b
• > 0 if $a > $b

<strlen $str>

Returns length of each given string value.

<strncmp $a $b $len>
<strnicmp $a $b $len>

<strrev $str>

Returns each value of $str, reversed.

<strstr $needle $haystack>
<strstri $needle $haystack>

Return value in $ret: Indices into each $haystack value of each $needle value. Indices start at 0 for first character; if not found value is -1.
<strstr> is case-sensitive, <strstri> is not.

<strtonum $text>

Returns: each $text string converted to a number.

<submit [options] URL=$theURL [var=$value ...]>

Option	Value
URL=$theURL	URL of form. FTP URLs supported (for GET, PUT, and DELETE methods).
FROMFILE=$file	File to send content from; ``-'' for stdin.
TOFILE=$file	File to write received content to; ``-'' for stdout. Content-derived fields in <urlinfo> etc. will be unavailable.
METHOD=$method	HTTP method (OPTIONS, GET, HEAD, POST, PUT, DELETE or TRACE)
DATA=$data	Raw data to send instead of variables.
CONTENT-TYPE=$type	Content-type of raw data to send.
var=$value	Variable to send with form

Return value in $ret: Result document from form submission

<substr $str $offset $len>

Returns substring starting at offset (0 is first char, < 0 is from string end) of length $len (< 0 is rest of string) for each corresponding string $str.

<sum $fmt $var [$var2 ...]>

Values of $var	Return value in $ret
All numeric (integers or simple floats)	Sum of values, then formatted and cast with $fmt code
One or more non-numeric	Each value formatted with $fmt, then string-concatenated

<syscp $what [$arg ...]>

Controls various system or OS-level parameters.

Command	Action
chdir	Change the current working directory to the argument

<sysinfo $what [$arg]>

Gets system/OS information.

What	Returns
time	Wall-clock time; same as date, but sub-second precision (double)
proctime	User, system, and real time used by process, in seconds (double). If given, $arg is subtracted
procmem	Virtual size and resident set size of process, in bytes.
physmem	Physical memory (RAM) of machine, in bytes.
getpid	Process id
getppid	Parent process id
procexists	1 if process id $arg exists, 0 if not
loadavg	3-value system load average (double)
cwd or pwd or getcwd	Current working directory
platform	Hardware-vendor-OS-filebits-ptrbits platform description string
regquery or regqueryvalue	(Windows only) Registry value, given key subkey name [default]
maxdescriptors	Max allowed open file descriptors (-1 if unknown)
opendescriptors	Number of currently open file descriptors (-1 if unknown)

<sysutil $action [options ...] $arg [...]>

Action and Args	What It Does
mkdir or md $dir [...]	Creates directories
chown or lchown $owner $file [...]	Changes ownership of files (or symlinks if lchown). $owner is [owner][.][group]. Unimplemented under Windows.
chgrp or lchgrp $group $file [...]	Changes the group of files. Unimplemented under Windows.
chmod $mode $file [...]	Changes file permissions to symbolic or numeric $mode.
rm or del or delete $file [...]	Removes files.
rmdir or rd $dir [...]	Removes directories.
mv or rename $src [...] $dest	Moves each $src file to $dest.
cp or copy $src [...] $dest	Copies each $src file to $dest.
link or symlink $target [...] $name	Creates a hard or soft link $name to the given $target. Unimplemented under Windows.

Option	What It Does
msg	Turns on putmsg for this call
nomsg	Turns off putmsg for this call
glob	Wildcard-expand source file argument(s)
recurse	Recursively descend source directories
force	Force action, even if no file write permission
preserve	Preserve file attributes for cp
asis	Copy non-directories as-is instead of as files for cp
parents	Create needed parent directories too for mkdir
nnn or [ugoa]=[rwx]	Directory permissions for mkdir
--	End of options: following arguments are files

Returns empty string on success, otherwise a string describing the first error.

<uniq [SORTED] $var [ICASE|NUM] [DESC|ASC] [$var2 ...]>

Returns unique, sorted list of the given variables.

Flag	Action
SORTED	Assume vars are already sorted as per given flags
ICASE	Ignore case, and sort alphabetically
NUM	Sort numerically
DESC	Sort descending
ASC	Sort ascending

SORTED flag applies globally.
Return value in $ret if single variable, otherwise variables modified in place.

<uniqcount>

<upper $str>

Returns each value of $str in all uppercase.

<urlcp $param [$value1 [$value2]]>

Possible $param settings: integer

Name	Type	Controls	Default
timeout	integer	Timeout, in seconds, for <fetch>, <submit>, <nslookup>	30
maxhdrsize	integer	Maximum header size, in bytes	128KB
maxpgsize	integer	Maximum raw page size, in bytes	512KB
maxredirs	integer	Maximum number of redirects	5
getframes	boolean	Whether to fetch frames	no
getiframes	boolean	Whether to fetch iframes	no
maxframes	integer	Maximum number of frames fetched per document	5
offsiteok	boolean	Whether to fetch off-site redirects and frames	yes
user	string	User name for password-protected pages (Basic authentication)	none
pass	string	Password for password-protected pages	none
proxyuser	string	User name for password-protected proxy (Proxy-Authorization header)	none
proxypass	string	Password for password-protected proxy	none
redirmsgs	boolean	Whether to show redirect informational messages	no
useragent	string	User agent header	Mozilla-compatible
8bithtml	boolean	Whether to leave 8-bit (foreign) HTML characters as-is	yes
alttxt	boolean	Whether to include ALT text from image tags	yes
strike	boolean	Whether to include <STRIKE> text in formatted output (from <urltext>)	yes
del	boolean	Whether to include <DEL> blocktext in formatted output (from <urltext>)	yes
select	boolean	Whether to include <SELECT> block text in formatted output (from <urltext>)	yes
input	boolean	Whether to include <INPUT TYPE=text> values in formatted output (from <urltext>)	yes
textarea	boolean	Whether to include <TEXTAREA> block text in formatted output (from <urltext>)	yes
formtxt or formtext	boolean	Alias for select, input and textarea together	yes
minclrdiff	integer	Minimum foreground/background color difference for formatted text to appear	0
eatlinkspace	boolean	Whether to trim leading/trailing whitespace from links before processing.	true
urlnonprint	string	How to treat non-printable (outside ASCII range 32-126) characters in URL links: asis, strip or encode	asis
urlcanonslash	Boolean	Whether to canonicalize backslashes in URLs to forward slashes	True
proxy	string	The URL to use as a proxy server	none
reparentmode	string	URL reparenting mode: abs, reroot, mirror or off	None
reparent	string	Re-parent links in HTML of fetched documents. If absolute URL, sets reparentmode abs and links made absolute, as if page were that URL. If full path, sets reparentmode reroot, links made relative (if possible) as if that were document root.	Empty
reparentimg	Boolean	Re-parent image links also (if reparent set).	False
ifmodsince	string	The HTTP If-Modified-Since conditional-get header	Unset
accept	2 string lists	The HTTP Accept header list	Everything (*/*)
header	list, 2 arguments	Set the HTTP headers given in first arg to second arg values.	None
clearheaders	no args	Clears any headers set with header.	--
dnsmode	sys or int	Host name lookup routines to use	int
nameservers	list	Nameservers used for host name lookup	System dependent
dnsretrans	integer	Initial host name lookup retry interval	System dependent
dnsretry		Max number of retries for host name lookups	System dependent
dnsservices	List of host or bind	Services to use and in what order, for host name lookup	System dependent
dnsdomains	List of domains	Domains to search when resolving hostnames without embedded dots	Host dependent (from /etc/resolv.conf)
dnsignoretrunc or dnsigntc	Boolean	Whether to ignore truncated-message flag in DNS replies	True
fileexclude	List of file prefixes	File prefixes to exclude when fetching local file:// URLs	None on Windows; /dev/, /proc/ and /debug/ on Unix
fileinclude	List of file prefixes	File prefixes to include (require) when fetching local file:// URLs	None
filenonlocal off|unc|ftp	String	How to treat non-localhost file:// URLs	off
fileroot	File prefix	File root for local file:// URLs	None
filetypes [add|del|set] [file|dir|device|symlink|other ...]	List	File types to allow for local file:// URLs	None
scriptvaryevents	Boolean	Whether to vary form elements' values to generate more JavaScript events	True
scripturldecode	Boolean	Whether to URL-decode javascript: links	True
protocols [add|del|set] [$protocols ...]	list	List of allowed protocols to fetch	http, https, ftp, gopher, javascript
linkprotocols [add|del|set] [$protocols|allowed|unknown ...]	list	List of protocols to allow in returned links from a page	unknown, http, https, ftp, gopher, javascript, file
pagecachesz	integer	Size of page cache for frames/iframes/scripts	5MB
getscripts	Boolean	Whether to fetch <SCRIPT SRC=...> scripts	True
javascript	Boolean	Whether to execute inline JavaScript <SCRIPT>...</SCRIPT> blocks and enable other JavaScript features	False
scriptevents [add|del|set] [$events ...]	list	List of JavaScript events to trigger after page fetch	all
scriptlinks	Boolean	Whether to run javascript: links on a page before returning from fetch	True
scriptmaxtimer	integer	Max time in seconds to run JavaScript timers	3
scriptrealtimers	Boolean	Whether to run JavaScript timers in real time	False
scriptmem	integer	Max memory to allow JavaScript engine to allocate	20MB
scriptmsgs	Boolean	Whether to print JavaScript errors	False
scripttimeout	integer	Max total time to allow JavaScript to run on a page	5
cookies	Boolean or integer	Sets acceptcookies and sendcookies together	3
acceptcookies	Boolean	Whether to acceptcookies	True
sendcookies	Boolean	Whether to send cookies	True
cookiejar $text [append]	string	Sets the cookie jar	As received from fetches
cookiemsgs	Boolean	Whether to print errors for invalid cookies	False
embedsecurity off|nodecrease|noincrease|sameprotocol	string	Whether to fetch embedded objects on a page based on main/embedded protocol security	off
addentropy $entropy [$fraction]	string	Seed SSL plugin with random data $entropy of fractional randomness $fraction	Check entropy daemon if needed
entropypipe $file	string	Path to entropy daemon pipe for SSL plugin init	Check texis.cnf, or etc/egd-pool in install dir
verbose	integer	Sets debugging verbosity level: 1 for commands received, 2 for commands sent too.	0
tracedns	integer	Host name lookup tracing (debug)	0 (off)
traceskt	integer	Socket level tracing (debug)	0 (off)
defaults	No arguments	Resets all settings

<urlinfo $name [$which]>

Name	Type	What
actualurl	string	The last URL retrieved. It may differ from the argument to <fetch> or <submit> if redirects were followed.
contenttype	string	The MIME content type of the page. This may be derived from the Content-Type header, a <META HTTP-EQUIV> tag, or the URL extension, depending on what is available.
contenttypeparams	list	The names of parameters in the MIME content type, if any.
contenttypeparam	list	The value(s) of the content type parameter(s) named $which. Multiple values may be given in $which.
frames	list	The list of frame URLs in the document. If the <urlcp> setting getiframes is true, the list is empty since the iframes have been fetched and inserted into the document.
iframes	list	The list of iframe URLs in the document. If the
headers	list	The names of HTTP headers received with the document.
header	list	The value(s) of the HTTP header(s) listed in $which.
errnum	integer	The last error code (not HTTP response code).
errmsg	string	The last error message (not HTTP message).
httpcode	integer	The value of the HTTP response code.
httpmsg	string	The HTTP response message.
images	list	The list of image URLs in the document, eg. <IMG> tags, background images, etc.
links	list	The list of non-image link URLs in the document, eg. <A HREF> tags, <FORM> tags, etc. Same as the return value of the <urllinks> function.
metaheaders	list	The names of <META HTTP-EQUIV> tags in the document.
metaheader	list	The value(s) of the <META HTTP-EQUIV> tag(s) listed in $which.
metanames	list	The names of <META NAME> tags in the document.
metaname	list	The value(s) of the <META NAME> tag(s) listed in $which.
rawdoc	string	The raw document source. Same as the return value of the original <fetch> or <submit>.
redirs	integer	The number of redirects encountered.
text	string	The formatted text of the document. Same as the return value of the <urltext> function.
title	string	The title of the document.
time	double	The total time taken to fetch the document.
dnstime	double	The time taken to resolve the hostname(s) while fetching the document.
transfertime	double	The time taken to transfer the data for the document.
cookiejar [all]	string	The contents of the cookie jar in Netscape cookie-file format.
prngdpid [$path]	integer	The process ID of the prngd daemon running at Unix pipe $path.

<urltext>
<urllinks>

<urlutil $action [$arg1 [$arg2]]>

URL parsing utility.

Action	What it does
split	Splits URL $arg1 into parts $arg2 (protocol, host, port, path, type, anchor or query)
abs	Absolutes (fully-qualifies) relative links $arg2 as if they were on absolute URL $arg1

<userstats $REMOTE_ADDR [license|user]>

Return value in $ret: Number of accesses by given user in past minute, either <loguser> stats (default or if user given), or license stats (license). Empty $REMOTE_ADDR gives current number of users; high gives highest ever seen.

<varinfo $action [$arg]>

Action	What it returns
list	Names of all variables in current scope. Optional $arg can be URL, ENV, CONTENT, CGI or COOKIE for specific source.
size	Size in bytes of each value of variable named in $arg.
type	SQL type of variable named by $arg.
filename	Filename for variable named by $arg, from Content-Disposition header, if multi-part file upload.
contenttype	The Content-Type for the variable named by $arg, if multi-part file upload.

<vxcp setting [args ...]>

Setting	Controls	Arguments
putmsg	Error message disposition	2 args: call, log, print or all, and a boolean value
stack	Maximum Vortex stack depth	Integer
timeout	Vortex script timeout	Integer or Texis date
trap	Trap signals or not (debug)	Boolean value
connreset	Trap connection-reset or not	Boolean value
filemsg	putmsg for <sysutil> actions or not	Boolean value
htmlmode	Whether to HTML-encode variables on output	Boolean value
libpath	The path to search for loadable modules	String value

<vxinfo $what [$arg] ...>

Gets miscellaneous Vortex information.

What	Returns
outputsz	Number of bytes printed by the script so far
db	The current database
globaldb	The global database for the script
release	Release date of this Vortex executable
version	Vortex version string
installdir	Directory Texis was installed in
htmlmode	1 if variables are being HTML-escaped, 0 if not
license	License value $arg, which is one of violationmsg, violationtime, gentime, expiretime, verifytime, verifytrytime, serial, curhits, curhitstime, maxhits, curtblrows, maxtblrows, curtblsz, maxtblsz, curdbrows, maxdbrows, curdbsz, maxdbsz, curtotrows, maxtotrows, curtotsz, maxtotsz, texismonitorpid, equivpath, uequivpath, vortexlog, defaultdb, defaultscript, vortexflags

<wordlist $table [$field [$prefix]]>
<wordcount>

Returns a list of words in $table, as found in a Metamorph index. Uses index for $field if given. $prefix can be a single word, or prefix followed by * to limit words.

<xtree $command [$values] [$treename]>
  or
<xtree [options] $command [$values] [$treename]>
  ...
</xtree>

Manages an in-memory list of string values. Non-looping syntax by default. If looping option(s) given, loop syntax expected and $ret.count, $ret.seq, $loop and $next set as well.

Loop option	What it does
ROW	Do not accumulate values in $ret
SKIP=$n	Skip the first $n return values
MAX=$n	Return at most $n values

Command	Action
CLEAR	Clear tree (reset counts to 0)
DUMP	Dump tree to $ret
COUNT	Dump corresponding value counts to $ret (non-loop only; loop sets $ret.count during search/dump)
SEQ	Dump corresponding sequence numbers to $ret (non-loop only; loop sets $ret.seq)
INSERT	Insert $values into tree; returns $values and counts/sequences
DELETE	Delete $values from tree
SEARCH	Return list of matching $values in tree
SET	Set options given in $values (see below)
FLUSH	Remove tree and free its memory

Option	Action
NOADD	Don't add new values to tree during INSERT
CLEAR	Also clear tree during DUMP

<$var = [value1 ...]>

<$var = (expression)>

• Make the query as precise as possible. The fewer the rows to be returned, the faster the response will be.
• The index is most useful for finding keywords and concepts, so try to include at least one such term in the query.

... LIKE 'iraq gulf %qadhafi'

... LIKE '~cat'

If you want to supply the search terms which comprise the query term yourself you can do so as follows:

... LIKE '(cat,tiger,kitten)'

... WHERE text LIKE 'dog';

... WHERE text LIKE 'dog cat';

... LIKE 'cancer #>0<1'

• ... 20 percent of cancer victims ...
• ... two-thirds of those suffering from lung cancer ...

... LIKEP 'john smith'

John Smith
John D. Smith
Smith, John

... LIKE '"John Smith"'

Expression	Meaning
line	within a line
sent	within a sentence
para	within a paragraph
page	within a page
all	within the field
NUMBER	within NUMBER characters
REX Expression	use expression as delimiter

Examples:

Find cat and dog withing a paragraph

...LIKE 'cat dog w/para'

...LIKE 'president congress w/\RSECTION'

'john david michael +smith @1'

+

Required set. Not counted in permutation.

-

Excluded set. Do not include hits which match this term.

@N

Permutations. Requires N intersections of remaining sets. (N + 1 of the sets must match) Default is all of them.

In the ranking queries (LIKER, LIKEP), the intersects operator is ignored, and the + and - take on a different meaning. Normally the weight of a word is determined by its frequency. A + will give that word the maximum possible weight. A - will take the weight of the term, and negate it, so a document containing the weight will have its rank reduced.

... LIKE 'screwdriver /\alpha{3}\digit{4} w/line'

Within Rex the meaning of various characters is as follows:

Flags	Meaning
\R	Respect Case
\I	Ignore Case (default)
\L	Toggle Literal mode. Removes special meaning from characters
\F \P	Followed/Preceded By. Allows the finding of an expression following or in front of another. The preceding or following expression is not counted as part of the hit.
!	NOT the following expression
>>	Use the following expression as the root, which means it will be found first.
Character Class	Meaning
\alpha \upper \lower \digit \xdigit \alnum \space \punct \print \graph \cntrl \ascii	Match the same characters that are would be matched by the respective is...() function, e.g. isalpha()
[]	Matches any single character from those contained between the brackets. Ranges can be given with a '-' (e.g. a-z0-9). A '^' as the first character inverts the set. Case is respected.
Single Characters	Value
\xNN	NN is the hexadecimal code of the character
$	End of line. (Appropriate for operating system)
.	Any Single Character
Other	Matches itself
\\	Matches a backslash
Repetition Operators	Meaning
{x}	expression repeated x times
{x,y}	repeated between x and y times
*	zero or more occurences
+	one or more occurences

Note that the repetition operators apply to entire preceding expression, not just the last character. So

\digit{1}abc{2,3}