by a webctrl.ini... this allows redirect scripts to notice.

Also, clear extra_path_info when we handle an internal redirect so it doesn't
use stale data.

If present, this JS code will be executed before any SSJS or Rewrite code
for this request.

Also, populate js.startup_dir and js.exec_dir as the directory the webctrl.ini
is in so that you cna do something like:
JSPreExec=load(js.exec_dir+'/coolstuff.js');

This now allows super-cool rewrite thingies.

work for rewrite scripts.

Fix up the rewrite logging to be more helpful and a bit less verbose.

Added by keys beginning with "Rewrite", ie:

RewriteDetail=var n=http_request.request_string.replace(/_detail\/(.*)/,"lib/exe/detail.php?media=$1"); if(n != http_request.request_string) { http_request.request_string=n; true }

If the expression returns "true", reparses http_request.request_string as
an internal redirect.

I plan to have another key that will pre-execute a script from a file so that
these don't need to be an entire script on one line in the INI file.

This API is subject to change.

May contain peanuts.

the extra path.

Set the SCRIPT_NAME to include the index filename... this was apparently
changed to emulator a specific mod_rewrite rule for a specific package and
is wrong.

update the path info data and re-extract the filename so that widcards in
section names can match path info index files.

longer than 127 bytes.

the script until the entire POST data is read, so is sub-optimal, but it
works.

Now tested on FreeBSD (should work for any *nix).
CGI and FastCGI still needs testing on Win32.

Add support for FastCGI with an eye to making PHP significantly faster.  In
the webctl.ini file, add something like the following:

[*.php]
FastCGI=localhost:9000

To pass all PHP requests under that directory through to a PHP fastCGI server.
You can start the PHP FastCGI server with a command something like this:
php-cgi -q -b localhost:9000

This change has also refactored normal CGI support, especially the Win32
support.  Since this patch is currently untested, it's best to assume that
it breaks both *nix and Win32 CGI support and doesn't work for FastCGI either
at this point.

(read email)

- Improved displaying of messages with ANSI codes. The reader will fall back to a non-scrolling user interface for displaying messages with ANSI codes, since there doesn't seem to be an easy way for the reader to handle ANSI with the scrolling interface (i.e., with ANSI cursor movements, etc.).
- Bug fix related to interpreting colors from other BBS software (WWIV, PCBoard, Wildcat, Celerity, Renegade)
- Work in progress: More kludge lines displayed (with the 'K' key), and all message header lines are now displayed (with the 'H' key) (this is a sysop feature)

1. If the PATH kludge line only contains 1 address, don't use it for circular
path detection.
2. If the last address of the PATH kludge line is the system's address, don't
add the system's address to the PATH.

to be more globally friendly.

If the following filenames don't exist, don't try to display them (i.e. don't
report/log an error if the file doesn't exist):
text/sbbs.msg
text/system.msg
text/newuser.msg
text/feedback.msg
Removed hard-coded clear-screen and pause sequences before/after each of these
files are displayed (instead, moved the clear-screens to @-codes in the stock
files). This gives the sysop more control over what is displayed to the user
during the new-user registration process.
NOTE: You may want to review your new-user process and modify your copies/
versions of the above mentioned filenames to make sure they display to the user
as you wish.

(needed for ircmsg.js send-failure handling).

due to age (date in header). Would log file path in addition to "!Filtering QWK
message from <author> due to age: x days" message. Found by Deuce's code
review.

Use jsdoor_revision* instead.

Added sort and find functions.
Added syncterm.lst generation function.

Instead, wait 0-seconds (by default) for a keypress after successful transfers
and 60-seconds (by default) for a keypress after unsuccessful transfers.
These durations can be changed via the TransferSuccessKeypressTimeout
and TransferFailureKeypressTimeout values in the syncterm.ini file.
ToDo: abort the wait more immediately upon key input
ToDo: display the count down timer
ToDo: add menu options in SyncTERM to edit these timeout values

directory.  This is the beginnings of the ability to write a door in JS for
all BBSes, not just Synchronet (door kit pending).

This requires moving js_CreateCommonObjects() into main.cpp, which requires
constifying the name parameter to js_CreateClientObject() due to C++ being
more paranoid about using a string literal as a char * parameter
(warning fix).

path only if needed.

When parsing the SBBSecho configuration file (for the USEPACKER, AREAFIX, or
ECHOLIST items), flags or packer names would be ignored if they began with the
same sequence of characters which made up a previous configuration item
(flag or packer name). This appears to be a *very* long standing bug going back
to the origination of this program by Allen Christiansen (King Drafus).
Also, long overdue brace style update. Steve Deppe's influence is diminishing.
-(

in the system's areas.bbs file, SBBSecho would terminate prematurely with an
error: "No areas defined in <path/to>areas.bbs". This would prevent SBBSecho
from importing inbound netmail and packing outbound netmail, for example.

Added minLength option. Input must be at least this lengthy before datasource functions will be called.  (Default: 1)

Datasource functions can return arrays of strings or objects. Restrict width and/or girth of input text. Updated comments.