nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Trying to use extended constructs in proxy map

awizardly

Does anyone know if the iwproxy server is supposed to support extended constructs?
I keep getting a failure to validate while using the following line in the [iwproxy_preconnect_remap]

_regex=(.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/(?!es)(.*)=$1/US/SITE/beta/en/working/STAGING/htdocs/$2

I've tested the regex in a perl script and it works just fine

This is a short script testing a similar regex
#!/global/site/vendor/InterWoven/iw-home/iw-perl/bin/iwperl -w
use re 'debug';
my $value = "htdocs/foo/bar/index.html";

if ( $value =~ m#.*htdocs/(?!en/)(.*)# ){
print "$value matches pattern\n";
} else {
print "$value doesn't match pattern\n";
}
print "$1\n";

The use re 'debug' is a pragma for testing regular expressions.
Also from the print statement you will see that $1 gets set to everything after htdocs/

Basically I am trying to get the proxy to match on anything that isn't in the htdocs/es directory and redirect it to another branch in this case the english branch.

I keep getting
failed: regex pattern set for: (.*)/US/Chrysler/us/es/working/WORKAREA/[^/]+/htdocs/(?!es)(.*)
when I run iwproxy -d

Does anyone know if iwproxy actually uses the perl interpreter, or is it really only using a subset when parsing mappings?

Lee
Organic Inc
http://www.organic.com

Edited by awizardly on 12/09/05 04:56 PM (server time).

Find more posts tagged with

Comments

Adam Stoller

I'm pretty sure that proxy does *not* use a Perl RE engine - I think it uses a REGEX(5) library - possibly REGEX(5) extended - but whichever I'm almost positive it doesn't understand the (?!...) syntax.

I think you'd have to do something like:

_regex=(.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^e][^s].*)=$1/US/SITE/beta/en/working/STAGING/htdocs/$2

Frankly, iwproxy's regex processing is probably overdue for an overhaul - for reasons such as this - so you should probably file a FR for this.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

awizardly

I've made some more progress, thanks to ghoti's help. I've abandoned the (?!es/) construct that seemed to work so well in perl,
and I am now using ([^e][^s][^/].*) to accomplish the same basic match.

in iw.cfg
_regex=(.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^e][^s][^/].*)=$1/US/SITE/us/en/working/STAGING/htdocs/$2

This gets me the actual html page of the initial request, but it doesn't seem to correctly map the rest of the assets the same way.

What I am effectively trying to do is have all assets on in the es directory come from another branch location, I get the html page, but the rest of my assets don't seem to be pulled in correctly. I am watching iwproxy -x go by and I am seeing something that puzzles me

Checking (.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^e][^s][^/].*) -> $1/US/SITE/us/en/working/WORKAREA/ognc_det/htdocs/$2
bmap_pair::rewrite() regex: regex compare failed against '/iw-mount/default/main/US/SITE/us/es/working/WORKAREA/ognc_det/htdocs/js/hbx.js': 1:regexec() failed to match

This doesn't make sense because beforehand I have
================================================================
[Request from client:]
GET /iw-mount/default/main/US/SITE/us/es/working/WORKAREA/ognc_det/htdocs/brand/index.html HTTP/1.1
... sanitized ...
IWPROXY_PATHTRACK=/iw-mount/default/main/US/SITE/us/es/working/WORKAREA/ognc_det/htdocs/js/hbx.js
Host: localhost:1080
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.10) Gecko/20050716 Firefox/1.0.6
X-Forwarded-For: 192.168.240.126
X-Forwarded-Host: analog.organic.com
X-Forwarded-Scheme: http
X-Forwarded-Server: analog.organic.com

[Using pathtrack cookie: '/iw-mount/default/main/US/Chrysler/us/es/working/WORKAREA/ognc_det/htdocs/js/hbx.js']
Checking (.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^e][^s][^/].*) -> $1/US/SITE/us/en/working/WORKAREA/ognc_det/htdocs/$2
bmap_pair::rewrite() regex: initial expansion loop
initial expansion complete: /iw-mount/default/main/US/Chrysler/us/en/working/WORKAREA/ognc_det/htdocs/brand/index.html
[iwproxy_preconnect_remap:]
/iw-mount/default/main/US/SITE/us/es/working/WORKAREA/ognc_det/htdocs/brand/index.html
---> /iw-mount/default/main/US/SITE/us/en/working/WORKAREA/ognc_det/htdocs/brand/index.html

Lee
Organic Inc
http://www.organic.com

Edited by awizardly on 12/12/05 09:38 AM (server time).

jbonifaci

Your regular expression is checking for:

../htdocs/ followed by not an e followed by not an s followed by not a / followed by anything.

The path coming through is ../htdocs/js/ So it is doesn't match for two reasons. The second charaster after /htdocs/ is an s and the third character after /htdocs/ is a /. So the match fails.

~Jeff

jbonifaci

Try doing something like this:
_regex=(.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^e]...*|.[^s]..*|..[^/].*)=$1/US/SITE/us/en/working/STAGING/htdocs/$2~Jeff

awizardly

I see the problem you are pointing out.

Because js also has an s as a second character it's failing the match correctly.

A colleague of mine worked out a simpler solution from what you had up there. It involves making the match inside the square braces atomic with parens

([^(es/)].*)

_regex=(.*)/US/SITE/us/es/working/WORKAREA/[^/]+/htdocs/([^(es/)].*)=$1/US/SITE/us/en/working/WORKAREA/ognc_det/htdocs/$2

Thanks to reddy who is also on devnet, but didn't post his solution instead poked me in the next cube with the solution.

Lee
Organic Inc
http://www.organic.com

reddyl

Thanks Lee.

jbonifaci

I'm fairly good with regex, but I'll have to admit, I didn't know you could put strings in character classes. I always use extended negate matches for that, so it's usually not an issue, but it's good to know in situations where you can't use them like this.

~Jeff