[racket-users] Re: resources for learning JS / React?

[Sean Kemplay]

This is a good (free) course that takes you the lates best practices of JS 
(getting more functional), react and then react native.

https://www.edx.org/course/cs50s-mobile-app-development-with-react-native


On Saturday, January 25, 2020 at 6:56:57 PM UTC, 'John Clements' via 
users-redirect wrote:
>
> I have a graduate student that wants a self-guided introduction to JS and 
> React. The problem here, to some degree, is that there are so *many* 
> introductions. Does anyone here have specific references that might be 
> helpful? (Say, e.g., if Gregor Kiczales did a JS course on coursera… that 
> would be pretty much perfect.) 
>
> John 
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/c433373a-3135-426d-90f3-f5d79032d38c%40googlegroups.com.

Re: [racket-users] Create an identifier in syntax

[Sean Kemplay]

Hi Ryan,

Thank you SO much for that explenation! Everything now clicks!!

Sean

On Thursday, January 23, 2020 at 4:12:44 PM UTC, Ryan Culpepper wrote:
>
> On 1/23/20 3:59 PM, Sean Kemplay wrote: 
> > Hello, 
> > 
> > I am exploring macros and am trying to define a variable at the top 
> > level (with the goal in mind to dynamically define a group of functions 
> > from a macro). 
> > 
> > with-syntax works fine however I was just wondering if it is possible to 
> > directly inject an identifier as syntax within syntax - something like 
> > the following which does not work! 
> > 
> > (define-syntax x 
> >  (lambda (x) 
> >#`(define ,#'y "y val"))) 
> > 
> > (x) 
> > y => y: undefined; 
> >   cannot reference an identifier before its definition 
>
> First, to escape a quasisyntax (#`) template you need to use unsyntax 
> (#,), not unquote (,). 
>
> Second, due to hygiene the y from the macro has an extra scope, so you 
> can't refer to it by typing y at the top level. But you can do this, for 
> example: 
>
>(define-syntax (x2 stx) 
>  #`(begin (define #,#'y "y val") y)) 
>
> Or you can write a macro that defines a y with the lexical context of 
> the macro use: 
>
>(define-syntax (x3 stx) 
>  #`(define #,(datum->syntax stx 'y) "y val")) 
>
> You could also write this macro with with-syntax instead. The way that 
> you insert an identifier into a syntax template (quasisyntax/unsyntax vs 
> with-syntax) is independent of the way you create the identifier. 
>
> (Note: using the lexical context of the macro use works here, but it's 
> not always the right answer. Unhygienic macros are complicated.) 
>
> Ryan 
>

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/ceaef6ec-8bca-428f-af8c-cfb71eddd08f%40googlegroups.com.

[racket-users] Create an identifier in syntax

[Sean Kemplay]

Hello,

I am exploring macros and am trying to define a variable at the top level 
(with the goal in mind to dynamically define a group of functions from a 
macro).

with-syntax works fine however I was just wondering if it is possible to 
directly inject an identifier as syntax within syntax - something like the 
following which does not work!

(define-syntax x
(lambda (x)
  #`(define ,#'y "y val")))

(x)
y => y: undefined;
 cannot reference an identifier before its definition

this is just to satisfy my own curiosity :-)

Cheers,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/5836fddb-fbc4-464c-87b8-68cff47b69f2%40googlegroups.com.

Re: [racket-users] Abstract run times (HTDP)

[Sean Kemplay]

Thanks Matthias, I think I have it now, much appreciated.

On Friday, May 24, 2019 at 2:24:57 PM UTC+1, Matthias Felleisen wrote:
>
>
>
> > On May 24, 2019, at 6:27 AM, Sean Kemplay  > wrote: 
> > 
> > Hi all, 
> > 
> > I trying to work out the ART of the sum-tree question from HTDP2e. 
> > 
> > I have worked through the book before however am going through it again, 
> this time am doing every exercise - this is not homework! 
> > 
> > Here is my data defenitions, examples, a function ror summing up the 
> contents and tests - 
> > 
> > ; A Pair is a list of two items 
> > 
> > ; A Number-Tree is one of: 
> > ; Number 
> > ; [Pair-of Number-Tree] 
> > ; Examples - 
> > ; 0 
> > ; (list 0 1) 
> > ; (list (list 1 2) 3) 
> > ; (list (list 1 2) (list 3  4)) 
> > ; (list (list (list 1 2) (list 3 4)) (list (list 5 6) (list 7 8))) 
> > ; (list (list (list (list (list (list (list 8 7) 6 ) 5) 4) 3) 2) 1) 
> > 
> > ; Number-Tree -> Number 
> > ; Sum up all numbers in nt 
> > (check-expect (sum-list 0) 0) 
> > (check-expect (sum-list (list 1 2)) 3) 
> > (check-expect (sum-list (list (list 1 2) (list 3  4))) 10) 
> > (check-expect (sum-list(list (list (list 1 2) (list 3 4)) (list (list 5 
> 6) (list 7 8 36) 
> > (check-expect (sum-list (list (list (list (list (list (list (list 8 7) 6 
> ) 5) 4) 3) 2) 1)) 36) 
> > (define (sum-list nt) 
> >   (cond [(number? nt) nt] 
> > [else (+ (sum-list (first nt)) 
> >  (sum-list (second nt)))])) 
> > 
> > 
> > Using a balanced tree my analysis shows an ART of n^2 to run sum-list 
> where n is the depth of the tree 
> > (sum-list(list (list (list 1 2) (list 3 4)) (list (list 5 6) (list 7 
> 8 
> > 
> > 
> > Using avery unbalanced tree with all nodesheading down to the left my 
> analysis shows an ART of 2n - or just n when the constant is removed. 
> > (sum-list (list (list (list (list (list (list (list 8 7) 6 ) 5) 4) 3) 2) 
> 1)) 
> > 
> > So the unbalanced tree in this case where all nodes need to be visited 
> would be better? 
>
>
> Equally good. 
>
>
> > A BST a search function would be better with a balanced tree but not 
> sum-tree? 
>
> Yes. 
>
>
> > Could someone possibly comment whether I am on the right track here? 
>
> Good job. 
>
>
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/491b3a19-824b-4e8b-9c2e-7e2f8ff1eabe%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Abstract run times (HTDP)

[Sean Kemplay]

Hi all,

I trying to work out the ART of the sum-tree question from HTDP2e. 

I have worked through the book before however am going through it again, 
this time am doing every exercise - this is not homework!

Here is my data defenitions, examples, a function ror summing up the 
contents and tests -

; A Pair is a list of two items

; A Number-Tree is one of:
; Number
; [Pair-of Number-Tree]
; Examples -
; 0
; (list 0 1)
; (list (list 1 2) 3)
; (list (list 1 2) (list 3  4))
; (list (list (list 1 2) (list 3 4)) (list (list 5 6) (list 7 8)))
; (list (list (list (list (list (list (list 8 7) 6 ) 5) 4) 3) 2) 1)

; Number-Tree -> Number
; Sum up all numbers in nt
(check-expect (sum-list 0) 0)
(check-expect (sum-list (list 1 2)) 3)
(check-expect (sum-list (list (list 1 2) (list 3  4))) 10)
(check-expect (sum-list(list (list (list 1 2) (list 3 4)) (list (list 5 6) 
(list 7 8 36)
(check-expect (sum-list (list (list (list (list (list (list (list 8 7) 6 ) 
5) 4) 3) 2) 1)) 36)
(define (sum-list nt)
  (cond [(number? nt) nt]
[else (+ (sum-list (first nt))
 (sum-list (second nt)))]))


Using a balanced tree my analysis shows an ART of n^2 to run sum-list where 
n is the depth of the tree
(sum-list(list (list (list 1 2) (list 3 4)) (list (list 5 6) (list 7 8


Using avery unbalanced tree with all nodesheading down to the left my 
analysis shows an ART of 2n - or just n when the constant is removed.
(sum-list (list (list (list (list (list (list (list 8 7) 6 ) 5) 4) 3) 2) 1))

So the unbalanced tree in this case where all nodes need to be visited 
would be better?

A BST a search function would be better with a balanced tree but not 
sum-tree?

Could someone possibly comment whether I am on the right track here?

Kind regards,
Sean



-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-users/4210ccdc-9d9e-434a-8dab-12b3d7b424d0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] lists.racket-lang.org seems to be down

[Sean Kemplay]

H All,

Just reporting that https://lists.racket-lang.org/ seems to be down.

I have tried on both my laptop and phone.

Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [racket-users] Subproceses in Windows

[Sean Kemplay]

Thanks George, that explains it perfectly.

On Monday, November 5, 2018 at 5:11:04 PM UTC, gneuner2 wrote:
>
>
>
> On 11/5/2018 11:49 AM, Sean Kemplay wrote:
>
> Hi All,
>
> I am trying to open windoes explorer from Racket using the following -
>
> (system* "cmd" "start" "explorer.exe")
>
> However it is not working and #f is being returned. This works fine from 
> Go and even VBScript!
>
> Anyone know what I am missing?
>
> Kind regards,
> Sean
>
>
>
> On Windows the path isn't searched (for whatever reason).  You need to do 
> something like:
>
> (let [
>   (path (find-executable-path "explorer.exe"))
>  ]
>   (system* path)
>   )
>
>
> George
>

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Subproceses in Windows

[Sean Kemplay]

Hi All,

I am trying to open windoes explorer from Racket using the following -

(system* "cmd" "start" "explorer.exe")

However it is not working and #f is being returned. This works fine from Go 
and even VBScript!

Anyone know what I am missing?

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Webserver HTTP 2

[Sean Kemplay]

Hello,

Does anyone know if the Racket webserver will support http2 at any stage?

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [racket-users] [Library Announcement] CSS-expressions, Pollen Component and Extensible Functions

[Sean Kemplay]

I will definitely check out the CSS package this weekend!

On 13 Feb 2017 13:48, "Leandro Facchinetti"  wrote:

> Hi all,
>
> I’m here to announce three packages I published recently, because I want
> feedback from the community:
>
> - CSS-expressions: S-expression-based CSS.
>   https://docs.racket-lang.org/css-expr/index.html
>
> - Pollen Component: Component-based development for Pollen.
>   https://docs.racket-lang.org/pollen-component/index.html
>
> - Extensible Functions: A solution to the expression problem in Typed
> Racket.
>   https://docs.racket-lang.org/extensible-functions/index.html
>
> Thank you.
> --
> Leandro Facchinetti 
> https://www.leafac.com
> GPG: 0x5925D0683DF3D583
>
> --
> You received this message because you are subscribed to the Google Groups
> "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to racket-users+unsubscr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [racket-users] Safely using port->string where nothing has been written to the port

[Sean Kemplay]

Hi David,

Indeed it does!! Thanks for your help.

Sean

On Monday, January 23, 2017 at 7:16:58 PM UTC, David K. Storrs wrote:
> Hi Sean,
> 
> Does (byte-ready?) work for your case?
> 
> https://docs.racket-lang.org/reference/Byte_and_String_Input.html?q=peek#%28def._%28%28quote._~23~25kernel%29._byte-ready~3f%29%29
> 
> 
> 
> On Mon, Jan 23, 2017 at 6:05 AM, Sean Kemplay <sean.k...@gmail.com> wrote:
> Hello,
> 
> 
> 
> I am using the URL library to interact with a REST API.
> 
> 
> 
> One of the Endpoints sends returns no content for a POST request with a 
> status code of 204 (a little unusual however it is an external API so we have 
> to cater to it).
> 
> 
> 
> I can check the status code for 204 and only apply string->port if it isn't a 
> 204 - however I was wondering if there is any check that can be made at the 
> port level that no content will be written - in case we encounter a malformed 
> response which does not send through a body.
> 
> 
> 
> Currently without the checks port->string will hang on a 204 response.
> 
> 
> 
> This would be good to know more generally when dealing with ports as well.
> 
> 
> 
> Kind regards,
> 
> Sean
> 
> 
> 
> --
> 
> You received this message because you are subscribed to the Google Groups 
> "Racket Users" group.
> 
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to racket-users...@googlegroups.com.
> 
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Safely using port->string where nothing has been written to the port

[Sean Kemplay]

Hello,

I am using the URL library to interact with a REST API. 

One of the Endpoints sends returns no content for a POST request with a status 
code of 204 (a little unusual however it is an external API so we have to cater 
to it).

I can check the status code for 204 and only apply string->port if it isn't a 
204 - however I was wondering if there is any check that can be made at the 
port level that no content will be written - in case we encounter a malformed 
response which does not send through a body. 

Currently without the checks port->string will hang on a 204 response.

This would be good to know more generally when dealing with ports as well.

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [racket-users] Testing an HTTPS (over HTTP CONECT) proxy

[Sean Kemplay]

Hi Tim,

This is great, the whole proxy thing has been an issue for me as well - I took 
a stab at doing something like this last year but couldn't get it working.

I have access to a trend and scansafe proxy so can physically test against 
those if you want.

Kind regards,
Sean

On Tuesday, August 9, 2016 at 2:37:03 PM UTC+1, Jay McCarthy wrote:
> Here are some testing options:
> 1) Implement a proxy server in Racket
> 
> 2) Fake a proxy server that just repeats what you know to be a good
> session (by dumping an interaction with a real server)
> 
> 3) Require squid or something to be present when testing
> 
> 4) Find a public proxy server that can be used to test
> 
> --
> 
> I recommend 2.
> 
> Jay
> 
> On Tue, Aug 9, 2016 at 9:28 AM, Tim Brown  wrote:
> > Folks,
> >
> > Fed up with not being able to install racket (or generally use raco) from 
> > behind
> > a firewall, I have a PR https://github.com/racket/racket/pull/1411 open 
> > which allows
> > me to use HTTPS over an HTTP CONNECT (e.g. squid) proxy.
> >
> > I need to document and test my changes.
> >
> > Documentation is a straightforward enough exercise.
> >
> > However, I’m a bit stuck with testing. I have a new clone of 
> > tim-brown/racket
> > and a `make` is building, DOWNLOADING and INSTALLING the packages. The 
> > capitals
> > are because I am so excited! So, as far as I can tell: it “works for me”.
> >
> > I guess that isn’t what most engineers would call a test-suite, however.
> > Since the code deals with proxy servers -- which are an 
> > installation-specific
> > thing. I can write tests like the ones in pkgs/net-test/tests/net/url.rkt 
> > -- which
> > test parsing proxy server names. But there is nothing to make sure that 
> > proxying
> > connections works.
> >
> > I was wondering if anyone has suggestions for writing tests for this.
> >
> > Tim
> >
> > PS: The make is no longer building.
> > It is fully builded!
> >
> > --
> > Tim Brown CEng MBCS 
> > 
> > City Computing Limited · www.cityc.co.uk
> >   City House · Sutton Park Rd · Sutton · Surrey · SM1 2AE · GB
> > T:+44 20 8770 2110 · F:+44 20 8770 2130
> > 
> > City Computing Limited registered in London No:1767817.
> > Registered Office: City House, Sutton Park Road, Sutton, Surrey, SM1 2AE
> > VAT No: GB 918 4680 96
> >
> > --
> > You received this message because you are subscribed to the Google Groups 
> > "Racket Users" group.
> > To unsubscribe from this group and stop receiving emails from it, send an 
> > email to racket-users+unsubscr...@googlegroups.com.
> > For more options, visit https://groups.google.com/d/optout.
> 
> 
> 
> -- 
> Jay McCarthy
> Associate Professor
> PLT @ CS @ UMass Lowell
> http://jeapostrophe.github.io
> 
>"Wherefore, be not weary in well-doing,
>   for ye are laying the foundation of a great work.
> And out of small things proceedeth that which is great."
>   - D 64:33

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Transforming data for plot / discreet-histogram

[Sean Kemplay]

Hello,

Something like the following would be ideal to build a histogram from a 
sequence of values read in via CSV

#lang racket
(require math)
(require plot)

(discrete-histogram (samples->hash '(1 2 3 4 4)))

However discreet histogram takes a list of vectors. Is there a way to achieve 
something like the above without managing a list of vectors for 500,000+ rows 
of data?

Reading the docs, I thought discrete-histogram would also take a list of lists 
and that I could do something like (discrete-histogram (hash->list 
(samples->hash '(1 2 3 4 4 but it looks like hash->list returns a list of 
dotted pairs.

Sorry, I dip in and out of Racket so I may be missing something obvious. Maybe 
run through hash-map?

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Web-server multi core

[Sean Kemplay]

Hi all,

There have been multiple queries about utilising multiple cores with Racket's 
webserver and also some ideas put forward by Jay and others.

Has anyone got something working they would be willing to share? I am really 
hoping to use racket over clojure for a new project but making use of all 
server CPU cores is important.

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Nested HTML templates

[Sean Kemplay]

Hello,

Has anyone had any experience using web-server/template?

If so do you have any suggestions on how to best/compose so there is a base 
template with placeholders being filled by other templates (which could in turn 
have even deeper nesting themselves)?

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] Re: raco pkg install behind proxy

[Sean Kemplay]

Hi Tim,

The net/http-client library (last time I looked) does not currently support SSL 
tunneling which would be required for ssl connections through a proxy such as 
squid. 

net/url builds on net/http-client and also does not have that support.

Link to what Squid has to say about tunneling - 

http://wiki.squid-cache.org/Features/HTTPS 

I was looking into SSL tunneling myself a few months ago and got partway there 
but it is on the backburner at the moment. If it were to be baked into 
net/http-client though, everything else would benefit.

Kind regards,
Sean

On Friday, October 9, 2015 at 2:39:26 PM UTC+1, Tim Brown wrote:
> Folks,
> 
> I have created a package, which needs to be deployed locally behind a
> Squid HTTP proxy.
> 
> My info.rkt has:
> (define deps
>  (list
>   "db-lib"
>   "web-server-lib"
>   ...))
> 
> I want to keep my racket installation as minimal as possible.
> 
> My initial problem comes because raco doesn’t honour any proxy settings
> (in .racketrc, or in the environment). That means that I don’t even get
> off my network.
> 
> I tweaked net/url.rkt with some hard-coded proxies (subject of another
> email):
> 
> (define current-proxy-servers
>   (make-parameter '(("http" "proxyname" 3128)
> ("https" "proxyname" 3128)) ...))
> 
> This then got me the following error:
> 
> ssl-connect: connect failed (error:140770FC:SSL
> routines:SSL23_GET_SERVER_HELLO:unknown protocol)
> 
> I’m stumped as to how I resolve this, since I don’t know where this is
> thrown up from -- squid, openssl, mzssl, git, what?
> 
> Can anyone help me out here?
> 
> Full transcript of woe below the dotted line.
> 
> Tim
> 
> . (dotted line)
> 
> PLTSTDERR=debug@net/url minimal-racket/bin/raco pkg install
>--fail-fast -i mypackage.tgz
> 
> The following uninstalled packages are listed as dependencies of
> mypackage:
>db-lib
>web-server-lib
>...
> Would you like to install these dependencies? [Y/n/a/c/?] Y
> 
> 00: Resolving "db-lib" via http://pkgs.racket-lang.org
> tcp-connect: connection failed
> detail: host not found
> address: pkgs.racket-lang.org
> port number: 80
> step: 1
> system error: Name or service not known; errno=-2
> context...:
> .../minimal-racket/share/racket/collects/net/http-client.rkt:224:0
> .../minimal-racket/share/racket/collects/racket/contract/private/arrow-val-first.rkt:324:3
> .../minimal-racket/share/racket/collects/net/url.rkt:77:0:
> http://getpost-impure-port
> .../minimal-racket/share/racket/collects/net/url.rkt:179:2: redirection-loop
> .../minimal-racket/share/racket/collects/racket/contract/private/arrow-val-first.rkt:324:3
> .../minimal-racket/share/racket/collects/pkg/private/network.rkt:58:3
> 
> 
> I can circumvent this by hard-coding some proxy-servers into
> current-proxy-servers in net/url.rkt, and repeat the exercise:
> 00: Resolving "db-lib" via http://pkgs.racket-lang.org
> Downloading repository git://github.com/racket/db/?path=db-lib
> ssl-connect: connect failed (error:140770FC:SSL
> routines:SSL23_GET_SERVER_HELLO:unknown protocol)
> context...:
> .../minimal-racket/share/racket/collects/openssl/mzssl.rkt:1401:8: loop
> .../minimal-racket/share/racket/collects/openssl/..:261:28
> .../minimal-racket/share/racket/collects/openssl/..:259:25
> .../minimal-racket/share/racket/collects/net/http-client.rkt:224:0
> .../minimal-racket/share/racket/collects/racket/contract/private/arrow-val-first.rkt:324:3
> .../minimal-racket/share/racket/collects/net/url.rkt:77:0:
> http://getpost-impure-port
> .../minimal-racket/share/racket/collects/net/url.rkt:179:2: redirection-loop
> .../minimal-racket/share/racket/collects/racket/contract/private/arrow-val-first.rkt:324:3
> .../minimal-racket/share/racket/collects/net/git-checkout.rkt:204:0:
> initial-connect
> .../minimal-racket/share/racket/collects/net/git-checkout.rkt:40:2:
> retry-loop
> .../minimal-racket/share/racket/collects/pkg/private/download.rkt:101:2:
> download!
> .../minimal-racket/share/racket/collects/file/cache.rkt:63:2:
> fetch-and-continue
> .../minimal-racket/share/racket/collects/racket/contract/private/arrow-val-first.rkt:324:3
> .../minimal-racket/share/racket/collects/pkg/private/download.rkt:93:0:
> download-repo!24
> .../minimal-racket/share/racket/collects/pkg/private/stage.rkt:299:11
> 
> -- 
> Tim Brown CEng MBCS 
> 
> City Computing Limited · www.cityc.co.uk
>   City House · Sutton Park Rd · Sutton · Surrey · SM1 2AE · GB
> T:+44 20 8770 2110 · F:+44 20 8770 2130
> 
> City Computing Limited registered in London No:1767817.
> Registered Office: City House, Sutton Park Road, Sutton, Surrey, SM1 2AE
> VAT No: GB 918 4680 96

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from 

Re: [racket-users] HTTPS connection through proxy (CONNECT HTTP Method)

[Sean Kemplay]

On Wednesday, August 12, 2015 at 9:54:23 PM UTC+1, Sean Kemplay wrote:
 On Wednesday, August 12, 2015 at 12:54:10 PM UTC+1, Jay McCarthy wrote:
  On Tue, Aug 11, 2015 at 10:24 AM, Sean Kemplay sean.kemp...@gmail.com 
  wrote:
   Hi All,
  
   Sending an http request through our corporate proxy works as follows for 
   http requests -
  
   (define-values (x y z)
 (http-sendrecv 10.0.0.200 http://www.example.com;
#:port 8080
#:headers '(
Proxy-Authorization: Basic 
   base64encodedcredentials
Proxy-Connections: keep-alive
)
#:ssl? #f
#:method GET))
  
   However fails for HTTPS requests (as expected).
  
   What I need to do is make a request like the above using the #:method 
   CONNECT and then make a secondary request through a returned connection.
  
   Does anyone know how I would go about doing that in Racket?
  
  http-sendrecv combines calls to http-conn-open, http-conn-send!,
  http-conn-recv!, then http-conn-close!. I suspect that you just need
  to break up that one big call into a few calls like open, send, recv,
  send, recv, close. I'd be happy to work on it with you, but I don't
  have such a proxy on hand, so I'll need helping testing it.
  
  Jay
  
  
   Kind regards,
   Sean
  
   --
   You received this message because you are subscribed to the Google Groups 
   Racket Users group.
   To unsubscribe from this group and stop receiving emails from it, send an 
   email to racket-users+unsubscr...@googlegroups.com.
   For more options, visit https://groups.google.com/d/optout.
  
  
  
  -- 
  Jay McCarthy
  http://jeapostrophe.github.io
  
 Wherefore, be not weary in well-doing,
for ye are laying the foundation of a great work.
  And out of small things proceedeth that which is great.
- DC 64:33
 
 Hi Jay,
 
 Thanks for that, yes I think you are right. I have just installed squid at 
 home which also supports http tunnelling. I'll see how I get on and post my 
 results - whether they be good or bad!
 
 It would be really good to at least get an example on the wiki for others to 
 build from, as I suspect a lot of corporate networks are behind proxies and 
 this would be problematic in using Racket to make calls to REST APIs etc 
 which my job at least requires a lot of.
 
 Kind regards,
 Sean

Hi Jay,

Unfortunately I am not getting very far with this.

Our app servers where our production code sits are not behind a proxy, so at 
the end of the day it doesn't rule out using Racket for some of the tasks I 
have in mind. It would be nice to be able to get through the proxy from my 
desktop to test code though.

I tried the following but the connection is ending early -

#lang racket
(require net/http-client)

(define conn (http-conn-open 10.0.0.200 #:port 8080))

(http-conn-send! conn https://news.ycombinator.com/; #:method CONNECT 
#:headers '(Proxy-Authorization: Basic base64encodedcredentials Connection: 
Keep-Alive) #:version #1.1)
(define-values (a b c)(http-conn-recv! conn #:close? #f))
(http-conn-send! conn / #:method GET); #:headers '(Proxy-Authorization: 
Basic Y3hnXHNlYW4ua2VtcGxheTpBdWd1c3QyMDE0 Connection: Keep-Alive) #:version 
#1.1)
(define-values (e f g) (http-conn-recv! conn))
(http-conn-close! conn)

I am basing the above on this S/O post but am not entirely sure if this is even 
the path I should be pursuing!

http://stackoverflow.com/questions/11697943/when-should-one-use-connect-and-get-http-methods-at-http-proxy-server

Here is an example using Node, it opens the connection to the uri through the 
proxy, then writes to that connection through an SSL connection which is 
something I can't see how to do in Racket.

http://blog.vanamco.com/proxy-requests-in-node-js/

I will keep digging, if you have any Ideas or anything I could test against our 
network it would be much appreciated. 

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
Racket Users group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [racket-users] HTTPS connection through proxy (CONNECT HTTP Method)

[Sean Kemplay]

On Friday, August 14, 2015 at 10:22:03 AM UTC+1, Sean Kemplay wrote:
 On Wednesday, August 12, 2015 at 9:54:23 PM UTC+1, Sean Kemplay wrote:
  On Wednesday, August 12, 2015 at 12:54:10 PM UTC+1, Jay McCarthy wrote:
   On Tue, Aug 11, 2015 at 10:24 AM, Sean Kemplay sean.kemp...@gmail.com 
   wrote:
Hi All,
   
Sending an http request through our corporate proxy works as follows 
for http requests -
   
(define-values (x y z)
  (http-sendrecv 10.0.0.200 http://www.example.com;
 #:port 8080
 #:headers '(
 Proxy-Authorization: Basic 
base64encodedcredentials
 Proxy-Connections: keep-alive
 )
 #:ssl? #f
 #:method GET))
   
However fails for HTTPS requests (as expected).
   
What I need to do is make a request like the above using the #:method 
CONNECT and then make a secondary request through a returned 
connection.
   
Does anyone know how I would go about doing that in Racket?
   
   http-sendrecv combines calls to http-conn-open, http-conn-send!,
   http-conn-recv!, then http-conn-close!. I suspect that you just need
   to break up that one big call into a few calls like open, send, recv,
   send, recv, close. I'd be happy to work on it with you, but I don't
   have such a proxy on hand, so I'll need helping testing it.
   
   Jay
   
   
Kind regards,
Sean
   
--
You received this message because you are subscribed to the Google 
Groups Racket Users group.
To unsubscribe from this group and stop receiving emails from it, send 
an email to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
   
   
   
   -- 
   Jay McCarthy
   http://jeapostrophe.github.io
   
  Wherefore, be not weary in well-doing,
 for ye are laying the foundation of a great work.
   And out of small things proceedeth that which is great.
 - DC 64:33
  
  Hi Jay,
  
  Thanks for that, yes I think you are right. I have just installed squid at 
  home which also supports http tunnelling. I'll see how I get on and post my 
  results - whether they be good or bad!
  
  It would be really good to at least get an example on the wiki for others 
  to build from, as I suspect a lot of corporate networks are behind proxies 
  and this would be problematic in using Racket to make calls to REST APIs 
  etc which my job at least requires a lot of.
  
  Kind regards,
  Sean
 
 Hi Jay,
 
 Unfortunately I am not getting very far with this.
 
 Our app servers where our production code sits are not behind a proxy, so at 
 the end of the day it doesn't rule out using Racket for some of the tasks I 
 have in mind. It would be nice to be able to get through the proxy from my 
 desktop to test code though.
 
 I tried the following but the connection is ending early -
 
 #lang racket
 (require net/http-client)
 
 (define conn (http-conn-open 10.0.0.200 #:port 8080))
 
 (http-conn-send! conn https://news.ycombinator.com/; #:method CONNECT 
 #:headers '(Proxy-Authorization: Basic base64encodedcredentials 
 Connection: Keep-Alive) #:version #1.1)
 (define-values (a b c)(http-conn-recv! conn #:close? #f))
 (http-conn-send! conn / #:method GET); #:headers '(Proxy-Authorization: 
 Basic Y3hnXHNlYW4ua2VtcGxheTpBdWd1c3QyMDE0 Connection: Keep-Alive) 
 #:version #1.1)
 (define-values (e f g) (http-conn-recv! conn))
 (http-conn-close! conn)
 
 I am basing the above on this S/O post but am not entirely sure if this is 
 even the path I should be pursuing!
 
 http://stackoverflow.com/questions/11697943/when-should-one-use-connect-and-get-http-methods-at-http-proxy-server
 
 Here is an example using Node, it opens the connection to the uri through the 
 proxy, then writes to that connection through an SSL connection which is 
 something I can't see how to do in Racket.
 
 http://blog.vanamco.com/proxy-requests-in-node-js/
 
 I will keep digging, if you have any Ideas or anything I could test against 
 our network it would be much appreciated. 
 
 Kind regards,
 Sean

Just an update on this, looking at the code for http-client I now understand 
that http-conn is a struct with an input and output port from a tcp connection.

Based on the node.js example I am thinking of instead of calling 
http-conn-send! a second time with a different method I need to write a 
function along the lines of http-conn-tunnel! which somehow pipes ssl input and 
output ports from the tcp input and output ports from http-conn's input and 
output pipes.

-- 
You received this message because you are subscribed to the Google Groups 
Racket Users group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https

Re: [racket-users] HTTPS connection through proxy (CONNECT HTTP Method)

[Sean Kemplay]

On Friday, August 14, 2015 at 3:41:04 PM UTC+1, Sean Kemplay wrote:
 On Friday, August 14, 2015 at 10:22:03 AM UTC+1, Sean Kemplay wrote:
  On Wednesday, August 12, 2015 at 9:54:23 PM UTC+1, Sean Kemplay wrote:
   On Wednesday, August 12, 2015 at 12:54:10 PM UTC+1, Jay McCarthy wrote:
On Tue, Aug 11, 2015 at 10:24 AM, Sean Kemplay sean.kemp...@gmail.com 
wrote:
 Hi All,

 Sending an http request through our corporate proxy works as follows 
 for http requests -

 (define-values (x y z)
   (http-sendrecv 10.0.0.200 http://www.example.com;
  #:port 8080
  #:headers '(
  Proxy-Authorization: Basic 
 base64encodedcredentials
  Proxy-Connections: 
 keep-alive
  )
  #:ssl? #f
  #:method GET))

 However fails for HTTPS requests (as expected).

 What I need to do is make a request like the above using the #:method 
 CONNECT and then make a secondary request through a returned 
 connection.

 Does anyone know how I would go about doing that in Racket?

http-sendrecv combines calls to http-conn-open, http-conn-send!,
http-conn-recv!, then http-conn-close!. I suspect that you just need
to break up that one big call into a few calls like open, send, recv,
send, recv, close. I'd be happy to work on it with you, but I don't
have such a proxy on hand, so I'll need helping testing it.

Jay


 Kind regards,
 Sean

 --
 You received this message because you are subscribed to the Google 
 Groups Racket Users group.
 To unsubscribe from this group and stop receiving emails from it, 
 send an email to racket-users+unsubscr...@googlegroups.com.
 For more options, visit https://groups.google.com/d/optout.



-- 
Jay McCarthy
http://jeapostrophe.github.io

   Wherefore, be not weary in well-doing,
  for ye are laying the foundation of a great work.
And out of small things proceedeth that which is great.
  - DC 64:33
   
   Hi Jay,
   
   Thanks for that, yes I think you are right. I have just installed squid 
   at home which also supports http tunnelling. I'll see how I get on and 
   post my results - whether they be good or bad!
   
   It would be really good to at least get an example on the wiki for others 
   to build from, as I suspect a lot of corporate networks are behind 
   proxies and this would be problematic in using Racket to make calls to 
   REST APIs etc which my job at least requires a lot of.
   
   Kind regards,
   Sean
  
  Hi Jay,
  
  Unfortunately I am not getting very far with this.
  
  Our app servers where our production code sits are not behind a proxy, so 
  at the end of the day it doesn't rule out using Racket for some of the 
  tasks I have in mind. It would be nice to be able to get through the proxy 
  from my desktop to test code though.
  
  I tried the following but the connection is ending early -
  
  #lang racket
  (require net/http-client)
  
  (define conn (http-conn-open 10.0.0.200 #:port 8080))
  
  (http-conn-send! conn https://news.ycombinator.com/; #:method CONNECT 
  #:headers '(Proxy-Authorization: Basic base64encodedcredentials 
  Connection: Keep-Alive) #:version #1.1)
  (define-values (a b c)(http-conn-recv! conn #:close? #f))
  (http-conn-send! conn / #:method GET); #:headers 
  '(Proxy-Authorization: Basic Y3hnXHNlYW4ua2VtcGxheTpBdWd1c3QyMDE0 
  Connection: Keep-Alive) #:version #1.1)
  (define-values (e f g) (http-conn-recv! conn))
  (http-conn-close! conn)
  
  I am basing the above on this S/O post but am not entirely sure if this is 
  even the path I should be pursuing!
  
  http://stackoverflow.com/questions/11697943/when-should-one-use-connect-and-get-http-methods-at-http-proxy-server
  
  Here is an example using Node, it opens the connection to the uri through 
  the proxy, then writes to that connection through an SSL connection which 
  is something I can't see how to do in Racket.
  
  http://blog.vanamco.com/proxy-requests-in-node-js/
  
  I will keep digging, if you have any Ideas or anything I could test against 
  our network it would be much appreciated. 
  
  Kind regards,
  Sean
 
 Just an update on this, looking at the code for http-client I now understand 
 that http-conn is a struct with an input and output port from a tcp 
 connection.
 
 Based on the node.js example I am thinking of instead of calling 
 http-conn-send! a second time with a different method I need to write a 
 function along the lines of http-conn-tunnel! which somehow pipes ssl input 
 and output ports from the tcp input and output ports from http-conn's input 
 and output pipes.

I haven't given up... yet!

I exported http-conn-from and http-conn

Re: [racket-users] HTTPS connection through proxy (CONNECT HTTP Method)

[Sean Kemplay]

On Wednesday, August 12, 2015 at 12:54:10 PM UTC+1, Jay McCarthy wrote:
 On Tue, Aug 11, 2015 at 10:24 AM, Sean Kemplay sean.kemp...@gmail.com wrote:
  Hi All,
 
  Sending an http request through our corporate proxy works as follows for 
  http requests -
 
  (define-values (x y z)
(http-sendrecv 10.0.0.200 http://www.example.com;
   #:port 8080
   #:headers '(
   Proxy-Authorization: Basic 
  base64encodedcredentials
   Proxy-Connections: keep-alive
   )
   #:ssl? #f
   #:method GET))
 
  However fails for HTTPS requests (as expected).
 
  What I need to do is make a request like the above using the #:method 
  CONNECT and then make a secondary request through a returned connection.
 
  Does anyone know how I would go about doing that in Racket?
 
 http-sendrecv combines calls to http-conn-open, http-conn-send!,
 http-conn-recv!, then http-conn-close!. I suspect that you just need
 to break up that one big call into a few calls like open, send, recv,
 send, recv, close. I'd be happy to work on it with you, but I don't
 have such a proxy on hand, so I'll need helping testing it.
 
 Jay
 
 
  Kind regards,
  Sean
 
  --
  You received this message because you are subscribed to the Google Groups 
  Racket Users group.
  To unsubscribe from this group and stop receiving emails from it, send an 
  email to racket-users+unsubscr...@googlegroups.com.
  For more options, visit https://groups.google.com/d/optout.
 
 
 
 -- 
 Jay McCarthy
 http://jeapostrophe.github.io
 
Wherefore, be not weary in well-doing,
   for ye are laying the foundation of a great work.
 And out of small things proceedeth that which is great.
   - DC 64:33

Hi Jay,

Thanks for that, yes I think you are right. I have just installed squid at home 
which also supports http tunnelling. I'll see how I get on and post my results 
- whether they be good or bad!

It would be really good to at least get an example on the wiki for others to 
build from, as I suspect a lot of corporate networks are behind proxies and 
this would be problematic in using Racket to make calls to REST APIs etc which 
my job at least requires a lot of.

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
Racket Users group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[racket-users] HTTPS connection through proxy (CONNECT HTTP Method)

[Sean Kemplay]

Hi All,

Sending an http request through our corporate proxy works as follows for http 
requests -

(define-values (x y z)
  (http-sendrecv 10.0.0.200 http://www.example.com;
 #:port 8080
 #:headers '(
 Proxy-Authorization: Basic 
base64encodedcredentials
 Proxy-Connections: keep-alive
 )
 #:ssl? #f
 #:method GET))

However fails for HTTPS requests (as expected).

What I need to do is make a request like the above using the #:method CONNECT 
and then make a secondary request through a returned connection.

Does anyone know how I would go about doing that in Racket?

Kind regards,
Sean

-- 
You received this message because you are subscribed to the Google Groups 
Racket Users group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.