joshkunz

[1] 03:14:47 Joshkunz:~
$ ping google.com
PING google.com (74.125.225.200): 56 data bytes
64 bytes from 74.125.225.200: icmp_seq=0 ttl=57 time=37.938 ms
64 bytes from 74.125.225.200: icmp_seq=1 ttl=57 time=601.127 ms
64 bytes from 74.125.225.200: icmp_seq=2 ttl=57 time=338.328 ms
64 bytes from 74.125.225.200: icmp_seq=3 ttl=57 time=947.589 ms
64 bytes from 74.125.225.200: icmp_seq=4 ttl=57 time=323.217 ms
Request timeout for icmp_seq 5
Request timeout for icmp_seq 6

joshkunz

# Basic ruby conditional
if true
    puts "true 1"
end

# But you can also use the python syntax
if true:
    puts "true 2"
end

joshkunz

-- Language: Haskell
-- To Run: $ ghc fizzbuzz_doc.hs && ./fizzbuzz

-- The syntax a -> a -> b is just 'function composition'
-- a -> b is a function that take a parameter of type a
-- and returns something of type b.
-- a -> a -> b is a function of type (a -> (a -> b))
-- which means that it takes an 'a' and returns
-- a function that takes an 'a' and returns a 'b'
-- All functions in haskell are specified this way.

joshkunz

/* Build, (with your favorite compiler) run: 
 * cc -o vlcwrap vlcwrap.c */

#include <unistd.h>
#include <signal.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>

#define VLC_BIN "/Applications/VLC.app/Contents/MacOS/VLC"

joshkunz

Simple classes for building compatible DFAs in a more intuitive way. (source is below the documentation)

Build a DFA manually

Using it manually is pretty simple, basically you create a new DFA object, and add states to it. States are created automatically when you try and use them, so you don't have to worry about creating them explicitly. Here's an example of creating the DFA shown here.

joshkunz

#include <limits.h>
#include <stdio.h>

/* magic constant */
#define SIGN_MASK INT_MIN 
/* negative shift for arbitrary size int */
#define NS ((sizeof(int)<<3) - 1)
/* Positive shift for arbitrary size int */
#define PS (NS - 1)

joshkunz

diff --git a/src/client_file.c b/src/client_file.c
index 2ee4333..8174c3b 100644
--- a/src/client_file.c
+++ b/src/client_file.c
@@ -44,12 +44,18 @@ client_allow_file(const struct client *client, const char *path_fs,
 		   instance */
 		return true;
 
+    /* Pretty sure this check will never work. 
 	if (uid <= 0) {

joshkunz

#include <stdio.h>
#include <stdlib.h>

FILE* next_file(char delim) {
    if (feof(stdin) || ferror(stdin)) { return NULL; }
    char * file_name = NULL;
    size_t buffer_size = 0;
    FILE * out = NULL;
    getdelim(&file_name, &buffer_size, delim, stdin);
    if (file_name) {

joshkunz

Testing: ./armage17/triangle
~~~ Failed test. ~~~
[(1, 2147483647), (0, 2147483647), (0, 0)]
[Expected] 'scalene right\n'
[Got]      'isosceles acute\n'
~~~ Failed test. ~~~
[(0, 0), (1073741822, 1073741824), (2147483645, 2147483647)]
[Expected] 'scalene obtuse\n'
[Got]      'isosceles obtuse\n'
~~~ Failed test. ~~~

joshkunz

Here's the script I wrote that automatically subscribes you to all forum threads. It's built to be run as a cron job.

You'll need to generate an API token for it which can be done at the bottom of the [profile settings page][profile]. It uses canvas's REST API to do the actual subscribing, the documentation for the api [is here][capi].

I also hate Canvas's WYSIWYG editor, so I wrote some scripts to render markdown, and then inline the styles. Typically I'll write my posts in a text file, and then run:

$ fpost name_of_file.md | pbcopy    # pbcopy is an OSX thing, xclip -in works for xwindows.

and then just post the raw HTML into the WYSIWYG's 'html entry' portion.