codegordi

# best practices for web scraping in R #

# function should be used with ldply
# eg: 

ldply(urls, scrape)

# add a try to ignore broken links/ unresponsive pages
# eg: 

codegordi

### a simple coordinate conversion function to convert latitude, longitude in DMS to decimal-degress
### christina gutierrez aka github@codegordi
#library(gsubfn)
require(gsubfn)
options(digits=15)
# point.set == data set with at least latitute and longitude in degrees-(arc)minutes-seconds[-fractional seconds]; 
#   no projection or datum assumed;
#   format of DD[D]:MM:SS[*] is assumed
# 	e.g. coords <- c("122:45:45", "-69:38:27")
# 	e.g  coords <- c("122:45:45.84", "-69:38:27.76")

codegordi

 
# use plyr package
library(plyr)
 
### READ all files in working directory into a list of dataframes
d.list=lapply(list.files(getwd(), full = F), FUN =  read.table, header=T, sep="\t", fill=T) # name data
 
### SEPARATE out data frames from list -- using a for-loop in this example
i=0
f.coll=list.files()

codegordi

# JDBC driver library for R
library("RJDBC")

user <- "cgutierrez"  # enter your own username
#pwd <-               # enter in URL str below

# I chose open-source JTDS for a particular reason in this example (not related to R); you might choose another one
# Check that appropriate .jar is in indicated directory
drv <- JDBC("net.sourceforge.jtds.jdbc.Driver", "/usr/share/java/drivers/jdbc/jtds-1.2.7.jar", identifier.quote="`")  
cxn <- dbConnect(drv, "jdbc:jtds:sqlserver://MYSERVER01ABC;useNTLMv2=true;domain=MY.DOMAIN.COM", user, #####) # enter pwd at runtime

codegordi

### manage memory on a large data set using multicore library
library(multicore)

# read in tab-delimited file from working dir
df = read.table(getwd(), sep="\t", header=T)
d.lines = as.character(df$charvar)            # $charvar is character-class variable you want to grep

grep_wrap <- function (pattern, x, ignore.case = FALSE, perl = FALSE, value = FALSE, fixed = FALSE, useBytes = FALSE, invert = FALSE) {
    ret = rep(0, length(x))
    ret[grep(pattern, x, ignore.case, perl, value, fixed, useBytes, invert)] = 1

codegordi

library('bitops')
library('RCurl')
library('RJSONIO')

query = function(querystring) {
  h = basicTextGatherer()
  curlPerform(url="http://<your host IP>/db/data/ext/CypherPlugin/graphdb/execute_query",
              postfields=paste('query',curlEscape(querystring), sep='='),
              writefunction = h$update,
              verbose = FALSE

codegordi

def cumRelFreqDistn(tups):
    
    # create bins of increment 0.01
    a = [i*-0.01 for i in range(100)]
    a = a[1:len(a)]
    b = [i*0.01 for i in range(101)]
    a.extend(b)
    a.sort()
    bins = a
    

codegordi

1) Get the FileGDB API http://www.esri.com/apps/products/download/

2) Extract it somewhere on your system and remember the path :)

3) Do:
> mkdir build   #directory where we are playing around
> cd build
> git clone https://github.com/OSGeo/gdal.git
> cd gdal
> ./configure --with-fgdb=/path/to/your/FileGDB_API

codegordi

# alteRyx_install_packages.R
#  > code to install packages via R Developer tool in Alteryx(R) module
#  > note call to custom Alteryx-R function wrte.Alteryx()

altx.repo <- getOption("repos")
altx.repo["CRAN"] <- "http://cran.rstudio.com"  # set your primary repo if you haven't already
options(repos = altx.repo)
#write.Alteryx(getOption("repos"), 1)  # DEBUG
install.packages("XML")

codegordi

## install Haskell Platform on Mac OS

(1)  Download from https://www.haskell.org/platform/
       > Contains the GHCi REPL shell (console-interpreter), common vetted libraries, Cabal, etc.
       > See https://www.haskell.org/platform/doc/2014.2.0.0/start.html for more information

(2)  Um ... that should be it on Mavericks (v. 10.9.+). You may run into problems if you had GHC/i installed already on your Mac and then upgraded to Mavericks. Look up @cartazio or @katchuang gists in these cases.

(3) Do you a Haskell for a Great Good!