I am trying to figure out how to parse a webpage - r

Question

I am working on a summer project. To grab course information from my school website.
I start off by going here: http://www.uah.edu/cgi-bin/schedule.pl?file=fall2015.html&segment=
to gather the course departments.
Then I grab info from pages like this one.
I have what I need filtered down to a list like:
[1] "91091 211 01 PRINC OF FINANCIAL ACCOUNTING 3.0 55 22 33 0 MW 12:45PM 02:05PM BAB 106 Rose-Green E"
[2] "91092 211 02 PRINC OF FINANCIAL ACCOUNTING 3.0 53 18 35 0 TR 09:35AM 10:55AM BAB 123 STAFF"
[3] "91093 211 03 PRINC OF FINANCIAL ACCOUNTING 3.0 48 29 19 0 TR 05:30PM 06:50PM BAB 220 Hoskins J"
[4] "91094 212 01 MANAGEMENT ACCOUNTING 3.0 55 33 22 0 MWF 11:30AM 12:25PM BAB 106 Hoskins J"
[5] "91095 212 02 MANAGEMENT ACCOUNTING 3.0 55 27 28 0 TR 02:20PM 03:40PM BAB 106 Bryson R"
However my issues are as follows:
www.uah.edu/cgi-bin/schedule.pl?file=fall2015.html&segment=CS
I need to add the department from each url. In the link I gave, the department was "CS". I need to have that included with each entry.
I need to turn this into a table, or some other object where I can reference the data like
Max Wait
CRN Course Title Credit Enrl Enrl Avail List Days Start End Bldg Room Instructor
------ ---------- ------------------------------ ------ ---- ---- -------- ---- ------- ------- ------- ----- ---------- --------------------
Basically how the data is displayed on the page.
So my end goal is to go through each of those links I grab, get all the course info(except the section type). Then put it into a giant data.frame that has all the courses like this.
Department CRN Course Title Credit MaxEnrl Enrl Avail WaitList Days Start End Bldg Room Instructor
ACC 91095 212 02 MANAGEMENT ACCOUNTING 3.0 55 27 28 0 TR 02:20PM 03:40PM BAB 106 Bryson R
So far I have this working
require(data.table)
require(gdata)
library(foreach)
uah <- readLines('http://www.uah.edu/cgi-bin/schedule.pl?file=fall2015.html&segment=')
uah <- substring(uah[grep('fall2015', uah)], 10)
uah <- sub("\\\"(.*)", "", uah)
uah <- paste("http://www.uah.edu" , uah , sep = "")
gatherClasses <- function(url){
dep <- readLines(url)
dep <- dep[grep('[[:digit:][:digit:][:digit:][:digit:]][[:digit:][:digit:][:digit:]] [[:digit:][:digit:]]', dep)]
dep <- substring(dep, 6)
dep <- foreach(i=dep) %do% i[grep('[[:digit:][:digit:][:digit:][:digit:]][[:digit:][:digit:][:digit:]] [[:digit:][:digit:]]', i)]
dep <- foreach(i=dep) %do% trim(i)
dep <- dep[2:length(dep)]
return(dep)
}
x <- gatherClasses(uah[1])
x <-unlist(x)
I am having trouble split the data in the right places. I am not sure what I should try next.
EDIT:(Working Now)
require(data.table)
require(gdata)
library(foreach)
uah <- readLines('http://www.uah.edu/cgi-bin/schedule.pl?file=sum2015b.html&segment=')
uah <- substring(uah[grep('sum2015b', uah)], 10)
uah <- sub("\\\"(.*)", "", uah)
uah <- paste("http://www.uah.edu" , uah , sep = "")
gatherClasses <- function(url){
L <- readLines(url)
Fields <- sub(" *$", " ", grep("---", L, value = TRUE))
widths <- diff(c(0, gregexpr(" ", Fields)[[1]]))
Data <- grep("\\d{5} \\d{3}", L, value = TRUE)
classes <- read.fwf(textConnection(Data), widths, as.is = TRUE, strip.white = TRUE)
classes$department <- unlist(strsplit(url, '='))[3]
return(classes)
}
allClasses = foreach(i=uah) %do% gatherClasses(i)
allClasses <- do.call("rbind", allClasses)
write.table(mydata, "c:/sum2015b.txt", sep="\t")

Answer 1

Read the lines into L, grab the "--- ---- etc." line into Fields and ensure that there is exactly one space at the end. Find the character positions of the spaces and difference them to get the field widths. Finally grep out the data portion and read it in using read.fwf which reads fixed width fields. For example, for Art History:
URL <- "http://www.uah.edu/cgi-bin/schedule.pl?file=fall2015.html&segment=ARH"
L <- readLines(URL)
Fields <- sub(" *$", " ", grep("---", L, value = TRUE))
widths <- diff(c(0, gregexpr(" ", Fields)[[1]]))
Data <- grep("\\d{5} \\d{3} \\d{2}", L, value = TRUE)
read.fwf(textConnection(Data), widths, as.is = TRUE, strip.white = TRUE)
giving:
V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13 V14 V15
1 90628 100 01 ARH SURV:ANCIENT-MEDIEVAL 3 35 27 8 0 TR 12:45PM 02:05PM WIL 168 Joyce L
2 90630 101 01 ARH SURV:RENAISSANCE-MODERN 3 35 14 21 0 MW 12:45PM 02:05PM WIL 168 Stewart D
3 90631 101 02 ARH SURV:RENAISSANCE-MODERN 3 35 8 27 0 MW 03:55PM 05:15PM WIL 168 Stewart D
4 92269 101 03 ARH SURV:RENAISSANCE-MODERN 3 35 5 30 0 TR 11:10AM 12:30PM WIL 168 Shapiro Guanlao M
5 90632 101 04 ARH SURV:RENAISSANCE-MODERN 3 35 13 22 0 TR 02:20PM 03:40PM WIL 168 Shapiro Guanlao M
6 90633 301 01 ANCIENT GREEK ART 3 18 3 15 0 MW 02:20PM 03:40PM WIL 168 Joyce L
7 92266 306 01 COLLAPSE OF CIVILIZATIONS 3 10 4 6 0 TR 12:45PM 02:05PM SST 205 Sever T
8 W 90634 309 01 CONTEMPORARY ART & ISSUES 3 18 10 8 0 TR 09:35AM 10:55AM WIL 168 Stewart D
9 90635 320 01 ST: MODERN ARCHITECTURE 3 12 0 12 0 TR 11:10AM 12:30PM WIL 172 Takacs T
10 90636 400 01 SENIOR THESIS 3 0 0 0 0 TBA TBA TBA TBA Joyce L
11 90637 400 02 SENIOR THESIS 3 0 0 0 0 TBA TBA TBA TBA Stewart D

Answer 2

I wrote and donated that schedule.pl script about 20 years ago because they simply published the flat mainframe files of all the courses on offer for each session. The script's job is to break up the whole set and present it in human-consumable chunks. (That, and back then a browser would choke on that much data.) I understand from one of the former UAH IT people that they tried to do away with it once, but got a great hew and cry from users, so they figured out how to keep it working.
It would be easier for you to ask the UAH IT folks if you can't just retrieve the underlying flat file. It used to be on a public-facing URL, but like I said, that was about 20 years ago, so I don't recall the specifics. The output you see when viewing courses is the same as the flat file, but the flat file contains every department, so you don't have to fetch each separately.