-
Notifications
You must be signed in to change notification settings - Fork 0
/
01_getPic_data.Rmd
207 lines (175 loc) · 7.09 KB
/
01_getPic_data.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
---
title: "flickr map Switzerland"
author: "Duc-Quang Nguyen"
date: "14 May 2016"
output: html_document
---
## FlickR
* [timelyportfolio R and flickr](http://timelyportfolio.github.io/rCharts_Rflickr/iso_httr.html)
* [flickr color analysis](http://beautifuldata.net/2013/05/color-analysis-of-flickr-images/)
* [](https://github.com/furukama/flickr/blob/master/flickr_get.R)
* [my API page](https://www.flickr.com/services/apps/by/112725067@N03)
* [official doc API](https://www.flickr.com/services/api/)
```{r setup, include=FALSE}
require(httr)
library(RCurl)
library(magrittr)
library(dplyr)
library(jsonlite)
library(parallel)
startDate <- as.Date("2014-01-01")
endDate <- as.Date("2016-07-14") #Sys.Date()
freqTime <- "week"
getPics <- F
getInfoPics <- F
getFavs <- F
# Use procedure in http://timelyportfolio.github.io/rCharts_Rflickr/iso_httr.html works!
# save(api_key, secret, flickr.app, flickr.endpoint, tok, file = "~/swissinfo/_helpers/secrets.Rdata")
load("~/swissinfo/_helpers/secrets.Rdata")
if(getPics) {
# flickr search API function
flickrSearch <- function(
bbox,
content_type = 1,
min_taken_date = format( Sys.Date() - 7, "%Y-%m-%d"),
max_taken_date = format( Sys.Date(), "%Y-%m-%d"),
api_key = api_key,
tok = tok
) {
search <- GET(url=sprintf(
"https://api.flickr.com/services/rest/?method=flickr.photos.search&api_key=%s&bbox=%s&content_type=%s&min_taken_date=%s&max_taken_date=%s&format=json&nojsoncallback=1"
, api_key
, bbox
, content_type
, min_taken_date
, max_taken_date
, tok$credentials$oauth_token
)
) %>%
content( as = "text", encoding="UTF-8") %>%
jsonlite::fromJSON ()
stopifnot(search[[2]] == "ok")
# subset the list to get only the relevant data
search %$% photos %$% photo
}
time.frames <- data.frame(
min_taken_date = seq(startDate, endDate, freqTime)
)
time.frames$max_taken_date = c(time.frames[-1, ] -1, Sys.Date() + 1)
# To define bbox, use http://bboxfinder.com/#45.809658,5.745850,47.813155,10.563354
bbox <- paste(c(5.745850,45.809658,10.563354,47.813155), collapse = ",")
pics <- do.call(rbind, lapply(1:nrow(time.frames), function(i) {
cat("\n", i)
flickrSearch(
bbox = bbox,
min_taken_date = time.frames[i,'min_taken_date'],
max_taken_date = time.frames[i,'max_taken_date'],
api_key = api_key,
tok = tok)
}))
pics <- pics %>% select(-isfriend, -isfamily)
write.csv(pics, file = paste0("data/", startDate, "_", endDate, "_by", freqTime, ".csv"), row.names = F)
} else {
pics <- read.csv(file=paste0("data/", startDate, "_", endDate, "_by", freqTime, ".csv"), stringsAsFactors = F)
}
stopifnot(!any(duplicated(pics)))
if(getInfoPics) {
# flickr API get info
getInfo <- function(id, api_key = api_key, tok = tok) {
call <- GET(url=sprintf(
"https://api.flickr.com/services/rest/?method=flickr.photos.getInfo&api_key=%s&photo_id=%s&format=json&nojsoncallback=1"
, api_key
, id
, tok$credentials$oauth_token
)) %>%
content( as = "text", encoding="UTF-8") %>%
jsonlite::fromJSON ()
if(call[[2]] == "ok") {
result <- call %$% photo
data.frame(
id = as.numeric(result$id),
lat = as.numeric(result$location$latitude),
lon = as.numeric(result$location$longitude),
locality = if(is.null(result$location$locality$`_content`)) "" else result$location$locality$`_content`,
county = if(is.null(result$location$county$`_content`)) "" else result$location$county$`_content`,
region = if(is.null(result$location$region$`_content`)) "" else result$location$region$`_content`,
country = if(is.null(result$location$country$`_content`)) "" else result$location$country$`_content`,
isFavorite = as.numeric(result$isfavorite),
dateTaken = as.character(result$dates$taken),
views = as.numeric(result$views),
url = as.character(result$url$url$`_content`)
)
} else {
warning("\nAPI call for ", id, " failed!")
NULL
}
}
# http://gforge.se/2015/02/how-to-go-parallel-in-r-basics-tips/
# Calculate the number of cores
# Initiate cluster
cl <- makeCluster(detectCores(), outfile ="")
clusterExport(cl=cl, varlist=c("pics", "getInfo", "api_key", "tok", "%>%", "GET", "content", "%$%"))
infos <- do.call(rbind, parLapply(cl, 1:nrow(pics), function(i) {
cat("\n", i, "/", nrow(pics), "\t", pics[i, 'id'])
getInfo(pics[i, 'id'], api_key = api_key, tok= tok)
}))
stopCluster(cl)
pics$id <- as.numeric(pics$id)
pici <- right_join(pics, infos) %>% select(-isFavorite, -ispublic)
write.csv(pici, file = paste0("data/", startDate, "_", endDate, "_by", freqTime, "_info.csv"), row.names = F)
} else {
pici <- read.csv(file = paste0("data/", startDate, "_", endDate, "_by", freqTime, "_info.csv"), stringsAsFactors = F)
}
if(getFavs) {
# get favorites count, a different API call!!!
# https://www.flickr.com/services/api/flickr.photos.getFavorites.html --> flickr.photos.getFavorites
# flickr API get info
getFav <- function(id, api_key = api_key, tok = tok) {
call <- GET(url=sprintf(
"https://api.flickr.com/services/rest/?method=flickr.photos.getFavorites&api_key=%s&photo_id=%s&format=json&nojsoncallback=1"
, api_key
, id
, tok$credentials$oauth_token
)) %>%
content( as = "text", encoding="UTF-8") %>%
jsonlite::fromJSON ()
if(call[[2]] == "ok") {
result <- call %$% photo
data.frame(
id = as.numeric(result$id),
faveCount = as.numeric(result$total)
)
} else {
warning("\nAPI call for ", id, " failed!")
NULL
}
}
# http://gforge.se/2015/02/how-to-go-parallel-in-r-basics-tips/
# Calculate the number of cores
# Initiate cluster
cl <- makeCluster(detectCores(), outfile ="")
clusterExport(cl=cl, varlist=c("pici", "getFav", "api_key", "tok", "%>%", "GET", "content", "%$%"))
favs <- do.call(rbind, parLapply(cl, 1:nrow(pici), function(i) {
cat("\n", i, "/", nrow(pici), "\t", pici[i, 'id'])
getFav(pici[i, 'id'], api_key = api_key, tok = tok)
}))
stopCluster(cl)
picif <- right_join(pici, favs)
write.csv(picif, file = paste0("input/", startDate, "_", endDate, "_by", freqTime, "_faved.csv"), row.names = F)
} else {
picif <- read.csv(file = paste0("input/", startDate, "_", endDate, "_by", freqTime, "_faved.csv"), stringsAsFactors = F)
}
#discard pics not taken in Switzerland
## get only grid points within Switzerland
#http://www.inside-r.org/packages/cran/sp/docs/point.in.polygon
require(sp)
require("maps")
world <- map_data("world")
ch <- world[which(world$region == "Switzerland"),]
ch.map <- ggplot(ch, aes(x = long, y = lat, group = group)) +
geom_path() + coord_equal() + theme_minimal()
picifs <- picif[which(point.in.polygon(picif$lon, picif$lat, ch$long, ch$lat) == 1),]
write.csv(picifs, file = paste0("input/", startDate, "_", endDate, "_by", freqTime, "_faved_CH.csv"), row.names = F)
# visual check
ch.map + geom_point(data = picifs, aes(x = lon, y = lat, group = 1), alpha = 0.3, size = 0.1)
```