从 postgresql 上的表中提取 R 中的 XML 数据

Question

I have a table on postgresql, which has an xml column and varchar/numeric columns. When trying to retrieve the data to save it into a data frame the xml is converted to character. Let's recreate the dataset:

my_dataset <- data.frame(id = c(1,1,1,1,2,2,2,2,2),
                         http_action = c("REQUEST","RESPONSE","REQUEST","RESPONSE","REQUEST","RESPONSE","REQUEST","RESPONSE","RESPONSE"),
                         http_data = c('"<?xml version="1.0" standalone="yes"?> <questions> <candidate> <lastname>GOMEZ</lastname> <name>BARNEY</name> </candidate> </questions>)"',
                                       '"<validating> <opnum>123</opnum> <q1>Daily activity?</q1> <a1>Drinking at Moes</a1></validating>"',
                                       '"<?xml version="1.0" standalone="yes"?> <questions> <option>1</option> </questions>"', 
                                       '"<validating> <code>XY936701</code> <date>12/03/2020</date> <time>19:07</time> <result>NONAUTHORIZED</result> <explanation>NON SUITABLE</explanation> </validating>"',
                                       '"<?xml version="1.0" standalone="yes"?> <questions> <candidate> <lastname>LEONARD</lastname> <name>LEN</name> </candidate> </questions>)"' ,
                                       '"<validating> <opnum>124</opnum> <q1>Daily activity?</q1> <a1>Work at Nuclear Power</a1></validating>"',
                                       '"<?xml version="1.0" standalone="yes"?> <questions> <option>1</option> </questions>"', 
                                       '"<validating> <code>XY936702</code> <date>15/03/2020</date> <time>16:12</time> <result>NONAUTHORIZED</result> <explanation>NON SUITABLE</explanation> </validating>"',
                                       '"<validating> <code>XY936702</code> <date>15/03/2020</date> <time>19:24</time> <result>AUTHORIZED</result> <explanation>SUITABLE</explanation> </validating>"'),
                         http_status = c(200,200,200,200,200,200,200,200,200),
                         stringsAsFactors = FALSE)

I receive the following warning:

In postgresqlExecStatement(conn, statement, ...) :
  RS-DBI driver warning: (unrecognized PostgreSQL field type xml (id:142) in column 4)

I can extract the information using string comparisons on lines containing the node , I tried the following:

my_dataset <- my_dataset %>% 
mutate(authorized = ifelse(str_extract(http_data,"<result>[w+]</result>")=="",NA,
                           ifelse(str_extract(http_data,"<result>[w+]</result>")=="NONAUTHORIZED",0,1)))

As a result I get a full NA column, which is not what I expect. Please, could you help me with this question? I mean, perhaps my regex is not well written. And, do you know if it's possible to extract that information directly from the query? Thank you in advance for the help you can provide.

Regards

Answer 1

You have a problem with your regex: it should be something like <result>(\\\\w+)</result> . Also to get the group matches str_extract is not enough. You can use str_match for groups. Take a look at str_match here .

As an alternative solution, you can use an XML parser.