SQLQuery with Paste in R & RODBC - mysql

Question

I have a problem with RODBC;this is the error :
chargerExp("C:\\test.csv",NE=1,NC=1,s=1)
Exeperience: 1 Execution: 1 Sujet: 1
> ajouter(new ("BDD"),new("Exp"))
[1] "42000 1064 [MySQL][ODBC 5.2(a) Driver][mysqld-5.6.17]You have an error in your SQL syntax;
check the manual that corresponds to your MySQL server version for the right syntax to use near ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/' at line 1"
[2] "[RODBC] ERROR: Could not SQLExecDirect 'INSERT INTO `test` (`NE`, `NC`, `E`, `X`, `Y`,
`Z`, `T`, `A`, `S`) VALUES ( , , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , , );'"
I'm using Mysql, RODBC and Rstudio with a method ajouter to insert in the data base
setMethod( f ="ajouter",signature =c(x="BDD",obj="Exp"),
def = function(x, obj)
{
channel <- odbcConnect(dsn="RSQL",uid="root",pwd="toor")
ne <- obj["ne"]
nc <- obj["nc"]
s <- obj["S"]
e<- encoder((obj["E"]))
x <- encoder((obj["X"]))
y <- encoder((obj["Y"]))
z <- encoder((obj["Z"]))
t<- encoder((obj["T"]))
a <- (obj["A"])
requeteSql.valeur <- paste("'",e,"'",",",
"'",x,"'",",",
"'",y,"'",",",
"'",z,"'",",",
"'",t,"'",",")
requetesql <- paste("INSERT INTO `test` (`NE`, `NC`, `E`, `X`, `Y`, `Z`, `T`,
`A`, `S`) VALUES (",ne,",",nc,", ",requeteSql.valeur, a,", ",s,");")
sqlQuery(channel, requetesql)
}
)
This is encoder method. It's main purpose is to convert my object to a text with some concatenation.
setMethod( f ="encoder", signature ="Para",
def =function(x, i, j, value)
{
s <- as.character(x["val"][1])
for(i in 2:length(x["val"]))
{
s <- paste(s,x["val"][i],sep="/")
}
return(s)
}
)
The call of my methods :
ajouter(new ("BDD"), new("Exp"))
This is my table in data base :
CREATE TABLE `test` (
`idTest` int(11) NOT NULL AUTO_INCREMENT,
`NE` int(11) DEFAULT NULL,
`NU` int(11) DEFAULT NULL,
`E` text,
`X` text,
`Y` text,
`Z` text,
T` text,
`A` float DEFAULT NULL,
`S` int(11) DEFAULT NULL,
PRIMARY KEY (`idTest`)
) ENGINE=InnoDB AUTO_INCREMENT=2 DEFAULT CHARSET=utf8;
My problem is from sqlquery, it said that my syntax are not correct, I don't know why.
This is the the object I want to load into the database (from dput so copy/pasteable):
new("Para"
, ne= 1
, nc = 1
, E = new("E"
, val = c(-13, -11, -11, -11, -11, -9, -10, -12, -12, -12, -11, -10,
-8, -8, -8, -9, -9, -7, -9, -10, -9, -9, -9, -9, -11, -9, -7,
-7, -7, -7, -7, -8, -7, -7, -5, -5, -7, -8, -7, -5, -5, -6, -6,
-9, -35, -76, -96, -62, 38, 167, 251, 251, 248, 157, 94, 56,
)
)
, X = new("Para"
, val = c(115, 116, 114, 113, 113, 114, 115, 114, 114, 113, 112, 111,
113, 114, 114, 115, 115, 116, 115, 115, 114, 116, 114, 115, 114,
113, 114, 114, 114, 114, 114, 113, 113, 114, 114, 114, 114, 115,
114, 115, 115, 115, 114, 115, 116, 114, 114, 114, 116, 115, 113,
)
)
, Y = new("Para"
, val = c(10, 11, 9, 9, 10, 9, 10, 9, 11, 10, 11, 10, 11, 11, 10, 11,
10, 10, 11, 10, 9, 10, 11, 12, 11, 10, 11, 11, 11, 12, 11, 11,
11, 10, 11, 11, 10, 12, 10, 11, 11, 11, 11, 11, 11, 10, 11, 11,
10, 11, 11, 11, 10, 11, 11, 11, 10, 11, 10, 12, 11, 10, 10, 10,
)
)
, Z = new("Para"
, val = c(-42, -42, -44, -43, -42, -41, -42, -42, -42, -42, -42, -43,
-40, -41, -41, -40, -41, -43, -41, -41, -41, -41, -41, -40, -40,
-41, -40, -40, -41, -40, -42, -41, -41, -41, -41, -41, -43, -42,
-43, -42, -41, -42, -42, -40, -41, -42, -40, -41, -41, -41, -42,
)
, T = new("Para"
, val = c(25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2,
25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2,
25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2,
25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2, 25.2,
)
)
, A = 1L
, S = 1
)

Answer 1

I downloaded your code and there seem to be many problems, the lest of which is actually running the query you tried. The error tells you it tried to run the SQL command
INSERT INTO `test` (`NE`, `NC`, `E`, `X`, `Y`, `Z`, `T`, `A`, `S`)
VALUES ( , , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' , ' NA/NA/NA/ ' ,
' NA/NA/NA/ ' , 'NA/NA/NA/ ' , , );'
And it should be clear that looks invalid. I don't think you can have emply values in an INSERT statement like that.
The blank values are from the lines
ne <- obj["ne"]
nc <- obj["nc"]
s <- obj["S"]
In your function definition, obj is your Exp (or Experimentation) and you simply haven't set any values for these slots yet. Since they are empty vectors, they do not render as anything in the paste statement.
The batman values (aka NA/NA/NA) are from the
e<- encoder((obj["E"]))
x <- encoder((obj["X"]))
y <- encoder((obj["Y"]))
lines. When the Signal class is empty (as it is when you create a new, empty object) the val slot is just an empty numeric vector. So when you go to encode it, you've set up a loop that's basically going
s <- NA
for(i in 2:1)
{
s <- paste(s,NA,sep="/")
}
which is resulting in that NA/NA/NA/ value. And really that loop in unnecessary, a simple
s <- paste(as.character(x["val"]), collapse="/")
would give you the string you want without the counting-backwards problem.
So really, running ajouter(new ("BDD"),new("Exp")) with empty objects like that doesn't make much sense. You really should test each of your classes first by themselves before putting them together like that. I can't believe you didn't get other errors or warnings before that. You should work on isolating your errors and testing each part individually.