Merge pull request #38 from Kevin-M-Smith/patch-1

Catch <values> without <value> error.

R/getWaterML1Data.r

@@ -60,20 +60,20 @@ getWaterML1Data <- function(obs_url){
     chunk <- xmlDoc(timeSeries[[i]])
     chunk <- xmlRoot(chunk)
     chunkNS <- xmlNamespaceDefinitions(chunk, simplify = TRUE)  
-      
-#     site <- as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:siteProperty[@name='hucCd']", namespaces = chunkNS, xmlValue))
+    
+    #     site <- as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:siteProperty[@name='hucCd']", namespaces = chunkNS, xmlValue))
     site <- as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:siteCode", namespaces = chunkNS, xmlValue))
     agency <- as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:siteCode/@agencyCode", namespaces = chunkNS))
     pCode <-as.character(xpathApply(chunk, "ns1:variable/ns1:variableCode", namespaces = chunkNS, xmlValue))
     statCd <- as.character(xpathApply(chunk, "ns1:variable/ns1:options/ns1:option/@optionCode", namespaces = chunkNS))
-
+    
     valuesIndex <- as.numeric(which("values" == names(chunk)))
     
     zoneAbbrievs <- c(as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:timeZoneInfo/ns1:defaultTimeZone/@zoneAbbreviation", namespaces = chunkNS)),
                       as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:timeZoneInfo/ns1:daylightSavingsTimeZone/@zoneAbbreviation", namespaces = chunkNS)))
-
+    
     names(zoneAbbrievs) <- c(as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:timeZoneInfo/ns1:defaultTimeZone/@zoneOffset", namespaces = chunkNS)),
-                          as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:timeZoneInfo/ns1:daylightSavingsTimeZone/@zoneOffset", namespaces = chunkNS)))
+                             as.character(xpathApply(chunk, "ns1:sourceInfo/ns1:timeZoneInfo/ns1:daylightSavingsTimeZone/@zoneOffset", namespaces = chunkNS)))
     
     for (j in valuesIndex){
       subChunk <- xmlRoot(xmlDoc(chunk[[j]]))
@@ -82,67 +82,69 @@ getWaterML1Data <- function(obs_url){
       
       methodID <- padVariable(methodID,2)
       
-      value <- as.numeric(xpathSApply(subChunk, "ns1:value",namespaces = chunkNS, xmlValue))  
-      datetime <- as.POSIXct(strptime(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS),"%Y-%m-%dT%H:%M:%S"))
-      tzHours <- substr(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS),
-                        24,
-                        nchar(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS)))
-      if(mean(nchar(tzHours),rm.na=TRUE) == 6){
-        tzAbbriev <- zoneAbbrievs[tzHours]
-      } else {
-        tzAbbriev <- rep(as.character(zoneAbbrievs[1]),length(datetime))
-      }
-      
-      timeZoneLibrary <- setNames(c("America/New_York","America/New_York","America/Chicago","America/Chicago",
-                                    "America/Denver","America/Denver","America/Los_Angeles","America/Los_Angeles",
-                                    "America/Anchorage","America/Anchorage","America/Honolulu","America/Honolulu"),
-                                  c("EST","EDT","CST","CDT","MST","MDT","PST","PDT","AKST","AKDT","HAST","HST"))
-      timeZone <- as.character(timeZoneLibrary[tzAbbriev])
-      if(length(unique(timeZone)) == 1){
-       datetime <- as.POSIXct(as.character(datetime), tz = unique(timeZone))
-      } else {
-        warning("Mixed time zone information")
-        for(i in seq_along(datetime)){
-          datetime[i] <- as.POSIXct(as.character(datetime[i]), tz = timeZone[i])
+      value <- as.numeric(xpathSApply(subChunk, "ns1:value",namespaces = chunkNS, xmlValue))
+      if(length(value)!=0){
+        datetime <- as.POSIXct(strptime(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS),"%Y-%m-%dT%H:%M:%S"))
+        tzHours <- substr(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS),
+                          24,
+                          nchar(xpathSApply(subChunk, "ns1:value/@dateTime",namespaces = chunkNS)))
+        if(mean(nchar(tzHours),rm.na=TRUE) == 6){
+          tzAbbriev <- zoneAbbrievs[tzHours]
+        } else {
+          tzAbbriev <- rep(as.character(zoneAbbrievs[1]),length(datetime))
         }
-      }
-      
-      qualifier <- as.character(xpathSApply(subChunk, "ns1:value/@qualifiers",namespaces = chunkNS))
-
-      valueName <- paste(methodID,pCode,statCd,sep="_")
-      qualName <- paste(methodID,pCode,statCd,"cd",sep="_")
-      valueName <- paste("X",valueName,sep="")
-      qualName <- paste("X",qualName,sep="")
-      
-      assign(valueName,value)
-      assign(qualName,qualifier)
-      
-      if(length(get(qualName))!=0){
-        df <- data.frame(rep(agency,length(datetime)),
-                         rep(site,length(datetime)),
-                         datetime,
-                         tzAbbriev,
-                         get(valueName),
-                         get(qualName),
-                         stringsAsFactors=FALSE)
         
-        names(df) <- c("agency_cd","site_no","datetime","tz_cd",valueName,qualName)
-      } else {
-        df <- data.frame(rep(agency,length(datetime)),
-                         rep(site,length(datetime)),
-                         datetime,
-                         tzAbbriev,
-                         get(valueName),stringsAsFactors=FALSE)
+        timeZoneLibrary <- setNames(c("America/New_York","America/New_York","America/Chicago","America/Chicago",
+                                      "America/Denver","America/Denver","America/Los_Angeles","America/Los_Angeles",
+                                      "America/Anchorage","America/Anchorage","America/Honolulu","America/Honolulu"),
+                                    c("EST","EDT","CST","CDT","MST","MDT","PST","PDT","AKST","AKDT","HAST","HST"))
+        timeZone <- as.character(timeZoneLibrary[tzAbbriev])
+        if(length(unique(timeZone)) == 1){
+          datetime <- as.POSIXct(as.character(datetime), tz = unique(timeZone))
+        } else {
+          warning("Mixed time zone information")
+          for(i in seq_along(datetime)){
+            datetime[i] <- as.POSIXct(as.character(datetime[i]), tz = timeZone[i])
+          }
+        }
         
-        names(df) <- c("agency_cd","site_no","datetime","tz_cd",valueName)       
-      }
- 
-      if (1 == i & valuesIndex[1] == j){
-        mergedDF <- df
-      } else {
-        similarNames <- intersect(names(mergedDF), names(df))
-        mergedDF <- merge(mergedDF, df,by=similarNames,all=TRUE)
-#         mergedDF <- merge(mergedDF, df,by=c("agency_cd","site_no","datetime","tz_cd"),all=TRUE)
+        qualifier <- as.character(xpathSApply(subChunk, "ns1:value/@qualifiers",namespaces = chunkNS))
+        
+        valueName <- paste(methodID,pCode,statCd,sep="_")
+        qualName <- paste(methodID,pCode,statCd,"cd",sep="_")
+        valueName <- paste("X",valueName,sep="")
+        qualName <- paste("X",qualName,sep="")
+        
+        assign(valueName,value)
+        assign(qualName,qualifier)
+        
+        if(length(get(qualName))!=0){
+          df <- data.frame(rep(agency,length(datetime)),
+                           rep(site,length(datetime)),
+                           datetime,
+                           tzAbbriev,
+                           get(valueName),
+                           get(qualName),
+                           stringsAsFactors=FALSE)
+          
+          names(df) <- c("agency_cd","site_no","datetime","tz_cd",valueName,qualName)
+        } else {
+          df <- data.frame(rep(agency,length(datetime)),
+                           rep(site,length(datetime)),
+                           datetime,
+                           tzAbbriev,
+                           get(valueName),stringsAsFactors=FALSE)
+          
+          names(df) <- c("agency_cd","site_no","datetime","tz_cd",valueName)       
+        }
+        
+        if (1 == i & valuesIndex[1] == j){
+          mergedDF <- df
+        } else {
+          similarNames <- intersect(names(mergedDF), names(df))
+          mergedDF <- merge(mergedDF, df,by=similarNames,all=TRUE)
+          #         mergedDF <- merge(mergedDF, df,by=c("agency_cd","site_no","datetime","tz_cd"),all=TRUE)
+        }
       }
     }
   }