I'm reading big csv files and mashing the lines into hierarchical objects for analysis and display. The csv might look like this:
YEAR,QUAD,SPEC,FIASP,COND,SZ2,SZ3,SZ4,SZ5,SZ6,SZ7,SZ8,SZ9
1984,4D1,1,129,1,1,0,1,1,0,1,3,0
1984,4D1,8,371,0,0,0,0,0,1,0,0,1
1984,4D1,9,375,1,0,0,0,0,0,0,1,0
1984,4D1,1,129,0,2,0,1,1,1,1,1,0
1984,4D1,3,012,0,0,1,0,3,1,0,0,0
...
2009,8I3,5,833,1,0,0,0,0,0,0,1,0
2009,8I3,7,316,1,0,0,0,0,0,0,0,1
2009,8I3,25,585,0,1,1,2,1,0,0,0,0
2009,8I3,7,316,0,0,0,0,0,1,1,0,1
My non-clever reading code works but I dislike it; it goes like this (could be typos from editing down):
function makePlotData() {
if(ground.reader.readyState==4) {
var lines = ground.reader.responseText.split(/\r\n?/);
for (var i=0; i<lines.length; i++) {
if (lines[i].match(/^\d\d\d\d,/)) { // then it's a regular csv line
var fieldsArr = lines[i].split(',');
var [year, quad, spec] = [fieldsArr[0],fieldsArr[1],fieldsArr[2]];
ground[quad] = ground[quad] || {};
ground[quad][year] = ground[quad][year] || {};
ground[quad][year][spec] = ground[quad][year][spec] || {};
ground[quad][year][spec].sum = ground[quad][year][spec].sum || 0;
ground[quad][year][spec].sizes = ground[quad][year][spec].sizes || [];
var sizeArr = fieldsArr.slice(5).map(Number);
if (ground[quad][year][spec].sizes.length) { // already have a size classes Arr
for(var s=0;s<sizeArr.length;s++){ // ...so add to it
ground[quad][year][spec].sizes[s] += sizeArr[s];}
} else {ground[quad][year][spec].sizes = sizeArr;}
} // ... continue with assignments and doing stuff
My objection is to that series of statements like ground[quad] = ground[quad] || {}; , which do what I need but seem like a lot of unnecessary assigning and reassigning; I bet it doesn't matter but it gives me a feeling of wasted cycles and room for errors; it's also ungainly and a nuisance to modify. Is there any idiom or anything that manages this more efficiently or is this as good as it gets with this kind of structure?
