home |
copyright ©2019, tjmenzie@ncsu.edu
syllabus |
src |
submit |
chat
Create a class Tbl Col Row (and reuse Num from last time).
Do not use any built-in CSV class (and definitely do not use Pandas).
Tbls has many rows, which is a list ofRow- Whenever a line from some comma-separated data is added to a
Tbl, then:- It becomes a
RowinTbl - And all the
Nums in thecolsare updated (using the code you wrote last week).
- It becomes a
Tbls reads named csv files (*).- The first line is a list of column names
- The other lines are data that we will read and store as
Rows in theTbls
$cloudCover, $temp, $humid, $wind, $playHours
100, 68, 80, 0, 3
0, 85, 85, 0, 0
0, 80, 90, 10, 0
60, 83, 86, 0, 4
100, 70, 96, 0, 3
100, 65, 70, 20, 0
70, 64, 65, 15, 5
0, 72, 95, 0, 0
0, 69, 70, 0, 4
80, 75, 80, 0, 3
0, 75, 70, 18, 4
60, 72, 83, 15, 5
40, 81, 75, 0, 2
100, 71, 91, 15, 0
Details:
- All my objects have a unique id
oid(that's optional for you). Tbls hold a list ofcols(one for each column in the `rows')- Every
Colknows its column name and whatposit has
- Every
Rowskeep a list ofcellsand another list calledcookedthat we won't talk about now.
My code reads the above and generates a
Tbl object with the following structure:
t.cols
| 1
| | add: Num1
| | col: 1
| | hi: 100
| | lo: 0
| | m2: 25635.7
| | mu: 52.1429
| | n: 14
| | oid: 2
| | sd: 44.407
| | txt: $cloudCover
| 2
| | add: Num1
| | col: 2
| | hi: 85
| | lo: 64
| | m2: 561.429
| | mu: 73.5714
| | n: 14
| | oid: 3
| | sd: 6.57167
| | txt: $temp
| 3
| | add: Num1
| | col: 3
| | hi: 96
| | lo: 65
| | m2: 1375.21
| | mu: 81.6429
| | n: 14
| | oid: 4
| | sd: 10.2852
| | txt: $humid
| 4
| | add: Num1
| | col: 4
| | hi: 20
| | lo: 0
| | m2: 820.857
| | mu: 6.28571
| | n: 14
| | oid: 5
| | sd: 7.94625
| | txt: $wind
| 5
| | add: Num1
| | col: 5
| | hi: 5
| | lo: 0
| | m2: 53.5
| | mu: 2.5
| | n: 14
| | oid: 6
| | sd: 2.02864
| | txt: $playHours
t.oid: 1
t.rows
| 1
| | cells
| | | 1: 100
| | | 2: 68
| | | 3: 80
| | | 4: 0
| | | 5: 3
| | cooked
| | dom: 0
| | oid: 7
| 2
| | cells
| | | 1: 0
| | | 2: 85
| | | 3: 85
| | | 4: 0
| | | 5: 0
| | cooked
| | dom: 0
| | oid: 8
| 3
| | cells
| | | 1: 0
| | | 2: 80
| | | 3: 90
| | | 4: 10
| | | 5: 0
| | cooked
| | dom: 0
| | oid: 9
| 4
| | cells
| | | 1: 60
| | | 2: 83
| | | 3: 86
| | | 4: 0
| | | 5: 4
| | cooked
| | dom: 0
| | oid: 10
| 5
| | cells
| | | 1: 100
| | | 2: 70
| | | 3: 96
| | | 4: 0
| | | 5: 3
| | cooked
| | dom: 0
| | oid: 11
| 6
| | cells
| | | 1: 100
| | | 2: 65
| | | 3: 70
| | | 4: 20
| | | 5: 0
| | cooked
| | dom: 0
| | oid: 12
| 7
| | cells
| | | 1: 70
| | | 2: 64
| | | 3: 65
| | | 4: 15
| | | 5: 5
| | cooked
| | dom: 0
| | oid: 13
| 8
| | cells
| | | 1: 0
| | | 2: 72
| | | 3: 95
| | | 4: 0
| | | 5: 0
| | cooked
| | dom: 0
| | oid: 14
| 9
| | cells
| | | 1: 0
| | | 2: 69
| | | 3: 70
| | | 4: 0
| | | 5: 4
| | cooked
| | dom: 0
| | oid: 15
| 10
| | cells
| | | 1: 100
| | | 2: 75
| | | 3: 80
| | | 4: 0
| | | 5: 3
| | cooked
| | dom: 0
| | oid: 16
| 11
| | cells
| | | 1: 0
| | | 2: 75
| | | 3: 70
| | | 4: 18
| | | 5: 4
| | cooked
| | dom: 0
| | oid: 17
| 12
| | cells
| | | 1: 60
| | | 2: 72
| | | 3: 90
| | | 4: 10
| | | 5: 4
| | cooked
| | dom: 0
| | oid: 18
| 13
| | cells
| | | 1: 40
| | | 2: 81
| | | 3: 75
| | | 4: 0
| | | 5: 5
| | cooked
| | dom: 0
| | oid: 19
| 14
| | cells
| | | 1: 100
| | | 2: 71
| | | 3: 91
| | | 4: 15
| | | 5: 0
| | cooked
| | dom: 0
| | oid: 20
Divide this problem into three parts:
- Part1: read a file from disk (or from a string) into a list of lists (where all the strings are turned into numbers, where appropriate).
- Part2: redo the above, handling some tedious special cases:
- skip blank likes;
- complain if any line is not the same size as line1 (number of cells);
- if any cell in the data contains "?", just skip over it;
- ifa column name contains a "?", then skip over that whole columns;
- Part3: using the Part2 code, generate something like the structure showing above (near enough is good enough).
To test Part2, run a test case like the following. To test Part3, write a dump
method for Tbl (to spit out something like the big print out
shown above).
if __name__=="__main__":
s="""
$cloudCover, $temp, ?$humid, <wind, $playHours
100, 68, 80, 0, 3 # comments
0, 85, 85, 0, 0
0, 80, 90, 10, 0
60, 83, 86, 0, 4
100, 70, 96, 0, 3
100, 65, 70, 20, 0
70, 64, 65, 15, 5
0, 72, 95, 0, 0
0, 69, 70, 0, 4
?, 75, 80, 0, ?
0, 75, 70, 18, 4
60, 72,
40, 81, 75, 0, 2
100, 71, 91, 15, 0
"""
for lst in cells(cols(rows(string(s)))):
print(lst)Output (something like the following, near enough is good enough)
['$cloudCover', '$temp', '<wind', '$playHours']
[100, 68, 0, 3]
[0, 85, 0, 0]
[0, 80, 10, 0]
[60, 83, 0, 4]
[100, 70, 0, 3]
[100, 65, 20, 0]
[70, 64, 15, 5]
[0, 72, 0, 0]
[0, 69, 0, 4]
['?', 75, 0, '?']
[0, 75, 18, 4]
E> skipping line 14
[40, 81, 0, 2]
[100, 71, 15, 0]
The following code handles the tedium of reading from disk, dividing into cells,
converting the strings into the right type. Not handled here are Row and Table and cols creation:
- Place your code in the
hw/2directory. - Place a text file
hw/2/outpart1.txtin that same directory, showing a transcript of Part1 (above) working. - Place a text file
hw/2/outpart2.txtin that same directory, showing a transcript of Part2 (above) working. - Place a text file
hw/2/outpart3.txtin that same directory, showing a transcript of Part3 (above) working. - Place a link to that code in http://tiny.cc/ase19give