Revision 1179
Added by Aaron Marcuse-Kubitza almost 13 years ago
| bin/map | ||
|---|---|---|
| 160 | 160 |
in_pkeys = {}
|
| 161 | 161 |
cur = sql.select(in_db, table=in_root, fields=None, conds=None, |
| 162 | 162 |
limit=end, start=0) |
| 163 |
col_names = list(sql.col_names(cur)) |
|
| 164 |
col_idxs = util.list_flip(col_names) |
|
| 163 | 165 |
|
| 166 |
mappings_new = [] |
|
| 167 |
for i, mapping in enumerate(mappings): |
|
| 168 |
in_, out = mapping |
|
| 169 |
if metadata_value(in_) == None: |
|
| 170 |
try: mapping = (col_idxs[in_], out) |
|
| 171 |
except KeyError: continue |
|
| 172 |
mappings_new.append(mapping) |
|
| 173 |
mappings = mappings_new |
|
| 174 |
|
|
| 164 | 175 |
def get_value(in_, row): |
| 165 |
try: return util.cast(str, row[in_]) |
|
| 166 |
except KeyError: return None |
|
| 167 |
|
|
| 168 |
col_names = list(sql.col_names(cur)) |
|
| 169 |
col_idxs = util.list_flip(col_names) |
|
| 176 |
try: return util.cast(str, row.list[in_]) |
|
| 177 |
except IndexError: return None |
|
| 170 | 178 |
def wrap_row(row): return util.ListDict(row, col_names, col_idxs) |
| 171 | 179 |
row_ct = map_rows(get_value, util.WrapIter(wrap_row, sql.rows(cur))) |
| 172 | 180 |
|
| ... | ... | |
| 183 | 191 |
else: # input is CSV |
| 184 | 192 |
map_ = dict(mappings) |
| 185 | 193 |
reader = csv.reader(sys.stdin) |
| 194 |
col_names = reader.next() |
|
| 195 |
col_idxs = util.list_flip(col_names) |
|
| 186 | 196 |
|
| 197 |
mappings_new = [] |
|
| 198 |
for i, mapping in enumerate(mappings): |
|
| 199 |
in_, out = mapping |
|
| 200 |
if metadata_value(in_) == None: |
|
| 201 |
try: mapping = (col_idxs[in_], out) |
|
| 202 |
except KeyError: continue |
|
| 203 |
mappings_new.append(mapping) |
|
| 204 |
mappings = mappings_new |
|
| 205 |
|
|
| 187 | 206 |
def get_value(in_, row): |
| 188 |
try: return util.none_if(row[in_], '') |
|
| 189 |
except KeyError: return None |
|
| 190 |
|
|
| 191 |
col_names = reader.next() |
|
| 192 |
col_idxs = util.list_flip(col_names) |
|
| 207 |
try: return util.none_if(row.list[in_], '') |
|
| 208 |
except IndexError: return None |
|
| 193 | 209 |
def wrap_row(row): return util.ListDict(row, col_names, col_idxs) |
| 194 | 210 |
row_ct = map_rows(get_value, util.WrapIter(wrap_row, reader)) |
| 195 | 211 |
|
Also available in: Unified diff
bin/map: DB, CSV inputs: Use column indexes instead of column names to look up each field (optimization to avoid repeated dict lookups of the same key)