Revision 1148
Added by Aaron Marcuse-Kubitza almost 13 years ago
| map | ||
|---|---|---|
| 121 | 121 |
doc0_root = doc0.documentElement |
| 122 | 122 |
if out_label == None: out_label = doc0_root.tagName |
| 123 | 123 |
|
| 124 |
def process_rows(process_row, rows, col_names=None):
|
|
| 124 |
def process_rows(process_row, rows): |
|
| 125 | 125 |
'''Processes input rows |
| 126 | 126 |
@param process_row(in_row, i) |
| 127 | 127 |
''' |
| ... | ... | |
| 130 | 130 |
if i < start: continue |
| 131 | 131 |
if end != None and i >= end: break |
| 132 | 132 |
process_row(row, i) |
| 133 |
|
|
| 134 | 133 |
row_ready(i, row) |
| 135 | 134 |
row_ct = i-start+1 |
| 136 | 135 |
return row_ct |
| ... | ... | |
| 159 | 158 |
|
| 160 | 159 |
in_db = connect_db(in_db_config) |
| 161 | 160 |
in_pkeys = {}
|
| 162 |
|
|
| 163 | 161 |
cur = sql.select(in_db, table=in_root, fields=None, conds=None, |
| 164 | 162 |
limit=end, start=0) |
| 165 |
col_idxs = util.list_flip(sql.col_names(cur)) |
|
| 166 | 163 |
|
| 167 |
mappings_new = [] |
|
| 168 |
for i, mapping in enumerate(mappings): |
|
| 169 |
in_, out = mapping |
|
| 170 |
if metadata_value(in_) == None: |
|
| 171 |
try: mapping = (col_idxs[in_], out) |
|
| 172 |
except KeyError: continue |
|
| 173 |
mappings_new.append(mapping) |
|
| 174 |
mappings = mappings_new |
|
| 164 |
def get_value(in_, row): |
|
| 165 |
try: return util.cast(str, row[in_]) |
|
| 166 |
except KeyError: return None |
|
| 175 | 167 |
|
| 176 |
row_ct = map_rows(lambda in_, row: util.cast(str, row[in_]), |
|
| 177 |
sql.rows(cur)) |
|
| 168 |
col_names = list(sql.col_names(cur)) |
|
| 169 |
col_idxs = util.list_flip(col_names) |
|
| 170 |
def wrap_row(row): return util.ListDict(row, col_names, col_idxs) |
|
| 171 |
row_ct = map_rows(get_value, util.WrapIter(wrap_row, sql.rows(cur))) |
|
| 178 | 172 |
|
| 179 | 173 |
in_db.close() |
| 180 | 174 |
elif in_is_xml: |
| ... | ... | |
| 189 | 183 |
else: # input is CSV |
| 190 | 184 |
map_ = dict(mappings) |
| 191 | 185 |
reader = csv.reader(sys.stdin) |
| 192 |
cols = reader.next() |
|
| 193 |
col_idxs = util.list_flip(cols) |
|
| 194 | 186 |
|
| 195 |
mappings_new = [] |
|
| 196 |
for i, mapping in enumerate(mappings): |
|
| 197 |
in_, out = mapping |
|
| 198 |
if metadata_value(in_) == None: |
|
| 199 |
try: mapping = (col_idxs[in_], out) |
|
| 200 |
except KeyError: continue |
|
| 201 |
mappings_new.append(mapping) |
|
| 202 |
mappings = mappings_new |
|
| 187 |
def get_value(in_, row): |
|
| 188 |
try: return util.none_if(row[in_], '') |
|
| 189 |
except KeyError: return None |
|
| 203 | 190 |
|
| 204 |
def get_value(in_, row): |
|
| 205 |
value = row[in_] |
|
| 206 |
if value != '': return value |
|
| 207 |
else: return None |
|
| 208 |
row_ct = map_rows(get_value, reader) |
|
| 191 |
col_names = reader.next() |
|
| 192 |
col_idxs = util.list_flip(col_names) |
|
| 193 |
def wrap_row(row): return util.ListDict(row, col_names, col_idxs) |
|
| 194 |
row_ct = map_rows(get_value, util.WrapIter(wrap_row, reader)) |
|
| 209 | 195 |
|
| 210 | 196 |
return row_ct |
| 211 | 197 |
|
Also available in: Unified diff
bin/map: Use new util.ListDict and util.WrapIter to simplify getting rows by column name instead of index, and to enable a row to be printed with its column names in error messages