carriel-sur weather station
altura : 8.0 mts.
database of wind from the carriel-sur station (january 1966 to the present)
data provided by the dirección meteorológica de Chile (meteochile.gob.cl).
by dr(c). claudio iturra
database of wind from the carriel-sur station (january 1966 to the present)
data provided by the dirección meteorológica de Chile (meteochile.gob.cl).
by dr(c). claudio iturra
171
data-processing routines for carriel-sur data
from pathlib import Path
import re
import pandas as pd
# -------- settings --------
HTML_DIR = Path("/home/cl/filen/database/doctoral-thesis/weather-stations/cs/")
# folder with your *.html files
GLOB = "20*.html" # e.g., 200611.html, 200612.html, etc.
OUT_CSV = HTML_DIR /
():
re.sub(, , (s).strip().lower())
():
s = (s).strip()
(re.(, s))
():
s = (s).strip()
(re.(, s))
():
i, c (cols):
nc = _norm(c)
(re.search(p, nc) p patterns):
i
():
cols = (df.columns)
ni = {i: _norm(c) i,c (cols)}
date_i = _col_idx_by_name(cols, [])
time_i = _col_idx_by_name(cols, [])
date_i time_i :
date_i, time_i
sample = df.head().astype()
i ((cols)):
sample.iloc[:, i].(_looks_like_date).mean() > :
date_i = i
j ((cols)):
sample.iloc[:, j].(_looks_like_time).mean() > :
time_i = j
date_i : date_i =
time_i : time_i =
date_i, time_i
():
s = _norm(s)
s {,,,,,,,}
():
cols = (df.columns)
ncols = [_norm(c) c cols]
dir_i = _col_idx_by_name(cols, [, , ])
spd_i = _col_idx_by_name(cols, [, , ])
vrb_i = _col_idx_by_name(cols, [])
dir_i spd_i :
date_i, time_i = _find_datetime_cols(df)
dt = pd.to_datetime(
df.iloc[:, date_i].astype()..strip() + +
df.iloc[:, time_i].astype()..strip(),
errors=,
dayfirst=
)
out = pd.DataFrame({
: dt,
: pd.to_numeric(df.iloc[:, dir_i], errors=),
: pd.to_numeric(df.iloc[:, spd_i], errors=),
})
vrb_i :
out[] = df.iloc[:, vrb_i].(_to_bool_vrb)
:
out[] = pd.NA
out[] = src_name
out.dropna(subset=[]).sort_values()
():
frames = []
:
tbl pd.read_html(path, flavor=, header=):
res = parse_one_table(tbl, path.name)
res res.empty:
frames.append(res)
Exception:
frames:
pd.DataFrame(columns=[,,,,])
df = pd.concat(frames, ignore_index=)
df = df.drop_duplicates(subset=[], keep=)
df
():
all_frames = []
p (html_dir.glob(pattern)):
df = parse_file(p)
df.empty:
()
:
all_frames.append(df)
all_frames:
pd.DataFrame(columns=[,,,,])
merged = pd.concat(all_frames, ignore_index=)
merged = (merged.sort_values([,])
.drop_duplicates(subset=[], keep=))
merged
__name__ == :
ts = parse_all()
ts.to_csv(OUT_CSV, index=)
()
