Python logger.info函数代码示例

def main():
    filetypes = Ska.Table.read_ascii_table('filetypes.dat')
    if len(sys.argv) == 2:
        filetypes = filetypes[ filetypes['content'] == sys.argv[1].upper() ]

    loglevel = pyyaks.logger.INFO
    logger = pyyaks.logger.get_logger(level=loglevel, format="%(message)s")

    for filetype in filetypes:
        ft.content = filetype.content.lower()

        orig_files_glob = os.path.join(orig_arch_files['contentdir'].abs, filetype['fileglob'])
        logger.info('orig_files_glob=%s', orig_files_glob)
        for f in glob.glob(orig_files_glob):
            ft.basename = os.path.basename(f)
            tstart = re.search(r'(\d+)', ft.basename).group(1)
            datestart = DateTime(tstart).date
            ft.year, ft.doy = re.search(r'(\d\d\d\d):(\d\d\d)', datestart).groups()

            archdir = arch_files['archdir'].abs
            archfile = arch_files['archfile'].abs

            if not os.path.exists(archdir):
                print 'Making dir', archdir
                os.makedirs(archdir)
                
            if not os.path.exists(archfile):
                # logger.info('mv %s %s' % (f, archfile))
                shutil.move(f, archfile)

def make_archfiles_db(filename, content_def):
    # Do nothing if it is already there
    if os.path.exists(filename):
        return

    datestart = DateTime(DateTime(opt.start).secs - 60)
    tstart = datestart.secs
    tstop = tstart
    year, doy = datestart.date.split(':')[:2]
    times, indexes = derived.times_indexes(tstart, tstop, content_def['time_step'])

    logger.info('Creating db {}'.format(filename))
    archfiles_def = open('archfiles_def.sql').read()
    db = Ska.DBI.DBI(dbi='sqlite', server=filename)
    db.execute(archfiles_def)
    archfiles_row = dict(filename='{}:0:1'.format(content_def['content']),
                         filetime=0,
                         year=year,
                         doy=doy,
                         tstart=tstart,
                         tstop=tstop,
                         rowstart=0,
                         rowstop=0,
                         startmjf=indexes[0], # really index0
                         stopmjf=indexes[-1],  # really index1
                         date=datestart.date)
    db.insert(archfiles_row, 'archfiles')

def try4times(func, *arg, **kwarg):
    """
    Work around problems with sqlite3 database getting locked out from writing,
    presumably due to read activity.  Not completely understood.

    This function will try to run func(*arg, **kwarg) a total of 4 times with an
    increasing sequence of wait times between tries.  It catches only a database
    locked error.
    """
    from django.db.utils import OperationalError

    for delay in 0, 5, 10, 60:
        if delay > 0:
            time.sleep(delay)

        try:
            func(*arg, **kwarg)
        except OperationalError as err:
            if 'database is locked' in str(err):
                # Locked DB, issue informational warning
                logger.info('Warning: locked database, waiting {} seconds'.format(delay))
            else:
                # Something else so just re-raise
                raise
        else:
            # Success, jump out of loop
            break

    else:
        # After 4 tries bail out with an exception
        raise OperationalError('database is locked')

def main():
    global opt
    opt = get_opt()
    info = {'date': opt.stop,
            'start': opt.start,
            'stop': opt.stop,
            'box_duration_months': opt.box_duration}

    asol_aimpoint = get_asol(info)

    asol_monthly = AsolBinnedStats(asol_aimpoint, 365.25 / 12)
    for det in ('ACIS-S', 'ACIS-I'):
        asol_monthly.det = det
        det_title = asol_monthly.det_title
        info[det_title] = asol_monthly.get_chip_x_y_info()
        asol_monthly.plot_chip_x_y(info[det_title])

    asol_monthly.plot_intra_obs_dy_dz()

    plot_housing_temperature()

    info_file = os.path.join(opt.data_root, 'info.json')
    with open(info_file, 'w') as fh:
        logger.info('Writing info file {}'.format(info_file))
        json.dump(make_pure_python(info), fh, indent=4, sort_keys=True)

def read_derived(i, filename, filetype, row, colnames, archfiles, db):
    """Read derived data using eng_archive and derived computation classes.
    ``filename`` has format <content>_<index0>_<index1> where <content>
    is the content type (e.g. "dp_thermal128"), <index0> is the start index for
    the new data and index1 is the end index (using Python slicing convention
    index0:index1).  Args ``i``, ``filetype``, and ``row`` are as in
    read_archive().  ``row`` must equal <index0>.  ``colnames`` is the list of
    column names for the content type.
    """
    # Check if filename is already in archfiles.  If so then abort further processing.

    if db.fetchall('SELECT filename FROM archfiles WHERE filename=?', (filename,)):
        logger.verbose('File %s already in archfiles - skipping' % filename)
        return None, None

    # f has format <content>_<index0>_<index1>
    # <content> has format dp_<content><mnf_step> e.g. dp_thermal128
    content, index0, index1 = filename.split(':')
    index0 = int(index0)
    index1 = int(index1)
    mnf_step = int(re.search(r'(\d+)$', content).group(1))
    time_step = mnf_step * derived.MNF_TIME
    times = time_step * np.arange(index0, index1)

    logger.info('Reading (%d / %d) %s' % (i, len(archfiles), filename))
    vals = {}
    bads = np.zeros((len(times), len(colnames)), dtype=np.bool)
    for i, colname in enumerate(colnames):
        if colname == 'TIME':
            vals[colname] = times
            bads[:, i] = False
        else:
            dp_class = getattr(Ska.engarchive.derived, colname.upper())
            dp = dp_class()
            dataset = dp.fetch(times[0] - 1000, times[-1] + 1000)
            ok = (index0 <= dataset.indexes) & (dataset.indexes < index1)
            vals[colname] = dp.calc(dataset)[ok]
            bads[:, i] = dataset.bads[ok]

    vals['QUALITY'] = bads
    dat = Ska.Numpy.structured_array(vals, list(colnames) + ['QUALITY'])

    # Accumlate relevant info about archfile that will be ingested into
    # MSID h5 files.  Commit info before h5 ingest so if there is a failure
    # the needed info will be available to do the repair.
    date = DateTime(times[0]).date
    year, doy = date[0:4], date[5:8]
    archfiles_row = dict(filename=filename,
                         filetime=int(index0 * time_step),
                         year=year,
                         doy=doy,
                         tstart=times[0],
                         tstop=times[-1],
                         rowstart=row,
                         rowstop=row + len(dat),
                         startmjf=index0,
                         stopmjf=index1,
                         date=date)

    return dat, archfiles_row

def make_msid_file(colname, content, content_def):
    ft['content'] = content
    ft['msid'] = colname
    filename = msid_files['data'].abs
    if os.path.exists(filename):
        return

    logger.info('Making MSID data file %s', filename)

    if colname == 'TIME':
        dp_vals, indexes = derived.times_indexes(opt.start, opt.stop,
                                                 content_def['time_step'])
    else:
        dp = content_def['classes'][colname]()
        dataset = dp.fetch(opt.start, opt.stop)
        dp_vals = np.asarray(dp.calc(dataset), dtype=dp.dtype)

    # Finally make the actual MSID data file
    filters = tables.Filters(complevel=5, complib='zlib')
    h5 = tables.openFile(filename, mode='w', filters=filters)
    
    n_rows = int(20 * 3e7 / content_def['time_step'])
    h5shape = (0,) 
    h5type = tables.Atom.from_dtype(dp_vals.dtype)
    h5.createEArray(h5.root, 'data', h5type, h5shape, title=colname,
                    expectedrows=n_rows)
    h5.createEArray(h5.root, 'quality', tables.BoolAtom(), (0,), title='Quality',
                    expectedrows=n_rows)

    logger.info('Made {} shape={} with n_rows(1e6)={}'.format(colname, h5shape, n_rows / 1.0e6))
    h5.close()

def move_archive_files(filetype, archfiles):
    ft['content'] = filetype.content.lower()

    stagedir = arch_files['stagedir'].abs
    if not os.path.exists(stagedir):
        os.makedirs(stagedir)

    for f in archfiles:
        if not os.path.exists(f):
            continue
        ft['basename'] = os.path.basename(f)
        tstart = re.search(r'(\d+)', str(ft['basename'])).group(1)
        datestart = DateTime(tstart).date
        ft['year'], ft['doy'] = re.search(r'(\d\d\d\d):(\d\d\d)', datestart).groups()

        archdir = arch_files['archdir'].abs
        archfile = arch_files['archfile'].abs

        if not os.path.exists(archdir):
            os.makedirs(archdir)

        if not os.path.exists(archfile):
            logger.info('mv %s %s' % (os.path.abspath(f), archfile))
            if not opt.dry_run:
                if not opt.occ:
                    shutil.copy2(f, stagedir)
                shutil.move(f, archfile)

        if os.path.exists(f):
            logger.verbose('Unlinking %s' % os.path.abspath(f))
            os.unlink(f)

def create_content_dir():
    """
    Make empty files for colnames.pkl, colnames_all.pkl and archfiles.db3
    for the current content type ft['content'].

    This only works within the development (git) directory in conjunction
    with the --create option.
    """
    dirname = msid_files['contentdir'].abs
    if not os.path.exists(dirname):
        logger.info('Making directory {}'.format(dirname))
        os.makedirs(dirname)

    empty = set()
    if not os.path.exists(msid_files['colnames'].abs):
        with open(msid_files['colnames'].abs, 'wb') as f:
            pickle.dump(empty, f, protocol=0)
    if not os.path.exists(msid_files['colnames_all'].abs):
        with open(msid_files['colnames_all'].abs, 'wb') as f:
            pickle.dump(empty, f, protocol=0)

    if not os.path.exists(msid_files['archfiles'].abs):
        archfiles_def = open('archfiles_def.sql').read()
        filename = msid_files['archfiles'].abs
        logger.info('Creating db {}'.format(filename))
        db = Ska.DBI.DBI(dbi='sqlite', server=filename, autocommit=False)
        db.execute(archfiles_def)
        db.commit()

def make_h5_col_file(dats, colname):
    """Make a new h5 table to hold column from ``dat``."""
    filename = msid_files['msid'].abs
    filedir = os.path.dirname(filename)
    if not os.path.exists(filedir):
        os.makedirs(filedir)

    # Estimate the number of rows for 20 years based on available data
    times = np.hstack([x['TIME'] for x in dats])
    dt = np.median(times[1:] - times[:-1])
    n_rows = int(86400 * 365 * 20 / dt)

    filters = tables.Filters(complevel=5, complib='zlib')
    h5 = tables.openFile(filename, mode='w', filters=filters)

    col = dats[0][colname]
    h5shape = (0,) + col.shape[1:]
    h5type = tables.Atom.from_dtype(col.dtype)
    h5.createEArray(h5.root, 'data', h5type, h5shape, title=colname,
                    expectedrows=n_rows)
    h5.createEArray(h5.root, 'quality', tables.BoolAtom(), (0,), title='Quality',
                    expectedrows=n_rows)
    logger.info('Made {} shape={} with n_rows(1e6)={}'
                .format(colname, h5shape, n_rows / 1.0e6))
    h5.close()

def get_obsid(obsid, dt=3.0):
    """
    Get an obsid
    """
    obsids = events.obsids.filter(obsid__exact=obsid)
    if len(obsids) == 0:
        raise ValueError('No obsid={} in kadi database'.format(obsid))

    dwells = events.dwells.filter(obsids[0].start, obsids[0].stop)
    obsid_dwells = [dwell for dwell in dwells if dwell.start > obsids[0].start]
    logger.info('Using obsid dwell(s): {}'
                .format(','.join(str(dwell) for dwell in obsid_dwells)))

    scs107s = events.scs107s.filter(obsid=obsid)
    tstart = DateTime(obsid_dwells[0].start).secs
    if len(scs107s) > 0:
        tstop = scs107s[0].tstop - 200
    else:
        tstop = DateTime(obsid_dwells[-1].stop).secs
    if tstop - tstart < 2000:
        raise ValueError('Observation interval too short {}'.format(tstop - tstart))
    telems, slots = get_archive_data(tstart, tstop)
    out = telems_to_struct(telems, slots)
    out['obsid'] = obsid

    return out

def copy_statfiles_to_test(stat, dt, tstart, tstop):
    ft['interval'] = stat
    colnames = pickle.load(open(msid_files['colnames'].abs))
    for colname in colnames:
        ft['msid'] = colname
        if os.path.exists(test_msid_files['stats'].abs):
            continue
        if os.path.exists(msid_files['stats'].abs):
            logger.info('Copying {0} stats for MSID {1}'.format(stat, colname))
            statdir = os.path.dirname(test_msid_files['stats.tmp'].abs)
            if not os.path.exists(statdir):
                os.makedirs(statdir)
            shutil.copy(msid_files['stats'].abs, test_msid_files['stats.tmp'].abs)
            h5 = tables.openFile(test_msid_files['stats.tmp'].abs, 'a')
            times = (h5.root.data.col('index') + 0.5) * dt
            row0, row1 = np.searchsorted(times, [tstart, tstop])
            #print colname, row0, row1, len(times), DateTime(times[row0]).date, DateTime(times[row1]).date,
            # Remove from row1-1 to end.  The row1-1 is because it is possible
            # to get the daily stat without the rest of the 5min data if
            # tstop is past noon of the day.  This messes up update_archive.
            h5.root.data.removeRows(row1 - 1, h5.root.data.nrows)
            h5.root.data.removeRows(0, row0)
            h5.copyFile(test_msid_files['stats'].abs, overwrite=True)
            newtimes = (h5.root.data.col('index') + 0.5) * dt
            #print len(newtimes), DateTime(newtimes[0]).date, DateTime(newtimes[-1]).date
            h5.close()
            os.unlink(test_msid_files['stats.tmp'].abs)

def del_stats(colname, time0, interval):
    """Delete all rows in ``interval`` stats file for column ``colname`` that
    occur after time ``time0`` - ``interval``.  This is used to fix problems
    that result from a file misorder.  Subsequent runs of update_stats will
    refresh the values correctly.
    """
    dt = {'5min': 328,
          'daily': 86400}[interval]

    ft['msid'] = colname
    ft['interval'] = interval
    stats_file = msid_files['stats'].abs
    if not os.path.exists(stats_file):
        raise IOError('Stats file {} not found'.format(stats_file))

    logger.info('Fixing stats file %s after time %s', stats_file, DateTime(time0).date)

    stats = tables.open_file(stats_file, mode='a',
                            filters=tables.Filters(complevel=5, complib='zlib'))
    index0 = time0 // dt - 1
    indexes = stats.root.data.col('index')[:]
    row0 = np.searchsorted(indexes, [index0])[0] - 1
    if opt.dry_run:
        n_del = len(stats.root.data) - row0
    else:
        n_del = stats.root.data.remove_rows(row0, len(stats.root.data))
    logger.info('Deleted %d rows from row %s (%s) to end', n_del, row0,
                DateTime(indexes[row0] * dt).date)
    stats.close()

def get_stats_over_time(start, stop=None, sp=False, dp=None, ir=False, ms=None,
                        slots='combined', t_samp=1000):
    """
    Equivalent to get_stats_per_interval, but concatenate the results for all
    obsids within the specified time interval.
    """
    # Get obsids in time range and collect all the per-interval statistics
    obsids = events.obsids.filter(start, stop, dur__gt=2000)
    stats_list = []
    for obsid in obsids:
        set_FILES_context(obsid.obsid, sp, dp, ir, ms, t_samp, slots)

        # First check that there is the raw dat file for this obsid.  Nothing
        # can be done without this.
        dat_file = FILES['dat.pkl'].rel
        if not os.path.exists(dat_file):
            logger.info('Skipping {}: {} not in archive'.format(obsid, dat_file))
            continue

        # Now get the stats for this obsid.  Hopefully it has already been computed and
        # is cached as a file.  If not, try to compute the stats (and cache).  If that
        # fails then press on but touch a file to indicate failure so subsequent attempts
        # don't bother.
        logger.info('Processing obsid {}'.format(obsid))
        try:
            stats = get_cached_stats()  # depends on the context set previously
        except FailedStatsFile:
            # Previously failed
            logger.info('  Skipping {}: failed statistics'.format(obsid.obsid))
            continue
        except NoStatsFile:
            logger.info('  Reading pickled data file {}'.format(dat_file))
            dat = pickle.load(open(dat_file, 'r'))
            try:
                logger.info('  Computing statistics')
                if slots == 'combined':
                    stats = get_stats_per_interval_combined(dat, sp, dp, ir, ms, t_samp)
                else:
                    stats = get_stats_per_interval_per_slot(dat, sp, dp, ir, ms, slots, t_samp)
            except ValueError as err:
                open(FILES['stats.ERR'].rel, 'w')  # touch file to indicate failure to compute stats
                logger.warn('  ERROR: {}'.format(err))

        stats['obsid'] = obsid.obsid
        stats_list.append(stats)

    stats = {}
    for case in STAT_CASES:
        stats[case] = {}
        for stat_type in STAT_TYPES:
            stats[case][stat_type] = np.hstack([x[case][stat_type] for x in stats_list])

    # Set corresponding array of obsids for back-tracing outliers etc
    stats['obsid'] = np.hstack([np.ones(len(x['obc']['std']), dtype=int) * x['obsid']
                                for x in stats_list])

    return stats

def cut_stars(ai):
    starfiles = glob(os.path.join(ai['outdir'],
                                  "*stars.txt"))
    shutil.copy(starfiles[0], starfiles[0] + ".orig")
    starlines = open(starfiles[0]).read().split("\n")
    for slot in ai['skip_slot']:
        starlines = [i for i in starlines
                     if not re.match("^\s+{}\s+1.*".format(slot), i)]
    logger.info('Cutting stars by updating {}'.format(starfiles[0]))
    with open(starfiles[0], "w") as newlist:
        newlist.write("\n".join(starlines))

def get_obsid_data(obsid):
    filename = os.path.join('data', str(obsid) + '.pkl')
    if os.path.exists(filename):
        dat = pickle.load(open(filename, 'r'))
    else:
        import update_flags_archive
        dat = update_flags_archive.get_obsid(obsid)
        pickle.dump(dat, open(filename, 'w'), protocol=-1)
        logger.info('Wrote data for {}'.format(obsid))

    return dat