d-chambers
commented
8 years ago Ok it looks like the problem is caused by detex splitting each part of the path apart to store less data in the SQLite database; this is crude representation of an enum (enumerated type).
Let me walk you through my reasoning for doing this, then we can think about fixes.
If we had a path /eq/POTTER/ContinuousWaveForms/120/event.mseed we could store the whole path in the SQLite database table where the data file info is stored but then the path field would require at least 35 chars. If we assume ascii chars (although they might be unicode, I am not sure) that would require 35 bytes. However, a lot of other events probably share much of the same path, so why not just store a single integer reference to each part of the path?
So in the 'indkey' table you could have something like:
| 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|
| eq | POTTER | ContinuousWaveForms | 2010 | 120, 121, 122 |
Then we have a fields in the "ind" table (one row for each data file) that look like this;
| path | name |
|---|---|
| 0,0,0,0,0 | event.mseed |
| 0,0,0,0,0 | event1.mseed |
To reconstruct the path we can use the path list and do an os.join from table "indkey" os.path.join(0[0], 1[0], 2[0], 3[0], 4[0]) = eq/POTTER/ContinuousWaveForms/210/120. This then allows us to save only 5 ints to represent the entire 32 char path. The savings could be big if you have a lot of data files.
One thing I have never checked on, however, is whether the sqlite database already does this type of compression (or a better type). If it does, I wasted my time and made the code unnecessarily complicated, so I will check on this and consider a revision.
So the problem in your case is in the reconstruction. In the example above we recover the path "eq/POTTER/ContinuousWaveForms/210/120" which looks like it should exist but it actually doesn't, it needs a "/" at the beginning, so the correct path is "/eq/POTTER/ContinuousWaveForms/210/120". If you are using the index in the same relative path you are running detex from there is no issue, but in your case there is an issue. A quick fix is to simply add another try except bracket in the _checkQuality function of getdata to try to read the path with a os.sep appended to the beginning. It looks like so:
def _checkQuality(stPath):
"""
load a path to an obspy trace and check quality
"""
try:
st = obspy.read(stPath)
except (TypeError, IOError): # if object is not obspy-readable
try:
st = osbpy.read(os.path.join(os.path.sep, stPath))
except:
return None
lengthStream = len(st)
gaps = st.get_gaps()
gapsum = np.sum([x[-2] for x in gaps])
starttime = min([x.stats.starttime.timestamp for x in st])
endtime = max([x.stats.endtime.timestamp for x in st])
duration = endtime - starttime
nc = len(list(set([x.stats.channel for x in st])))
netsta = st[0].stats.network + '.' + st[0].stats.station
outDict = {'Gaps': gapsum, 'Starttime' : starttime, 'Endtime' : endtime,
'Duration' : duration,There could, however, be a bigger issue if multiple people are using the same data directory and the index created is specific to the user. In which case, the indexDirectory function might need to use abspath and not normpath
def indexDirectory(dirPath):
"""
Create an index (.index.db) for a directory with stored waveform files
which also contains quality info of each file
Parameters
__________
dirPath : str
The path to the directory containing waveform data (any structure)
"""
columns = ['Path', 'FileName', 'Starttime', 'Endtime', 'Gaps', 'Nc', 'Nt',
'Duration', 'Station']
df = pd.DataFrame(columns=columns) # DataFrame for indexing
msg = 'indexing, or updating index for %s' % dirPath
detex.log(__name__, msg, level='info', pri=True)
# Create a list of possible path permutations to save space in database
pathList = [] # A list of lists with different path permutations
for dirpath, dirname, filenames in os.walk(dirPath):
dirList = os.path.abspath(dirpath).split(os.path.sep)
...
Trying to run SVD or detections on data has resulted in Detex being unable to read the Continuous data files.
I have tried running the data in pickle and mseed format. Neither has worked properly.
Using data that was pulled by previous versions of Detex I received this error (file format pickle):
When trying to run SVD on the older data, it skips all waveforms and ends up with no data to run SVD.
Thinking that there might be something wrong with the continuous waveform data, I downloaded a month's worth of continuous data to a new directory. This was completed in both pickle and mseed format using Detex 1.0.6. After the download completed, Detex began to auto-index the ContinuousWaveform directory with this result:
I tried running SVD (after terminating the auto indexing and Detex tried to index again), with this result:
Just to try it, I created a subspace with the new data:
I have been able to successfully use detex.pickTimes() and have been able to see those waveforms. Detex has also had no problem reading either pickle or mseed format EventWaveforms. The clusters have been produced without error. I am using fillZeros=True, but all other parameters (minus directory location variables) have remained unchanged.
My TemplateKey: TemplateKey.txt
My Station Key: StationKey.txt
*Both have been switched to ".txt" for uploading.
*If you choose to download the data, the four stations will be about 9 GB of data for the single month. The event waveforms take up approximately 50 MB of disk space.