[SOLVED] Iterator class failing to raise StopIteration on HDF5 data opened with h5py

Iterator class failing to raise StopIteration on HDF5 data opened with h5py

I'm trying to implement an iterable class for the h5 dataset.

class Argh():
    def __init__(self, data):
        self.data = data
        self.c_idx = 0 

    def __getitem__(self, idx):
        return self.data[idx]

    def __len__(self):
        return len(self.data)

    def __next__(self):
        try:
            x = self.data[self.c_idx]
        except IndexError or ValueError:
            raise StopIteration
        self.c_idx += 1
        return x

    def __iter__(self):
        return self

When I try to go through it as through sequence it fails to raise StopIteration exception before ValueError or IndexError occures. This doesn't happen if I just iterate through the h5 dataset directly or if I use iteration by indexing and getitem:

with h5py.File('test.h5', 'w') as f:
     f.create_dataset(name='matrix', data=np.arange(10), dtype=np.float32)

f =  h5py.File('test.h5', 'r')
A = Argh(np.arange(0,10))
B = Argh(f['matrix'])

for x in A: pass
for x in B.data: pass
for i in range(len(B)): B[i]
for x in f['matrix']: pass
for x in B: pass

ValueError  Traceback (most recent call last)
<ipython-input-7-1dcb814e7a79> in <module>
      3 for i in range(len(B)): B[i]
      4 for x in f['matrix']: pass
----> 5 for x in B: pass
ValueError: Index (10) out of range (0-9)

I've tested this on several other objects including zarr arrays but observed this behavior only for h5py opened h5 datasets.

Solution

IndexError or ValueError is an expression that evaluates to IndexError. The syntax for matching multiple exception types is

except (IndexError, ValueError):