I have encountered a zip archive, which Stata fails to unpack.
The specific message is "I/O error on read, error may be caused by incorrect password" r(601)

A bunch of other files from the same archive are unzipped correctly.
I have checked for the archive integrity using a bunch of alternative tools, (winrar and other) and none of them have reported any problem.
I have checked for problems with data transfer, no problem.
I have checked and confirmed that password protection was not engaged anywhere during the production of this file.
On failure the file is created with the zero length and remains open (locked) until the whole Stata session is closed. I have found no way to close the file and delete the zero-length file.
The offending file is characterized as "deflated, 84%" and the original size is 22,790 while compressed size is 3,606 bytes. All of the files I've checked from this archive use the same "deflated" compression.
I believe this is a Stata problem, rather than the zip-file problem.
Observed in Stata 15.0 for Windows and a bunch of earlier versions.
If the problem is confirmed, perhaps Stata could
A) skip the files it can't unpack and continue, reporting the number of files skipped.
B) provide a way to acquire the list of files in the archive and unpack selectively.
C) properly close the file when any error is encountered.
Thank you, Sergiy Radyakin
The specific message is "I/O error on read, error may be caused by incorrect password" r(601)
A bunch of other files from the same archive are unzipped correctly.
I have checked for the archive integrity using a bunch of alternative tools, (winrar and other) and none of them have reported any problem.
I have checked for problems with data transfer, no problem.
I have checked and confirmed that password protection was not engaged anywhere during the production of this file.
On failure the file is created with the zero length and remains open (locked) until the whole Stata session is closed. I have found no way to close the file and delete the zero-length file.
The offending file is characterized as "deflated, 84%" and the original size is 22,790 while compressed size is 3,606 bytes. All of the files I've checked from this archive use the same "deflated" compression.
I believe this is a Stata problem, rather than the zip-file problem.
Observed in Stata 15.0 for Windows and a bunch of earlier versions.
If the problem is confirmed, perhaps Stata could
A) skip the files it can't unpack and continue, reporting the number of files skipped.
B) provide a way to acquire the list of files in the archive and unpack selectively.
C) properly close the file when any error is encountered.
Thank you, Sergiy Radyakin
Comment