"Convert" zlib stream to gzip?-CodePudding

I have a blob of data which is compressed with zlib, now I need to decompress it but only have gzip at hand in a Linux environment. No possibility of getting any other tools unfortunately. Is there any way to "convert" the zlib data into something that gzip can decompress?

I read this one: https://unix.stackexchange.com/questions/22834/how-to-uncompress-zlib-data-in-unix

Where it's basically said to add a 10 bytes header and pass it into gzip. However, when I try this I don't get any output. The out file is empty.

printf "\x1f\x8b\x08\x00\x00\x00\x00\x00\x00\x00" |cat - /tmp/data |gzip -dc >/tmp/out

/tmp/data contains a chunk of data that starts with these two bytes: \x78\xDA (which I interpret as a zlib header).

Any suggestions? I've also tried removing these 2 bytes before prepending the gzip header.

Edit: After some more thinking and after reading the comments here I realized that the input data was in fact not correct. Using another source for test data I got a bit further. See below, encoded as base64 since I could not attach binary data.

I have the source data:

eJwNzbEvA1EcwPEcgsUs6UIisZ17v7tre2ehaR Kq/TaU0XEvddTd9yrtq8NFiQ1GGwisWnSQQSJwWoSJgMWxGBCYvEPEB2/y f7WIvWhf5629jWXE/ShJ/erNDVEEgMtwXIydZ9e2DxbtLQhda60LCrO6eEuuuSZSYGrelofBBBDEvxWZAVNRiRZMBofCYWtpLZEI5OGkomHc6O9GmEFpoeybEWkuOejGQQkSoiIG6ZmwXilHjvRIEui3FW5i6vcOLb/pQ5KgJCSAHQVOKvrWJLAklGEgRRmJRzLYTnXU1FQUVHKESqax4A6IrarDDNFYgHeqj50ESQKbP9BllinlEpc6fk24yRPHNwyV2x6RLjnmGl0tg0IokEzTPuYDM EaFVp TAABpAM8Pjtcc9u/PrYdHJFK f37vfAocdmxf5s5uUmJ73Lz86F/z93233c P1  mlOPR369GrVqzvnh8dH95qB/QfS/18rQ==

Then I strip the two first bytes, which gives me this:

Dc2xLwNRHMDxHILFLOlCIrGde7 7a3tnoWkfiqv02lNFxL3XU3fcq7avDRYkNRhsIrFp0kEEicFqEiYDFsRgQmLxDxAdv8vn 1iL1oX etvY1lxP0oSf3qzQ1RBIDLcFyMnWfXtg8W7S0IXWutCwqzunhLrrkmUmBq3paHwQQQxL8VmQFTUYkWTAaHwmFraS2RCOThpKJh3OjvRphBaaHsmxFpLjnoxkEJEqIiBumZsF4pR470SBLotxVuYur3Di2/6UOSoCQkgB0FTir61iSwJJRhIEUZiUcy2E511NRUFFRyhEqmseAOiK2qwwzRWIB3qo dBEkCmz/QZZYp5RKXOn5NuMkTxzcMldsekS455hpdLYNCKJBM0z7mAzPhGhVafkwAAaQDPD47XHPbvz62HRyRSvn9 73wKHHZsX bOblJie9y8/Ohf8/d9t93Pj9fvppTj0d vRq1as754fHR/eagf0H0v9fK0=

Writing that (in binary, not base64 - just to be clear) to a file: /tmp/inputdata I run the following shell command:

printf "\x1f\x8b\x08\x00\x00\x00\x00\x00\x00\x00" |cat - /tmp/inputdata| gzip -dc >/tmp/out

Which gives me this:

gzip: corrupted data

The contents of /tmp/out (again, base64 encoded):

0oRDoQEmoQRIf1sfUVIx8CBZAQ2kAWJERQQaYqh/zQYaYMdMTTkBA6EBpGF2gapiY2l4L1VSTjpVVkNJOjAxREUvSVoxMjM0NUEvMjFFMEpYRDdVUVk2RUNMTTNXVDdZRiM4YmNvYkRFYmRuAmJkdGoyMDIxLTA0LTAxYmlzdFJvYmVydCBLb2NoLUluc3RpdHV0Ym1hbU9SRy0xMDAwMzExODRibXBsRVUvMS8yMC8xNTA3YnNkAmJ0Z2k4NDA1MzkwMDZidnBqMTExOTM0OTAwN2Nkb2JqMTk2NC0wOC0xMmNuYW2kYmZuak11c3Rlcm1hbm5iZ25lRXJpa2FjZm50ak1VU1RFUk1BTk5jZ250ZUVSSUtBY3ZlcmUxLjAuMFhASoTSiWEI6NFgZVdxvtjgF9walgd6rmesxFMtVFxtseYIXm2N/YBp53na69PZcT/ xmpjtQNFOYWtmaCWxjiUYw==

CodePudding user response：

You got it right. Strip the first two bytes and prepend those ten bytes. Then gzip -dc will complain about an "unexpected end of file", but it will nevertheless decompress and write the result to your output file.

Unless the zlib stream is corrupt to begin with.