mz_zip_add_mem_to_archive_file_in_place() creates duplicate file entry if file being added already exists in archive

[nyq]

mz_zip_add_mem_to_archive_file_in_place() function seems to work great adding a new file into archive. However, if a file with the same name already exists in the archive, it creates a second entry with the same name instead of overwriting existing file in the archive. I think a better behavior would be:

1. Introduce MZ_ZIP_FLAG_ALLOW_OVERWRITE flag.
2. When mz_zip_add_mem_to_archive_file_in_place() is called with this flag, overwrite a file if it exists in the archive.
3. mz_zip_add_mem_to_archive_file_in_place without MZ_ZIP_FLAG_ALLOW_OVERWRITE flag, refuse to overwrite a file if it exists in the archive and return false.

Alternatively, it would be nice to add a high-level function to delete an existing file from the archive.

[rebirth123]

I just started using this version of miniz and have also discovered this problem. During my tests I called add to zip file (with the same "pArchive_name") a number of times. mz_zip_reader_get_num_files even includes the duplicates into the count. This doesn't feel like it's the proper behaviour. "_in_place" suggests it should overwrite if it doesn't exist.

I'm trying to figure out a workaround, but like I've said, I've only been using it for a couple of hours and only to create, add and read files from and to zip archives. A hint would be welcomed.

[rebirth123]

I got around it the hard way:

Try and open the zip file with "mz_zip_reader_init_file". If successful read the archive names and compare them to the new archive names to be added. If no duplicate found add them to a list of indices to keep. Create a write zip with "mz_zip_writer_init_file"using a temporary filename and clone the indices to keep from the read zip into the write_zip using "mz_zip_writer_add_from_zip_reader". When done, add the new data using either "mz_zip_writer_add_file" or "mz_zip_writer_add_mem"(if working from memory). Finally call "mz_zip_writer_finalize_archive" and then call mz_zip_writer_end and mz_zip_reader_end.

Then just delete the original zip file and rename the new zip file.

[ell1e]

For what it's worth, I vaguely remember having similar issues with Python. And since stuff shifts around when replacing entries, I assume solving it on miniz's side might not be that easy either. So maybe the best solution is to just rewrite changing archives like @rebirth123 did.