Description
| Q | A |
|---|---|
| Library version(s) affected: | 3.3.3 |
| PHP version(s): | 7.2 |
| OS (with bit depth): | Alpine Docker |
Description
When an archive contains a lot of entries (for example 22k+), library always loads it into memory (internal array) of entries, which leads to out of memory problems (adding more RAM is not a solution). Also the same is for zipping large amount of files (around 7k is enough) where is the same problem: entries are collected into an internal array.
Maybe it's kind of limitation of this library to zip/unzip relatively small amount of files as it's not probably intended to do such huge works; my original reason is making deployment packages using ZIP (both zip/unzip) in quite dark environment, so a lot of processed files are required in a bit restricted environment.
How to reproduce
Set memory limit of php script to for example 32MB and try zip/unzip folder with 10k files; it will die on memory usage.
Possible Solution
This could be hard to do as there is an internal array of entries; solution could be zipping on fly when an entry is added and also reading entries using for example stream/iterator. Functions like hasEntry() would work with quite low performance on huge entry set as it has to go through all entries in the stream.
Additional context
Nothing to say here.