Parse the zipped files into our representation of objects.

[dtbukowski]

Communicate with Nam with how the .wdz files look, then parse those files into the representation of objects that blake and noah decided on.

[dtbukowski]

Made progress on the current state of libzip in the CS Machines with a commit on wdl/parsing_zipped (thought the branch doesn't show up and I may need to push it). Although share objects exist in repositories /usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64/libzip.so/usr/lib/jvm/java-11-openjdk-amd64/lib/libzip.so
/usr/lib/vlc/plugins/access/libzip_plugin.so, I cannot find the zip.h file necessary to complete. There is either a mistake in my Makefile or libzip needs to be installed in usr/lib/x86_64-linux-gnu, so I will use zziplib for now and try asking techstaff to download libzip

[dtbukowski]

Was able to successfully open and close a zip file since I finally had the right libraries for zip.h and the header installed in the Virtual Machine. Libzip isn't installed on the CS machines so as of now I am not able to use libzip on the CS machines.

Will push my progress once I am able to fix errors pushing onto Github. Apparently there are errors with remotes.

[dtbukowski]

Pushed progress on using libzip where I figured out how to get file indexes and get files from name (has to be specific directory) or index.

Next I will use Noah's implementation to make JSON objects and push that onto master

[dtbukowski]

I have pushed code that will make JSON objects, but specifically only for the 5 JSON files in the sandbox in wdl in parsed_zipped directory.

However, the errors present are because of JSON not being present in the VMs so I will either install JSON myself tonight or, depending on how fast techstaff acts, use my own computer with libzip installed.

I have implemented a way to loop through all files in a zip directory, however I will need to figure out how to:
Get files that end in .json (check if filename ends in .json)
Get files that are images and files that are sounds (looked into it here: https://stackoverflow.com/questions/670546/determine-if-file-is-an-image)
Loop through all JSON objects in a .json file (current implementation of json_parse_test that uses the name of the field will not work because new field will require this file to constantly be updating.)

[dtbukowski]

Link to code: https://github.com/uchicago-cs/chiventure/tree/wdl/parsing_zipped/src/wdl/sandbox

[dtbukowski]

Pushed Pseudocode that I will work on tomorrow for parsing zip files into general objects. I will loop through the zip files and in any json files I will loop through them and add the key value pairs to the general object

[dtbukowski]

Turned most of Pseudocode into real code. However, the only parts I was able to test right now are to see if the functions that parsed directories as a string (like game/game/actions.json) was able to parse correctly. If we are able to install libzip soon then we should use the CS machnes. Otherwise anyone who wants to test will need to install json-c and libzip on a VM.

Things done:
Added onto the code right before the loop and added assert to the functions that use
makejsonobj to see if the function works
Code to get objtype from json file name.
Code to see if a file name is a json file.
Code to get specific types of json objects such as string or int to add to file.

Still need to get done:
TEST the file! (Install libzip and json-c on Virtual Machine if we have to wait a long time for libzip)
When objtype is object, currently has json_object and this needs to be converted into libobj. (lines 143-146)
Add necessary header files to parsetojsonobj - it is called parsetojsonobj even though it
parses to libobj implementation so either the name will be changed or to split the file into multiple
files.
Add a method to get assets associated with the .json files.

[dtbukowski]

Moving this issue to Sprint 4 since this issue wasn't complete by Sprint 3. Also assigning Nam to this issue since this is something we need to get done for teams to test their code.