Using Plex Sqlite3[sandbox] alongside other standard sqlite3 installs

Plex Media Server Metadata & Adding Files
1

Server Version#:MacOS Version 1.43.0.10492-121068a07
Player Version#: n/a
Using sqlite3 direct in Terminal (iTerm2)

I have been exploring EPG metadata (Plex pass) to see if I can find upcoming films across all channels. I am making progress BUT I believe when using the Plex binary (/Applications/Plex\ Media\ Server.app/Contents/MacOS/Plex\ SQLite) it impacts my home ~/.sqlite_history

It sees the V1 history file that my latest sqlite3 install (3.51.2 2026-01-09 17:27:48 b270f8339eb13b504d0b2ba154ebca966b7dde08e40c3ed7d559749818cb2075 (64-bit)) uses and decides to overwrite it with a new V2 file, sample:

head ~/.sqlite_history
HiStOrY_V2
.ex
.schema
select\040items.begins_at,\040tv.tag\040AS\040Channel,\040show.title\040AS\040Show,\040episode.summary\040AS\040’Episode’\040FROM\040metadata_items\040AS\040show\040\040JOIN\040metadata_items\040AS\040season\040ON\040show.id\040=\040season.parent_id\040JOIN\040metadata_items\040AS\040episode\040ON\040season.id\040=\040episode.parent_id\040JOIN\040media_items\040AS\040items\040ON\040episode.id\040=\040items.metadata_item_id\040JOIN\040tags\040AS\040tv\040on\040items.channel_id\040=\040tv.id\040order\040by\040items.begins_at;;
select\040items.begins_at,\040tv.tag\040AS\040Channel,\040show.title\040AS\040Show,\040episode.summary\040AS\040’Episode’\040FROM\040metadata_items\040AS\040show\040\040JOIN\040metadata_items\040AS\040season\040ON\040show.id\040=\040season.parent_id\040JOIN\040metadata_items\040AS\040episode\040ON\040season.id\040=\040episode.parent_id\040JOIN\040media_items\040AS\040items\040ON\040episode.id\040=\040items.metadata_item_id\040JOIN\040tags\040AS\040tv\040on\040items.channel_id\040=\040tv.id\040order\040by\040items.begins_at\040limit\0403;

This is a bit of disaster since I rely on recalling old sql lines from history for my other ongoing projects on environmental data etc. These are all in the V1 format, example:

.ex
select * from DayInfo order by dayID desc limit 19;
update dayinfo set daynote=‘s18,25,27,24’ where dayid=850;
update dayinfo set dayRainMM=1 where dayid=850;
update dayinfo set PVkwHx10=36 where dayid=850;
insert into DayInfo (dayCH,dayRainMM,dayFire,dayNote,dayXS_date ) values (1,0,‘’,‘’,‘2026/03/02’);
select * from DayInfo order by dayID desc limit 19;
.ex

How can I interoperate with both?

I don’t want to be setting environment variables before and after each use of the Plex variant if at all possible. I searched the Plex binary for the file name string in case I could try an edit in place but it was not seen…

Any suggestions??

2

One thought of my own is to open a copy of EPG cloud data with a read only mode of my normal sqlite3. Seems the Plex mods are around locking etc and ONLY impactful around WRITES. Which I hope not to need to do…

The events did make me realise I don’t know how/where/if the Plex invocations implement any environment variables or equivalents for paths etc. Like SQLITE_HISTORY to set the path to the command history.

While this workaround maybe solves a read only use the general solution to allow writes would be useful. At least some awareness or MENTION of the issue might help avoid data loss for others …