rohancme / zsh_history Goto Github PK

I had high hopes for this tool and spent some time pulling all the histories I could find from years of backup drives. I consider the command history to be the best kind of documentation queuing my mind to any stackoverflow answers or blog posts online with good solutions to problems... always first testing out the solution on the zsh (even python, or awk, or jq, or whatever else).
However, with said workflow I OFTEN have multi-line commands.

I ran the backup from all these files then I tried a restore on a test file and it seems like all commands are truncated to a single line. This is not particularly useful as I can't trust the restoration. Am I doing something wrong or is the design indeed only intended for single line commands?

I am considering implementing a fix, although initially I was gonna write something in awk. Perhaps a regex search to identify full commands might work in python instead of parsing lines 1 by 1.

Please let me know what you think.

Thank you for this, it looks like exactly what I want. However, trying to run it, I got an exception right off the bat. Is it obvious what the problem is? I might try to fix it myself but am not yet familiar with the code.

[patkujawa@LS2-00159 Mon Dec 10 10:56:41 MST 2018] ~/oss/zsh_history (master)
❯ python2 ./src/hist.py -b                                                                                                                                                      ⏎
Traceback (most recent call last):
  File "./src/hist.py", line 121, in <module>
    backup(args.path, args.dbname)
  File "./src/hist.py", line 59, in backup
    u" VALUES(?,?,?)", rows)
sqlite3.ProgrammingError: You must not use 8-bit bytestrings unless you use a text_factory that can interpret 8-bit bytestrings (like text_factory = str). It is highly recommended that you instead just switch your application to Unicode strings.

I had a multiline curl command but it does save all the lines into sql db instead just saved the first line only.