segfault when trying to run

[digitalsignalperson]

I'm arch linux with the 0.16.0 package

I ran astroid --new-config initially.

astroid --new-config
[2023-05-25 16:42:20.814875] [0x00007f7e265ae580] [info]    welcome to astroid! - 0.16.0
[2023-05-25 16:42:20.841616] [0x00007f7e265ae580] [info]    creating new config..
[2023-05-25 16:42:20.841626] [0x00007f7e265ae580] [debug]   HOME: /home/andy
[2023-05-25 16:42:20.841637] [0x00007f7e265ae580] [info]    writing default config to: /home/andy/.config/astroid/config
[2023-05-25 16:42:20.841640] [0x00007f7e265ae580] [info]    cf: loading: "/home/andy/.config/astroid/config"
[2023-05-25 16:42:20.841821] [0x00007f7e265ae580] [warning] cf: writing back config to: "/home/andy/.config/astroid/config"

If I run notmuch

Notmuch is configured and appears to have a database. Excellent!
...

trying to run astroid

astroid
[2023-05-25 16:44:57.737693] [0x00007f745c812580] [info]    welcome to astroid! - 0.16.0
[2023-05-25 16:44:57.756507] [0x00007f745c812580] [debug]   HOME: /home/andy
[2023-05-25 16:44:57.756520] [0x00007f745c812580] [info]    cf: loading: "/home/andy/.config/astroid/config"
[2023-05-25 16:44:57.756726] [0x00007f745c812580] [info]    cf: version: 11
[2023-05-25 16:44:57.756797] [0x00007f745c812580] [debug]   cf: check config..
[2023-05-25 16:44:57.756808] [0x00007f745c812580] [warning] cf: missing values in config have been updated with defaults (old version: 11, new: 11)
[16:44:57] [0x00007f745c812580] [M] [debug] log: stdout: yes
[16:44:57] [0x00007f745c812580] [M] [info] log: level: info
[16:44:57] [0x00007f745c812580] [M] [info] date: init.
[16:44:57] [0x00007f745c812580] [M] [info] db path: /home/andy/.local/share/local-mail
[16:44:57] [0x00007f745c812580] [M] [error] db: failed to open database, please check the manual if everything is set up correctly: db: error: no search.exclude_tags defined in notmuch-config
fish: Job 1, 'astroid' terminated by signal SIGSEGV (Address boundary error)

so I go add to the notmuch config

[search]
exclude_tags=

next up

astroid
[2023-05-25 16:47:12.395250] [0x00007f0556b01580] [info]    welcome to astroid! - 0.16.0
[2023-05-25 16:47:12.414198] [0x00007f0556b01580] [debug]   HOME: /home/andy
[2023-05-25 16:47:12.414213] [0x00007f0556b01580] [info]    cf: loading: "/home/andy/.config/astroid/config"
[2023-05-25 16:47:12.414438] [0x00007f0556b01580] [info]    cf: version: 11
[2023-05-25 16:47:12.414509] [0x00007f0556b01580] [debug]   cf: check config..
[2023-05-25 16:47:12.414520] [0x00007f0556b01580] [warning] cf: missing values in config have been updated with defaults (old version: 11, new: 11)
[16:47:12] [0x00007f0556b01580] [M] [debug] log: stdout: yes
[16:47:12] [0x00007f0556b01580] [M] [info] log: level: info
[16:47:12] [0x00007f0556b01580] [M] [info] date: init.
[16:47:12] [0x00007f0556b01580] [M] [info] db path: /home/andy/.local/share/local-mail
[16:47:12] [0x00007f0556b01580] [M] [error] db: failed to open database, please check the manual if everything is set up correctly: db: error: no maildir.maildir_synchronize_flags defined in notmuch-config
fish: Job 1, 'astroid' terminated by signal SIGSEGV (Address boundary error)

so I go add to the notmuch config

[maildir]
synchronize_flags=

now

astroid
[2023-05-25 16:48:08.358743] [0x00007f607c32b580] [info]    welcome to astroid! - 0.16.0
[2023-05-25 16:48:08.378232] [0x00007f607c32b580] [debug]   HOME: /home/andy
[2023-05-25 16:48:08.378249] [0x00007f607c32b580] [info]    cf: loading: "/home/andy/.config/astroid/config"
[2023-05-25 16:48:08.378460] [0x00007f607c32b580] [info]    cf: version: 11
[2023-05-25 16:48:08.378534] [0x00007f607c32b580] [debug]   cf: check config..
[2023-05-25 16:48:08.378548] [0x00007f607c32b580] [warning] cf: missing values in config have been updated with defaults (old version: 11, new: 11)
[16:48:08] [0x00007f607c32b580] [M] [debug] log: stdout: yes
[16:48:08] [0x00007f607c32b580] [M] [info] log: level: info
[16:48:08] [0x00007f607c32b580] [M] [info] date: init.
[16:48:08] [0x00007f607c32b580] [M] [info] db path: /home/andy/.local/share/local-mail
terminate called after throwing an instance of 'boost::wrapexcept<boost::property_tree::ptree_bad_data>'
  what():  conversion of data to type "b" failed
fish: Job 1, 'astroid' terminated by signal SIGABRT (Abort)

I'm trying changing the config, deleting the config. Still can't seem to get past this.

My setup is just having some maildirs in ~/.local/share/local-mail and a barebones notmuch config

[database]
path=/home/andy/.local/share/local-mail

[user]
name=Andy

[new]

[search]
exclude_tags=

[maildir]
synchronize_flags=

I was imagining just trying out the interface before I go full on with a complicated config. I just want to work offline with a test copy of a maildir.

[ibuclaw]

terminate called after throwing an instance of 'boost::wrapexceptboost::property_tree::ptree_bad_data'
what(): conversion of data to type "b" failed

This occurs because synchronize_flags doesn't have a boolean value (i.e: true, false).

I'm not able to reproduce the other SIGSEGV's locally when running astroid in a new empty environment, however I have strong reasons to believe that #753 would fix them.

[jorsn]

Does this still happen with the current HEAD, i.e. with 4b483ff applied?

[jorsn]

@digitalsignalperson Since this is probably fixed, I will assume it is for the new release (#750) unless you complain.
If you don't encounter the problem anymore, please close the issue.